This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Dear reader, we are very pleased, that with this new book series of our institute cluster, we succeeded in bringing together our scientifically diverse and widespread publications. Especially for such an interdisciplinary research institution like the ZLW/IMA & IfU the wish for an “integrated presentation” is quite great, since the represented disciplinary cultures - from engineering und natural sciences over humanities, social and communicational sciences up to economics - meet very different publication cultures. It is therefore of most importance for us to unite all contributions from a specified period in one book. The idea to publish an annual book edition, which contains a comprehensive selection of publications of the institute cluster, originated from this wish. Almost all publications were peer-reviewed and published in recognized journals or conference proceedings of the various disciplinary cultures. This is the first edition of this series. In the last months the institute cluster ZLW, IMA & IfU has been reorganized. Together with the division managers we summarized our existing research fields and those introduced by Sabina Jeschke in five department-overlapping research fields. These fields run transversely to the organizational structure of the five areas of operations (departments) of the institute cluster, which are depictured in the underlying matrix. The agile and turbulence-suitable processes field summarizes our research activities on systems that are turbulent in their level of complexity. Therein the transfer of the agility principle of the software development to the process management plays an important role. Our research and application areas lie in knowledge and technical intensive organizations. To these belong e.g. the Cross Sectional Processes, which deal with efficient networking and transparency of the research processes in large, distributed and highly networked research clusters. Examples are the Cluster of Excellence “Integrative Production Technology for High-wage Countries” and “Taylor- made Fuels from Biomass” at the RWTH Aachen University. - The challenge for the research project Med-on@Aix (a telematic-supported high-performance rescue system) is the formation of organizational and logistical consistency in highly complex and non-predictable rescue scenarios. - Projects like the “value proposition of the IT”, which deal with the determination of assessments in complex environments, build the bridge to the new research focus: Dynamical IT-Outsourcing/Cloud Computing. v
vi
Preface
Fig. 1 Research Profile of ZLW/IMA & IfU
The research field next-generation teaching and learning concepts does not only address the demand on new ways of teaching and teaching processes for the generation grown up with the internet. It also involves the different approaches to teaching and learning along different diversity categories with the focus on age and gender and with the consideration of different mental user models. A typical example for Universities is one of the largest projects of the institute cluster, the project KISSWIN. On behalf of the Federal Ministry of Education and Research (BMBF) a nationwide platform for young scientists is built and operated. This includes also an annual convention with around 1000 participants. - To address also the younger academic generation we opened the student laboratory RoboScope in July 2010, which focuses on robotics and is sponsored by the federal state North Rhine-Westphalia. This is the first building block of the RWTH Education Labs with the prime target group being school children. - The economical side within this research field is covered for example by the EU project RELOAD. In RELOAD microteaching units are developed for employees of DIY stores, which can be used for further qualification during work time. The research field cognitive IT-supported processes covers the numerous project areas, which primarily deal with the cognition of technical systems and the development of cognitive-supported models that base on these systems. Within this research field belongs the research area “partly automated systems” which has been ongoing for about 10 years and involves electronically coupled truck convoys. - In future we will intensify our research on automated elements up to completely automated driving. - Recently the conception of cognitive, artificial intelligence-supported controls to plan flexible, adaptive assembly systems for a larger diversity of products (keyword: “Factories of the Future”) have become a greater matter of interest. - Also the interaction of heterogeneous robot teams will be considered at the ZLW/IMA & IfU as well as research on humanoid robots.
Preface
vii
With target group-adapted user models we cover a research field, which has a long tradition at the institute cluster. The aim is the development of technologies and human-machine interfaces considering the user models of the different target groups, the human “going from servant to user”. An example is the by European forwarding agents driven development of a new type of container within the project Tellibox, which is sponsored by the European Union. The development is oriented towards the loading requirements of the haulers, which is especially a challenge for the construction teams. - Another socially extremely relevant topic is “accessibility”, in particular barrier-free concepts for web services, software concepts, devices and machines using user and model centered development and multi modal user concepts. The research questions within the field semantic networks and ontologies have been addressed in more and more of our projects over the last years. Especially in complex value chains, which are close to applications, this topic is an enormous challenge for researchers. Within the Cluster of Excellence “Integrative production technology for highwage countries” the institute makes an important contribution to the semantic coupling of the differently structured simulations of applications. This enables a continuous simulation of complex production procedures. Centerpiece is the development of an ontology basis through which the different parts of the simulations can communicate. - Development of ontologies and semantic networks of the focus of the project AsIsKnown for the determination of trends that overlap several sectors in the design of carpets, curtains and upholstered furniture. The network is used now European-wide. - Within the new research focus “web services, semantic web, cloud computing” ontologies play and important role for the design and implementation, and orchestration of single services to complex process chains. The five core research fields display the structure of this book. The complete work is now lying before us of which we are happy about. Our deepest thanks goes to our personal assistant Dr. Alicia Liedtke for her tireless and thorough sighting and compilation of the articles and the coordination of this publication project, which was an unusual task for an experimental physicist. Grateful acknowledgements are also due to those who supported us with the formatting and correction of the articles. Also, we thank the Springer Verlag for publishing this volume and thus helping us to position the research activities of our institute cluster into an international frame. We hope that this series of publications is contributing to the global promotion of broad und interdisciplinary research methods. Further we would like to express our gratitude to the RWTH Aachen University for providing an open, inspiring atmosphere and a great scientific infrastructure, which make the RWTH Aachen to an extraordinary University. Finally, we thank all employees from our heart for their never ending commitment, for their creativity and that they do make the institute to something very special! Aachen, August 2010
Sabina Jeschke Ingrid Isenhardt Klaus Henning
Contents
Part I Agile and turbulence-suitable processes How to Structure and Foster Innovative Research . . . . . . . . . . . . . . . . . . . . . . Ursula Bach, Ingo Leisten
3
Innovation Rules 2030 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Ursula Bach, Klaus Henning Gestaltungsansätze für ein systemisches Fakultätsmanagement . . . . . . . . . 27 Sabine Bischoff, Paul Flachskampf, Klaus Henning Prävention und Innovation - Strategische Ausrichtung, aktuelle Fragen und Ausblick . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Klaus Henning, Ursula Bach Yes, we can! Warum Deutschland den Kopf nicht in den Sand stecken sollte . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Klaus Henning, Frank Hees, Ursula Bach, Alan Hansen Management and Optimal Distribution of Large Student Numbers . . . . . . 71 Sabina Jeschke, Gerald Lach, Robert Luce, Olivier Pfeiffer, Erhard Zorn Spirit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 Sabina Jeschke, Barbara Burr, Peter Göhner, Wolfram Ressel, Wolfgang Schlicht Going diverse in the two Clusters of Excellence “Integrative Production Technology for High-wage Countries” and “Tailor-Made Fuels from Biomass” at RWTH Aachen University . . . . . . . . . . . . . . . . . . . . . . . . . . 103 Claudia Jooß, René Vossen, Anja Richert, Ingrid Isenhardt ix
x
Contents
Strategic Transfer Communication in Prevention Research as a Contribution to the Innovative and Competive Ability of Enterprises . . . . 107 Ingo Leisten, Frank Hees A Methodology to Reduce Technical Risk in the Development of Telematic Rescue Assistance Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 Matthias Müller, Michael Protogerakis, Klaus Henning Defining a universal actor content-element model for exploring social and information networks considering the temporal dynamic . . . . . . . . . . . 123 Claudia Müller, Benedikt Meuthrath, Sabina Jeschke A Composite Calculation for Author Activity in Wikis: Accuracy Needed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 Claudia Müller-Birn, Janette Lehmann, Sabina Jeschke Experiences from an International Student and Staff Exchange Program and Some Still Unsolved Mysteries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 Olivier Pfeiffer, Sabina Jeschke, Lars Knipping, Nina Reinecke, Erhard Zorn A System Architecture for a Telematic Support System in Emergency Medical Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163 Michael Protogerakis, Arno Gramatke, Klaus Henning Designing Agile Processes in Information Management . . . . . . . . . . . . . . . . 173 Uschi Rick, René Vossen, Anja Richert, Klaus Henning Kulturveränderung oder kulturbasierte Veränderung? Eine strategische Entscheidung . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 Robert Schmitt, Thilo Münstermann, Klaus Henning, Alexandra Ottong Network Management for Clusters of Excellence - A Balanced-Scorecard Approach as a Performance Measurement Tool . . . . . . . . . . . . . . . . . . . . . . . 195 Florian Welter, René Vossen, Anja Richert, Ingrid Isenhardt Part II Next-generation teaching and learning concepts for universities and the economy Application of Remote Technology to Electrical Power System Laboratories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211 Saleh Al-Jufout, Sabina Jeschke, Abdullah Y. Al-Zoubi, Jarir Nsour, Olivier Pfeiffer
Contents
xi
Environments for Work and Learning 2020 . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 Stefan Brall, Ursula Bach, Frank Hees Developing a PBL-based Rescue Robotics Course . . . . . . . . . . . . . . . . . . . . . 231 Frank Hees, Sabina Jeschke, Nicole Natho, Olivier Pfeiffer Networking Resources for Research and Scientific Education in BW-eLabs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241 Sabina Jeschke, Eckart Hauck, Michael Krüger, Wolfgang Osten, Olivier Pfeiffer, Thomas Richter Networked Virtual and Remote Laboratories for Research Collaboration in Natural Sciences and Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259 Sabina Jeschke, Arno Gramatke, Olivier Pfeiffer, Christian Thomsen, Thomas Richter Natural Sciences in the Information Society First Experiences . . . . . . . . . . 271 Grit Köppel, Sabina Jeschke, Nicole Natho, Lars Knipping, Grit Petschik, Christian Schröder, Erhard Zorn Bringing Problem Based Learning to Academic Engineering Education using Robotics as the Utility Vehicle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281 Nicole Natho, Sabina Jeschke, Lars Knipping, Olivier Pfeiffer, Ursula Vollmer, Marc Wilke New Media in Education and Research – a Sophomore Lecture at TU Berlin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291 Nicole Natho, Sabina Jeschke, Erhard Zorn Supporting Collaboration in Professional Soft-Skill Training Courses . . . . 301 Olivier Pfeiffer, Sabina Jeschke, Lars Knipping, Nicole Natho LiLa: A European Project on Networked Experiments . . . . . . . . . . . . . . . . . 307 Thomas Richter, David Boehringer, Sabina Jeschke VIDEOEASEL - A Flexible Programmable Simulation Environment for Discrete Many Body Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319 Thomas Richter, Sabina Jeschke, Olivier Pfeiffer An Intensive Course in Mathematics for Engineers: Experiences and Prospects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329 Mike Scherfner, Sabina Jeschke, Matthias Plaue
xii
Contents
Moderne Studienform: Galilea und der Bachelorstudiengang „Naturwissenschaften in der Informationsgesellschaft“ . . . . . . . . . . . . . . . . 337 Christian Schröder, Sabina Jeschke, Nicole Natho, Olivier Pfeiffer Microtraining for Workplace-Related Learning . . . . . . . . . . . . . . . . . . . . . . . 347 Anne Carina Thelen, Sascha Daniel Herr, Frank Hees, Sabina Jeschke Teachers need robotics-training, too . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359 Ursula Vollmer, Sabina Jeschke, Barbara Burr, Lars Knipping, Jörg Scheurich, Marc Wilke RELOAD - A Semantic-based Learning and Knowledge Platform for Employees of the Do-It-Yourself Industry . . . . . . . . . . . . . . . . . . . . . . . . . 365 Florian Welter, Olivier Pfeiffer, Anja Richert, Sabina Jeschke Pre-Freshmen Students Gearing up with Early Bird . . . . . . . . . . . . . . . . . . . 373 Erhard Zorn, Sabina Jeschke, Akiko Kato, Olivier Pfeiffer Part III Cognitive IT-supported processes for heterogeneous and cooperative systems Software Architecture, Knowledge Compiler and Ontology Design for Cognitive Technical Systems Suitable for Controlling Assembly Tasks . . . 383 Eckart Hauck, Daniel Ewert, Arno Gramatke, Klaus Henning Sustainable Transport - Knowledge and Innovations at RWTH Aachen University for Europe’s Systems of Tomorrow . . . . . . . . . . . . . . . . . . . . . . . . . 393 Klaus Henning, Leonie Petry, Richard Ramakers, Julie Meinhold Benefits of RFID for the Production of hybrid Micro Systems in flexible Production Networks of SMEs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407 Verena Jänen, Christian Tummel, Klaus Henning Automated Truck Platoons on Motorways – A Contribution to the Safety on Roads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415 Ralph Kunze, Max Haberstroh, Richard Ramakers, Klaus Henning, Sabina Jeschke Organization and Operation of Electronically Coupled Truck Platoons on German Motorways . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 427 Ralph Kunze, Richard Ramakers, Klaus Henning, Sabina Jeschke
Contents
xiii
Electronically Coupled Truck Platoons on German Highways . . . . . . . . . . . 441 Richard Ramakers, Klaus Henning, Stefan Gies, Dirk Abel, Max Haberstroh Determination of the Order of Electronically Coupled Trucks on German Motorways . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453 Christian Tummel, Ralph Kunze, Klaus Henning RENS – Enabling A Robot to Identify A Person . . . . . . . . . . . . . . . . . . . . . . . 467 Xin Yan, Sabina Jeschke, Hinrich Schütze, Amit Dubey, Marc Wilke Part IV Target group-adapted user models for innovation and technology development processes Open Innovation - Strategie der offenen Unternehmensgrenzen für KMU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 481 Sabine Bischoff, Gergana Aleksandrova, Paul Flachskampf Criteria for Age Based Design of Active Vehicle Safety Systems . . . . . . . . . 495 Max Haberstroh, Max Klingender, Richard Ramakers, Klaus Henning Dynaxibility for Innovation – Global trends in the field of “Working, Learning, Developing Skills” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 507 Klaus Henning, Frank Hees, Alan Hansen Größere Nutzfahrzeuge - länger und schwerer? Chancen und Risiken für Europa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 519 Max Klingender, Richard Ramakers, Klaus Henning In-depth Safety Impact Study on longer and/or heavier commercial vehicles in Europe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 539 Max Klingender, Richard Ramakers, Klaus Henning Economic Assessment of Innovations – Application of the Value Oriented Cost-Effectiveness Estimation on Electronically Coupled Trucks . . . . . . . . 553 Ralph Kunze, Sabine Bischoff, Paul Flachskampf User Acceptance as a Key to Success for the Implementation of a Telematic Support System in German Emergency Medical Services . . . . . 563 Marie-Thérèse Schneiders, Michael Protogerakis, Ingrid Isenhardt Benutzungsorientierte Entwicklung barrierefreier Benutzungsschnittstellen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 569 Helmut Vieritz, Sabina Jeschke, Olivier Pfeiffer
xiv
Contents
Using Web Accessibility Patterns for Web Application Development . . . . . 579 Helmut Vieritz, Sabina Jeschke, Olivier Pfeiffer Part V Semantic networks and ontologies for complex value chains and virtual environments Crystalline Ge1−x Snx Heterostructures in Lateral High-Speed Devices . . 597 Sabina Jeschke, Olivier Pfeiffer, Joerg Schulze, Marc Wilke Digitale Produktion via Enterprise Application Integration . . . . . . . . . . . . . 609 Tobias Meisen, Philipp Meisen, Daniel Schilberg, Sabina Jeschke Dynamische Gruppenarbeit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 623 Thilo Münstermann, Jens Völzke, Paul Flachskampf Knowledge Base Concepts in the KEA System Combined with Social Networking Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 635 Nicole Natho, Sabina Jeschke, Marc Wilke, Olivier Pfeiffer Enterprise Application Integration für die virtuelle Produktion . . . . . . . . . 651 Daniel Schilberg, Tobias Meisen, Philippe Cerfontaine, Sabina Jeschke Ontology Based Semantic Interconnection of Distributed Numerical Simulations for Virtual Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 661 Daniel Schilberg, Tobias Meisen, Klaus Henning Verkettung von Prozesssimulationen für die virtuelle Produktion . . . . . . . . 671 Daniel Schilberg, Arno Gramatke, Klaus Henning Monographs and Book Contributions Published at the ZLW/IMA & IfU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 679
Contributors
Dirk Abel Institut für Regelungstechnik (IRT), RWTH Aachen University, Steinbachstr. 54, 52074 Aachen, Germany Gergana Aleksandrova ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Saleh Al-Jufout Electrical Engineering Department, Tafila Technical University, P.O. Box (179), Tafila (66110), Jordan Abdullah Al-Zoubi Princess Sumaya University for Technology, P.O. Box 1438, Al-Jubaiha, 11941 Jordan Ursula Bach ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Sabine Bischoff Institut für Unternehmenskybernetik e.V., Schurzelter Str. 25, 52074 Aachen, Germany David Boehringer RUS Computing Center, University of Stuttgart Allmandring 30a, 70550 Stuttgart, Germany Stefan Brall ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Barbara Burr Institute of Information Technology Services (IITS), University of Stuttgart, Allmandring 30A, 70569 Stuttgart, Germany Philippe Cerfontaine Institute for Scientific Computing, RWTH Aachen University, Seffenter Weg 23, 52056 Aachen, Germany David Cremer ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Amit Dubey Institute for Communicating and Collaborative Systems, University of Edinburgh, 10 Crichton Street, Edinburgh EH8 9AB, UK Daniel Ewert ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany xv
xvi
Contributors
Paul Flachskampf Institut für Unternehmenskybernetik e.V., Schurzelter Str. 25, 52074 Aachen, Germany Stefan Gies Institut für Kraftfahrzeuge (ika), RWTH Aachen University, Steinbachstr. 7, 52074 Aachen Peter Göhner Institut für Automatisierungs- und Softwaretechnik, University of Stuttgart, Pfaffenwaldring 47, 70550 Stuttgart, Germany Wojtek Gora MuLF, Berlin University of Technology, Straße des 17. Juni 136, 10623 Berlin, Germany Arno Gramatke ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Max Haberstroh ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Jens U. Hahn Department of Information and Communication, Stuttgart Media University, Nobelstr. 322, 70191 Stuttgart, Germany Alan Hansen ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Eckart Hauck ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Frank Hees ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Leni Helmes Fachinformationszentrum Karlsruhe (FIZ Karlsruhe), Hermann-vonHelmholtz-Platz 1, 76344 Eggenstein-Leopoldshafen, Germany Klaus Henning IMA/ZLW & IfU - RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Sascha Daniel Herr ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Ingrid Isenhardt IMA/ZLW & IfU - RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Verena Jänen ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Sabina Jeschke IMA/ZLW & IfU - RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Claudia Jooß ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Akiko Kato Institut für Theoretische Physik, Berlin University of Technology, Hardenbergstr. 36, 10623 Berlin, Germany
Contributors
xvii
Max Klingender ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Lars Knipping MuLF, Berlin University of Technology, Straße des 17. Juni 136, 10623 Berlin, Germany Grit Köppel MuLF, Berlin University of Technology, Straße des 17. Juni 136, 10623 Berlin, Germany Walter Kriha Institute of Information Technology Services (IITS), University of Stuttgart, Allmandring 30A, 70569 Stuttgart, Germany Michael Krüger Freiburg Materials Research Centre (FMF), University of Freiburg, Stefan Meier Strasse 21, 79104 Freiburg, Germany Ralph Kunze ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Gerald Lach MuLF, Berlin University of Technology, Straße des 17. Juni 136, 10623 Berlin, Germany Mirjana Lach MuLF, Berlin University of Technology, Straße des 17. Juni 136, 10623 Berlin, Germany Janette Lehmann Institut für Informatik, Universität Potsdam, August-Bebel-Str. 89, 14482 Potsdam, Germany Ingo Leisten ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Andreas W. Liehr Freiburg Materials Research Centre (FMF), University of Freiburg, Stefan Meier Strasse 21, 79104 Freiburg, Germany Robert Luce MuLF, Berlin University of Technology, Straße des 17. Juni 136, 10623 Berlin, Germany Jan Lübbe MuLF, Berlin University of Technology, Straße des 17. Juni 136, 10623 Berlin, Germany Julie Meinhold ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Philipp Meisen Inform GmbH, Pascalstr. 23, 52076 Aachen, Germany Tobias Meisen ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Benedikt Meuthrath Institute of Computer Science, Freie Universität Berlin, Königin-Luise-Str. 24/26, 14195 Berlin, Germany Claudia Müller-Birn (formerly Müller) Institute for Software Research, Carnegie Mellon School of Computer Science, 5000 Forbes Avenue, Pittsburgh, PA 15213, USA
xviii
Contributors
Matthias Müller ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Thilo Münstermann Institut für Unternehmenskybernetik e.V., Schurzelter Str. 25, 52074 Aachen, Germany Nicole Natho MuLF, Berlin University of Technology, Straße des 17. Juni 136, 10623 Berlin, Germany Jarir Nsour Princess Sumaya University for Technology, P.O. Box 1438, Al-Jubaiha, 11941 Jordan Wolfgang Osten Institute of Technical Optics, University of Stuttgart, Pfaffenwaldring 9, 70569 Stuttgart, Germany Alexandra Ottong Fraunhofer-Institut für Produktionstechnologie IPT, Steinbachstr. 17, 52074 Aachen, Germany Leonie Petry ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Grit Petschik MuLF, Berlin University of Technology, Straße des 17. Juni 136, 10623 Berlin, Germany Olivier Pfeiffer MuLF, Berlin University of Technology, Straße des 17. Juni 136, 10623 Berlin, Germany Matthias Plaue School for Mathematics and Natural Sciences, TU Berlin, Straße des 17. Juni 136, 10623 Berlin, Germany Michael Protogerakis ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Richard Ramakers ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Nina Reinecke MuLF, Berlin University of Technology, Straße des 17. Juni 136, 10623 Berlin, Germany Wolfram Ressel Lehrstuhl für Straßenplanung und Straßenbau, University of Stuttgart, Pfaffenwaldring 7, 70569 Stuttgart, Germany Anja Richert ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Thomas Richter RUS Computing Center, University of Stuttgart Allmandring 30a, 70550 Stuttgart, Germany Uschi Rick ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Mike Scherfner Institute of Mathematics, Berlin University of Technology, Straße des 17. Juni 136, 10623 Berlin, Germany
Contributors
xix
Jörg Scheurich Institute of Information Technology Services (IITS), University of Stuttgart, Allmandring 30A, 70569 Stuttgart, Germany Daniel Schilberg ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Wolfgang Schlicht Institut für Sport- und Bewegungswissenschaft, University of Stuttgart, Allmandring 28, 70569 Stuttgart, Germany Robert Schmitt Fraunhofer-Institut für Produktionstechnologie (IPT), Steinbachstr. 17, 52074 Aachen, Germany Gerhard Schneider Computing Center, Albert-Ludwigs-Universität, HermannHerder-Strasse 10, 79085 Freiburg, Germany Marie-Thérèse Schneiders ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Christian Schröder MuLF, Berlin University of Technology, Straße des 17. Juni 136, 10623 Berlin, Germany Hinrich Schütze Institut für Maschinelle Sprachverarbeitung, University of Stuttgart, Azenbergstr. 12, 70174 Stuttgart, Germany Jörg Schulze Institut für Halbleitertechnik, University of Stuttgart, Pfaffenwaldring 47, 70569 Stuttgart, Germany Werner Stephan Universitätsbibliothek Stuttgart (UBS), University of Stuttgart, Holzgartenstr. 16, 70174 Stuttgart, Germany Anne Carina Thelen ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Christian Thomsen MuLF, Berlin University of Technology, Straße des 17. Juni 136, 10623 Berlin, Germany Christian Tummel ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Helmut Vieritz ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Ursula Vollmer Institute of Information Technology Services (IITS), University of Stuttgart, Allmandring 30A, 70569 Stuttgart, Germany Jens Völzke Institut für Unternehmenskybernetik e.V., Schurzelter Str. 25, 52074 Aachen, Germany René Vossen ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Karl-Heinz Weber Fachinformationszentrum Karlsruhe (FIZ Karlsruhe), Hermann-von-Helmholtz-Platz 1, 76344 Eggenstein-Leopoldshafen, Germany
xx
Contributors
Florian Welter ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany Marc Wilke Institute of Information Technology Services (IITS), University of Stuttgart, Allmandring 30A, 70569 Stuttgart, Germany Xin Yan Institute of Information Technology Services (IITS), University of Stuttgart, Allmandring 30A, 70569 Stuttgart, Germany Erhard Zorn MuLF, Berlin University of Technology, Straße des 17. Juni 136, 10623 Berlin, Germany
Part I
Agile and turbulence-suitable processes
How to Structure and Foster Innovative Research Ursula Bach, Ingo Leisten
Abstract Innovative research structures provide the possibility to address new stakeholders on the topic Preventive Occupational Health and Safety. Therefore the stakeholder to the Preventive Occupational Health and Safety can be appealed, as for example health insurance schemes, chambers of commerce or ministries. This means, visibility of the research community can be raised at the network and society level ([HLBH09]:26). To reach the aim “better visibility of the research community”, “improving integration of the partners in the research groups” and the “avoidance of the "fragmenting" of a research community”, different network management methods were applied in the founding priority Preventive Occupational Health and Safety ([Hue03]:119). The choice of the methods of the network management bases on the needs of a research network in three specific network phases: phase of initiation, phase of stabilisation and finally the phase of increasing steadiness. Every phase puts different challenges to the network management. Keywords Innovation · research · structure · network management · methodical support
Fig. 1 The Island of Research according to v. Alemann [vA84]. Own translation of inscriptions
activities, the Project Management Agency for the federal R&D programme on innovations in working life (PT-DLR) introduced innovative funding structures for granting proposals of the Founding Priority “Preventive Occupational Safety and Health”. On various levels, scientific work within the promotional focus takes place across different joint research projects ([HLBH09]: 14–17). Through bundling, systematizing and concentration of the diverse project actors new constellations of actors and institutes can be identified and addressed as target groups (see Fig. 2). Particularly on the levels of networks and society visibility of the research community can be enhanced, this in turn helps to address the stakeholders of Preventive Occupational Safety and Health ([HLBH09]: 26). To support the work of the different research groups a transfer project was initialized. This special research meta-project StArG (Strategischer Transfer im Präventiven Arbeits- und Gesundheitsschutz, engl. Strategical Transfer in the field of Preventive Occupational Safety and Health) provide research results for network management and research communication. Here the basic results will be introduced. It has proved feasible to map the constellations of actors and institutions onto the four levels of recursion. Also, with respect to the debate on the structure of the funding priority, it has become established practice to also map the individual research actors onto the same levels of recursion (see Fig. 3): Sub-Projects, Joint Projects, Focus Groups, Funding Priority describe the structures within the funding priority, on the other hand, the levels reaching beyond the funding priority point to the research program, the German Federal Ministry of Education and Research, the political space and the international level respectively.
How to Structure and Foster Innovative Research
5
Fig. 2 Depiction of new actor and institutional constellations based on the Transfer Diagnostic Workshops held by the StArG meta-project
Fig. 3 Recursive levels of the Funding Priority “Preventive Occupational Safety and Health” ([HLBH09]: 15)
6
Ursula Bach et al.
2 Network Management methods of the funding priority “Preventive Occupational Safety and Health” When choosing management methods, the funding priority understood itself as a research network. This is the case as it shows network-specific characteristics. Network theory gives four characteristics research networks have to feature in order to be referred to as such (cf. [Jan04]: 21): 1a) Research networks consist of several individually disjunct network partners. 2a) These network partners are related through regulated prearrangements. 3a) The network partners have the possibility to use synergies, e.g. through resource sharing or knowledge exchange. 4a) The network partners have a technological or social subsystem. Regarding the research network, these four characteristics can be translated as follows: 1b) The research network “Preventive Occupational Safety and Health” consists of 18 joint research projects and nine individual research projects that work on diverse research results and approaches within diverse research institutions and enterprises from diverse scientific disciplines ([pt208]: 4). 2b) The content-related relationships of the various projects are formulated through the composition of the focus groups. The relationships built through the funding priority and fostered through regular events are initiated, organized and carried out by the meta-project (for an overview on the activities of the funding priority see www.starg-online.de). 3b) The individual network and research partners have a multitude of possibilities to use resources within the funding priority, e.g. through joint surveys and synergetic public relations work or through annual conferences that serve as a platform for joint scientific work on topics of “Preventive Occupational Safety and Health”. 4b) The social sub-system of the research network is created by the structures of the funding priority (cf. Fig. 3). The technological sub-system is created among others through the interactive exchange and discussion platform as well as through the material resources of the individual projects. The following goals are generally intended when initiating research networks: Better integration of the joint research participants and “avoidance of `fragmentatio´” of a research community ([Hue03]:119). In joint research, a better integration generally means a better cooperation between science and entrepreneurial practice. In the case of the funding priority “Preventive Health and Safety” broaderranging aims are intended: To contribute to the competitive ability of German economy, to counter economic losses through insufficient preventive workplace health and safety measures and to stay abreast of changes in the modern work environment [BMB05].
How to Structure and Foster Innovative Research
7
The choice of network management methods thus has to allow for the project goals as well as the necessities of a research network during the individual network stages. According to Ahrens et al. [OH04], a research network, once approved, runs through three stages: the initiating phase, the stabilizing phase and the permanent phase. The initiating phase lays the foundations for joint work. Here, mutual trust is established and first attunements are made regarding thematic decisions, division of labor or structural agreements. The stabilizing phase is the most productive stage of a network. The chances for utilizing resources have become clear and trust can be deepened. In the permanent phase, the actual work of the network is heading toward the end and results have to be secured, processed and made available for subsequent stages (for an elaboration, cf. [Aea04]). Within the third phase, research networks with private and/or public funding have to pay special attention to later utilization of their projects’ results ([BMB06]:5). Leisten further suggests keeping appropriability in view across all of the research process [HLB09]. When designing research networks, it has to be noted that these phases rarely ever occur in stringently chronologic succession but rather tend to overlap, repeat themselves or get abandoned. Thus, every phase poses different challenges to network management (summarized in Table 1). Besides knowledge of the individual stages a diagnosis of the research network is necessary to facilitate strategic management. This diagnosis is structured along five lines: Structure of actors, communication and interactions, development and renewing, controlling, and the material equipment of the projects and focus groups ([Lea04]:39). The Actor line describes e.g. what projects are part of the funding priority, which value partners are integrated into the research processes or what focus groups have been initiated. The line of communication and interactions depicts what
Table 1 Own compilation of the phases and their requirements for a network management of the funding priority according to [Aea04]:19. Phases Initiating phase
Requirements
Choice of projects, Composition of focus groups, Matching expectations and values that are to be the foundations of joint efforts, Settling the means of exchange within the thematic superstructure of the focus groups, Stabilizing phase Attunement of task and resource allotments, Assigning individual tasks and resources. Create possibilities to build trust beyond the activities of the first stage, Work out a mutual understanding of the research matter, Synergies are being recognized and made use of, e.g. through joint surveys of research results. Permanent phase Securing a sustained impact of research results, Expansion of the network, Funding-priority-internal securing of results for funding-priority-external transfer, Work out visions for the future of the funding priority.
8
Ursula Bach et al.
ways of communication are used and what sorts of interaction are possible between the actors. Development and renewing shows the innovative potential of the research network. Controlling contains tools and instruments that a network wants to allow itself to check its alignment with its goals. Material equipment provides an overview of the financial, organizational and staff resources available to the research network. These findings describe the basis of decision-making on which the meta-project StArG has chosen and incorporated management methods during the term of the funding priority “Preventive Occupational Safety and Health” (see Fig. 4). The picture exemplarily shows one management method per network phase. Other methods and activities can be obtained at www.starg-online.de. During the stabilizing phase it has proved an advantage within the funding priority to hold a Method Workshop within the individual focus groups. The method workshop provides the scientific sub-projects with an opportunity to exchange on scientific methods on data survey, data analysis and data processing. This has proved valuable to facilitate exchange across project, discipline and domain borders and to make interdisciplinary work possible. It is one powerful tool to lay foundations for the future of the focus group and to make sure empirical surveys can be used across the board. From the vantage point of the meta-project StArG, it seems sensible to hold a Transfer-Diagnosis Workshop both in the joint research projects and in the focus groups. The Transfer-Diagnosis Workshop is a concept that has been tailordesigned for projects and focus groups of the funding priority “Preventive Workplace Health and Safety”. It serves to analyze transfer conditions within the individual research projects. Target of the workshops is to obtain and analyze information on the transfer situation of the project partners. Changes concerning transfer contents, goals, framework conditions, partners and target groups can be identified in order to react with new transfer strategies ([HLB09]:25f.).
Fig. 4 Table of the possible methods of network management within the funding priority considering the individual phases and available design elements
How to Structure and Foster Innovative Research
9
The Permanent Phase within a funding priority serves as preparation for the phase of sustainment where results and their appropriation are maintained. Here, we gave the example of the Position Paper. Position papers summarize the metatopics developed during the collaboration of the focus groups. Content parentheses are visible in the denomination of the focus groups, like “New Methods of Cost-Effectiveness Assessment” or “Corporate Prevention and Health Fostering in Knowledge Economy”. Through a dialog with the projects of the focus group demands for research can be extracted so that a thematic Future Bulletin on the topic of the focus group can be developed and published. The Position Paper provides an opportunity to give a scientific strategic statement during the phase of sustainment, but also to compile a summary of the scientific results. All the presented methods of network management must be able to bridge the gap between the requirements of research funding and the individual work within the project (according to [Sau05]:101).
3 Transfer activities of the funding priority “Preventive Occupational Safety and Health” in 2009 After having shed light on the “innards” of a research network, let us not forget that a funding priority is not established as an end in itself by Federal research funding. The funding priority is set the task to find practice-oriented approaches to the prevention dilemma and to enable value partners to make use of them ([PA06]:59). At the same time, it is intended to support intermediaries in their function as service providers ([BMB06]:263). In order to not let a funding priority become a “closed shop” but rather to let its results become part of public knowledge a multitude of transfer instruments are necessary which have to serve both transfer in width and depth and are suitable for a diversity of target groups (cf. Fig. 3). Transfer activities are carried out by the individual levels of the funding priority (cf. Fig. 3). Role of the meta-project is to initiate these individual transfer activities, to accompany them methodologically and to assist with implementation (print, layout, moderation, conceptualization etc.). Exemplarily for all transfer activities of the funding priority, we will present the methods used in 2009.
3.1 Transfer Level of Joint Research Projects Besides classical transfer methods like web page maintenance, keeping project flyers up to date and the publication of papers, some of the projects entering the phase of sustainment have organized and held Project Conferences. Not only those immediately involved were invited, but also persons and actors working in the research – and above all practical – field of “Preventive Workplace Health and Safety”. This enabled the projects to pass on research results and to acquire new appropriators and users.
10
Ursula Bach et al.
3.2 Transfer Level of Focus Groups For the focus groups, the last complete year of the funding priority’s term became the time of reaping the fruits of the mutual discussions, the scientific debates and the methodological arguments. The contents of the focus groups have been compiled in two joint publications. The Focus Group Brochures serve to thematically collect and concentrate the scientific work in the scope of the focus groups. These brochures are targeted at research funding. While the focus group brochures summarize the findings of the focus groups, the Position Papers offer an opportunity to look into the future. The position paper as a “future bulletin” serves to point at further demands for research and to identify blind spots.1 A Strategy Planner has been developed by the meta-project as support for the focus groups. Here, some 70 transfer methods can be evaluated, weighed and prioritized according to the focus groups’ own framework conditions. Based on these assessments the strategy planner suggests a top-ten list of the best measures for the design of transfer.
3.3 Transfer Level of Funding Priority Same as in 2007 and 2008, StArG prepared, organized and held an Annual Conference of the funding priority. Matters of content were directed by an editorial team consisting of representatives of the funding priority. 130 participants congregated to work out, discuss and pass the Aachen Impulse “Realign Prevention Research – Strengthen Innovative Ability”. The content-related dialog forums discussed the diverse building blocks of the Impulse and revised and perfected them within the individual forums. These revised text parts were then introduced to the plenum for a second discussion and final adoption. Through this means, the largest possible consent on the Impulse Bulletin could be guaranteed so that the funding priority was able to present research successes and further research demands in unison. This is a signal that an interdisciplinary research community on topics of “Preventive Occupational Safety and Health” has indeed been established over the last three years and stands ready for future challenges. Another transfer instrument made available by the meta-project is the “Interactive Work and Discussion Platform” (iDA). Having been originally intended as an instrument facilitating cooperation within projects and focus groups by means of a common data management system, directory management, event calendar and a pin board of the Project Management Agency, iDA and its underlying concept has evolved into a transfer instrument across founding priority boundaries. Since October 2009, all funding priorities of the program “Working, Learning, Developing
1
The position papers can be obtained at www.starg-online.de.
How to Structure and Foster Innovative Research
11
Skills. Potential for Innovation in a Modern Work Environment” can use iDA as a one-stop access to results of the various projects. Maintenance of the iDA is carried out by the individual meta-projects
3.4 Transfer Level of Research Program In order to bring the contributions of the funding priority and its integrated projects to the light of day, various transfer activities have been initiated and carried out by the meta-project in 2009. The German Federal Ministry of Education and Research held the 2nd Future Forum on Innovative Ability at the beginning of April. Here3, the meta-project conceptualized, organized and accompanied the exhibition of the various focus groups. Additionally, the actors of the funding priority took part in both the forum and the diverse workshops. The meta-project accomplished a targetgroup-tailored placement of the 2nd Conference Transcript during the Program Conference. 32 expert articles were published mirroring the current state of research. Furthermore, the Project Management Agency holds a series of Workshops for Junior Scientists hosted by the funding priorities. This facilitates the discussion of dissertations by junior scientists working in current R&D programs, the use of synergy effects and the sharing of experiences
4 Funding Priority and the learning program As has been suggested above, the funding priority is part of a current research program of the German Federal Ministry of Education and Research. This program explicitly views itself as a Learning Program ([BMB07]:9) and thus requires the individual funding priorities to get into a feedback loop with the overall program. This feedback loop has been constituted via the monitoring project IMO (“International Monitoring”) (cf. Fig. 5). As a meta-project, StArG has the task to strengthen the ability to do “Joint Research and Work On a Thematic Focus” in a way to facilitate addressing a common vision to research funding when entering the phase of sustainment. By way of joint research work in the focus groups and the interaction on funding priority levels the various actors from a multitude of scientific disciplines, different branches of work and numerous target groups were able to establish an inter- and transdisciplinary research community in the area of “Preventive Occupational Safety and Health”. In order to optimally moderate this process it was a prerequisite for success to have a meta-project running the systemic and strategic management of the funding priority. The very construct of this funding priority with all its underlying structures and processes made it possible to pass on condensed information in the form of the Aachen Impulse to the Monitoring Project and thus to the Learning Program.
12
Ursula Bach et al.
Fig. 5 Structure of the Learning Program "Working, Learning, Developing Skills. Potential for Innovation in a Modern Work Environment"
References [Aea04]
D. Ahrens and et al. Phasen der Netzwerkentwicklung und des Netzwerkmanagements. In R. Oertel and F. Hees, editors, Das Netzwerk-Kompendium – Theorie und Praxis des Netzwerkmanagements, pages 17–25. Aachen, 2004. [BMB05] BMBF. Bekanntmachung zur Förderung von Forschung und Entwicklung auf dem Gebiet “Präventiver Arbeits- und Gesundheitsschutz”. www.bmbf.de/foerderungen/ 4655.php, downloaded December 1, 2009, 2005. [BMB06] BMBF. Bundesbericht Forschung 2006. Berlin, 2006. [BMB07] BMBF. Arbeiten – Lernen – Kompetenzen entwickeln. Innovationsfähigkeit in einer modernen Arbeitswelt. BMBF-Forschungs- und Entwicklungsprogramm. Berlin, 2007. [HLB09] F. Hees, I. Leisten, and U Bach. Strategischer Transfer im Präventiven Arbeits- und Gesundheitsschutz (Metaprojektbroschüre). Aachen, 2009. in preparation. [HLBH09] K. Henning, I. Leisten, U. Bach, and F. Hees. Präventionsforschung und unternehmerische Praxis: Zwei Seiten einer Medaille. In Innovationsfähigkeit stärken – Wettbewerbsfähigkeit erhalten. Präventiver Arbeits- und Gesundheitsschutz als Treiber, pages 12–31. Aachen, 2009. Henning K.; Leisten I.; Hees F. eds. [Hue03] H. Huemer. Wissensnetzwerke als forschungspolitische Instrumente. In Graggober and et. al., editors, Wissensnetzwerke, Konzepte, Erfahrungen und Entwicklungsrichtungen, pages 115–130. Wiesbaden, 2003. [Jan04] Christoph Jansen. Scorecard für die Wissensmanagement-Performance in heterogenen Unternehmensnetzwerken. Düsseldorf, 2004. [Lea04] E. Lübcke and et al. Netzwerkfähigkeits-Check für Unternehmen. In R. Oertel and K. Henning, editors, Das Netzwerk-Kompendium – Theorie und Praxis des Netzwerkmanagements, pages 39 –49. Aachen, 2004. [OH04] R. Oertel and F. Hees. Das Netzwerk-Kompendium – Theorie und Praxis des Netzwerkmanagement. Aachen, 2004.
How to Structure and Foster Innovative Research [PA06]
[pt208]
[Sau05] [vA84]
13
U. Pröll and U. Ammon. Selbstständig und gesund, Prävention und Gesundheitsförderung bei selbstständiger Erwerbsarbeit. www.gesundheit-unternehmen.de/ 04_Service/Downloads/4_2006_selbststaendig_und_gesund__Werkstattbericht.pdf, downloaded November 19, 2009, 2006. PT im DLR: Themenheft Präventiver Arbeits- und Gesundheitsschutz. www.zlwima.rwth-aachen.de/forschung/projekte/starg/download/PAGS_Themenheft_2008.pdf, downloaded November 10, 2009, 2008. Bonn. J. Sauer. Förderung von Innovationen in heterogenen Forschungsnetzwerken und Evaluation am Beispiel des BMBF-Leitprojektes SENEKA. Aachen, 2005. H. von Alemann. Der Forschungsprozess: Eine Einführung in die Praxis der empirischen Sozialforschung. Stuttgart, 1984.
Innovation Rules 2030 Ursula Bach, Klaus Henning
Abstract Innovative research structures provide the possibility to address new stakeholders on the topic Preventive Occupational Health and Safety. Therefore the stakeholder to the Preventive Occupational Health and Safety can be appealed, as for example health insurance schemes, chambers of commerce or ministries. To reach the aim “better visibility of the research community”, “improving integration of the partners in the research groups” and the “avoidance of the “fragmenting” of a research community”, different network management methods were applied in the founding priority Preventive Occupational Health and Safety. The choice of the methods of the network management bases on the needs of a research network in three specific network phases. Every phase puts different challenges to the network management. Keywords Innovation · social innovation · future · working · learning · developing skills · 2030 · modern working environment
1 Challenges of a Modern Working Environment At the present time, all modern societies experience changes in socio-economic terms and go through far-reaching processes of change in economical and social structures. Examples of dilemmas which characterize these processes are tightened stress of competition, short product and innovation cycles, a continuous adjustment of qualification standards and an increase of flexible, unsecured employer-employee relationships [HL07]. Even Germany and Europe have to face these challenges to stay competitive on the global markets of the 21st century. Under these conditions, the future scenarios
agree on the perspective 2030 (Fig. 1): China, India, Americas and Europe together with Russia will be the new economic area of the future [Sea04, BK07]. Most likely, Europe will only be able to establish themselves to a worldwide main economic area in close cooperation with Russia, at the present time one of the most dynamic markets. How can Germany, as a part of the European Union, get to play an important role in this contradictory context furthermore? Which future perspectives have to be considered thereby? In order to keep Germany competitive in the perspective 2030, the expected tendencies of future social, economic and political developments have to be considered. In the following, the framework of requirements will be discussed in three clusters of innovation in the future: Working 2030, Learning 2030 and Development of competencies 2030.
2 The Future is Coming – Generation innovation 2030 2.1 Working 2030 – How may we enforce innovation! Only unique enterprises in Germany are able to export world wide The development, communication and retaining of unique factors of an enterprise are the core of innovation for new processes and products. Intersectoral cooperation conduce the definition of uniqueness and, moreover, they can recreate new milieus of innovation trough the use of learning aptitude of inter-organizational relations.
Innovation Rules 2030
17
In Germany, we can find such milieus of innovation in an unique way by looking at the owner managed companies, especially in the subcontracted supply of the automotive sector, mechanical and plant engineering [Rem07]. Internal and external mobility become the crucial competitive factor Mobility and transport always refer to individuals, goods and information. However, the means of transportation were broaden. Mobility does not only occur on rails, roads, on waterways and in the air, even the part of data and mains supply increases fast. The importance of mobility of employees, border-crossing cooperation and the reduction of language barriers will increase against the background of the European processes of enhancement [Mic07]. Nomadic working people will displace livelong jobholders Besides occupational mobility, employees in Germany are also characterized by an ongoing flexibility. Self-employment, free-lanced project work, temporal unemployment: Due to new types of employment and fixed term jobs, the ”job for live“ will become an exception and the secondary job besides the part-time employment will be standard. Many employees will increasingly become nomadic working people, multiple jobbers or mini jobbers, as well as precarious employees. But precarious employment is only parlous against the background of the idea of live-financing full-time work. Without the long-beared association „employment means security“, it is simply normal [Opa06, LS04, Gor00, BK07]. Germany needs knowledge management for innovations Education and knowledge are Germany’s essential location economies and the very factors of production which can not really be imitated by others. Therefore knowledge management is a strategic factor of success for the location Germany. In this context, a crucial point is to share knowledge in certain phases and be creative in global networks at the same time. A balance between the poles „Sharing knowledge“ and „Hiding knowledge“ has to be found [FWK+ 03] (Fig. 2). Out of this, a dynamic of innovation develops which is called „Business oriented Familiarity“.
2.2 Learning 2030 – How will we learn to innovate in the future? Learning in the process of working – a normal case for employees and enterprises The learning culture of an enterprise is an essential factor for its competitiveness. Learning has to be increasingly integrated in the process of working to advance the competencies of the employees as source of innovation. Learning on
18
Ursula Bach et al.
Fig. 2 Functional familiarity [HS06]
the job reduces therefore the effort of transfer in the everyday working life. It stands out because it orients itself to the real processes of change. It is also integrated into those and uses the corresponding potentials. In this context, employees and executives develop their competencies relating to the particular potential, problem or situation. However, this presupposes not only the further design of existing types of work and learning arrangements but also demands the establishment of skillsupporting cultures of learning and social relationships in a more comprehensive way. Against this background, further efforts are necessary to enable new learning situations on the job [Rea06].
Relearning as a life-long professional development The acceleration of the increasing of know-how and the social processes of change require a continuous development of competencies. Self-directed processes of learning get thereby a stronger weighting. Regarding living and learning, the individual of the future has to become active in an interpreneurial way. Personal key skills (i.e. the ability of self-organisation, creativity, emotional stability) have to reach the same level of valuation as professional qualifcation. Against the background of the demographic change of a “society of long live”, the necessity of life-long-learning is going to be amplified. For this reason, strategies of further education have to be structured in a way which allows also the educational disadvantaged persons to broaden their competencies. The forms of learning have to offer thereby different accesses for a wide range of target groups [Opa06].
Innovation Rules 2030
19
How fast can you unlearn? The expiration time of knowledge increases to shorten. Therefore, individuals not only have to practise learning, but also unlearning. Unlearning provides new energy that leads to expansion of creativity and therefore to knew knowledge. Basically, advancement consists in the capacity to unlearn, a goodness which obtains increasingly less in the management and is not imparted by our present education system. To acquire old and learn new knowledge and look for knew solutions permanently should therefore be essential attributes of modern business management. However, unlearning has not to be put a level with erasing computer data, it rather means the producing of new significances. In systems apart from the balance, constant unlearninig is the precondition to create new structures and to ensecure the survivability. Therefore, unlearning also means acceptance of chaos and imbalance, because a balanced system is basically not in need of new knowledge to describe its status [NOOS07].
2.3 Development of Competencies 2030 – How can we keep the innovation ability in the future? Dramatic fusions of technology enforce new ways of developing competencies for innovation A new culture of integrated learning, working and providing, needs appropriate technologies. The dramatic fusions of technology enforce new ways of developing competencies to create innovations and the dematerialization of technique brings new chances for innovations through integrated working and learning. In this process, the dimension of expected fusions of technologies – e.g. in the contradictory contexts between the variety of composition, material selection particularly in polymer materials, integration of nano-materials or materials with integrated smart computer devices – can hardly get overrated [Mic07]. Without „older people“ the change to more innovation in Germany can’t be made The demographic change points at the need to use the potential of older people appropriate. New models can bring a balance between the diverse requirements and possibilities of the participants. The process of aging is therefore seen as a perspective of organization for the over internal politics of the development of competencies. Analyzing aging, health and development of competencies in the connection and in their interrelation, is an essential research and activity field. Furthermore, the organization of the transfer of know-how to young employees gets more and more an important role. Even here you can see the growing together of the fields of learning and working. Trough appropriate methods and instruments
20
Ursula Bach et al.
the potential of old and young employees has to be used and at the same time made useful for the development of competencies of all parties [LS04].
3 How can we prepare ourselves for the innovation future 2030? With respect to this background, from the point of the national level, an objective has to be to get Germany a moving and steering role inside the European Union. The function of Germany as one of the economic and innovation driver of the European Union and the world market has to be kept. What can Germany do to make this happen? This can only happen if knowledge, in the parts in which Germany is traditionally strong, will be generated and transferred in much quicker cycles into products as well as services. The scientific focus will lie, among other things, on the machine and plant manufacture, the automotive industry, the energy management and the electro technology [GRRW06]. In this regard, the developmental dynamics of the owner-managed companies in Germany is essential, they provide over 75% of all jobs and most of them are small and medium sized companies [fMB06]. These calls for the understanding that knowledge in and out of these parts, is the essential factor of production and growth with these Germany can stay competitive [SUH+ 03]. Not only the pure quantity definition of the EU underlies the description of small and medium sized companies, it also contains specific attributes: Small and medium sized companies are characterised by • the unity of risk and administration, • the unity of independence of a choice as well as the bearing of responsibility, • the unity of economic existence of the owner and the existence of the company. These companies especially stand for the perception of social responsibility for the employees, for the successful integration of working and learning processes and for the connection with the location Germany. If you now ask how and where these processes get realized and what the formula of success for the creativity and innovation at the location Germany actually is, which lead Germany to the today’s strengths, you can’t easily pass the considerations of owner managed companies, which are so typical for Germany. The topic innovation and creativity mainly depends on how we are dealing with this corporate elite – with these humans who form with their (micro-)enterprises the business location and constantly have to find answers to the challenges in the context of working, learning and developing competencies (Fig. 3) [SUH+ 03]. This elite is affected by the following behavioural characteristics: • They think and plan in global structures and know the strengths of the regional innovative milieus [Hun03]. • They know that the future is shaped by dynamic biographies and constantly relearning will be the normal case.
Innovation Rules 2030
21
Fig. 3 Micro-enterprises in the turbulent environment
• Also generated through the I&K-Technologies the parts of working, learning and all other parts of life reach to one another. Multi-Jobs will become the rule. • Finally multicultural competencies will increasingly affect working and learning in teams [HI97]. On the global future markets, only those companies will be successful which affect the process of innovation effectively through the micro enterprises and focus on the process of these products and services in which we have already the world market leader at the location Germany, following the principle: Strengthen strengths. Such companies thereby are not only characterized themselves by their ability to identify innovation potential in early stages but they are also able to realize and use potentials systematically. The interaction between innovation management and competition is known by many companies. Yet in the constitution of the innovation management are immense unused potentials. While some companies intuitively incorporate this context into their entrepreneurial skills others choose the way of total standardisation of their innovation process. Often the strategic direction of an important und well categorized process of innovation does not define the action, but also the daily stress and the boundedness of resources [SUH+ 03]. This dilemma is the starting point and the challenge of scientific investigations in this field: To identify the worldwide practices of innovation processes which can reach under certain circumstances best possible results.
4 International Monitoring – How to achieve (global) competitiveness in innovation processes? A large research project International Monitoring – IMO sponsored by the German Federal Ministry of Education and Research to the research program „Working – Learning – Developing competencies; Potential for Innovation in a modern Working
22
Ursula Bach et al.
Fig. 4 Fields of Action (FA) of the International Monitoring for Best-Practice of Innovation Processes
Environment“ concentrates on this point and focuses on the below mentioned three objectives including the corresponding fields of action (Fig. 4).
4.1 Continuous acquisition and evaluation of national and international trends of the working, learning and innovation research – methods and results In the field of actions International Monitoring of Dilemmas (FA 1) and National Monitoring of Dilemmas (FA 5) the existing and future trends will be identified and evaluated in half-yearly survey periods in the form of a Delphi-study (cf. www.futureofworkandlearning.com). The questionnaire was applied as an explorative study, starting off an entire series of investigations accompanying the research program. With the help of the answers provided by experts, the conflict areas described in the research program are to be assessed, their actual significance determined on the basis of today’s research activities and, through denomination of future research requirements, additional demands for action are indicated. For this reason, and in order to exploit the research field in width, the main focus was placed on answers obtained through open questions. Hence, the main questions were:
Innovation Rules 2030
23
• Which subjects in the area of “Working, Learning, Developing Skills. Innovative Ability in a Modern Work Environment” currently influence your work? • Which problem fields / challenges in the area of “Working, Learning, Developing Skills. Innovative Ability in a Modern Work Environment” are especially pressing today? • Which research gaps in the area of “Working, Learning, Developing Skills. Innovative Ability in a Modern Work Environment” do you consider to be relevant in the coming years? • How do you assess the relevance of these conflict areas with respect to the creation of work and learning processes during the next ten years? Do you see additional areas of conflict which will become important in these ten years? The survey was carried out online to guarantee worldwide accessibility. An invitation to participate was sent via email to a total of 1,100 experts worldwide. The questionnaire was accessible from January to May, 2008, and available both in German and English. From altogether 214 returns, 140 questionnaires were filled out completely. The majority of participants came from Europe, 54 of them from Germany. In addition to the questionnaires, 18 in-depth interviews were carried out with the assistance of national experts.The majority of participants came from the field of Science. On the basis of constantly occurring trend workshops with national and international outstanding experts of Universities, Companies and intermediary Organizations research gaps will be shown in the subject area „Working – Learning – Developing competencies; Potential for Innovation in a modern Working Environment“ in the framework of the Trend Studies (FA 6). Besides the workshop, constantly trend studies of participating experts will be developed e.g. “Homo Zappiens and its Consequences for Learning, Working and Social Life“ of Prof. Dr. Wim Veen, “Open Innovation with Customers – Foundations, Competences and International Trens” of Prof. Frank Piller.
4.2 Identification and activation of best-practice-methods for the reduction of dilemmas – methods and results In the field of action Best-Practice (FA 2), with the help of different national and international practitioners a criteria catalogue for the identification of Best-PracticeSolutions in the subject area „Working – Learning – Developing competencies” will be developed, validated and constantly verified. This will be raised through different Innovation-Workshop, which takes place in several German companies, e.g. AIXO, Frötek or Nabaltec. Half-yearly Excursions (FA 3) to companies round the globe assure thereby the exemplification and internalization of used Best-Practice-Solutions. In the context of the excursions, through topic-centred workshops with the participants and companies, courses of action will be deduced off for the political, economical and scientific target group in Germany.
24
Ursula Bach et al.
The first national excursion to FESTO took place in July 2008. Main topics were the “Theory of Constraints” and the “Managements of Competencies” of the FESTO Company. Within the scope of the international excursions the companies Unichimtec (Russia) in the year 2008 and Tata Motors in the year 2009 will be visited.
4.3 Ensuring transfer results for the political, economical and scientific target group – methods and results The main focus of the action fields International Panel (FA 7), Expert Study Groups (FA 8) and Summer- and Winterschool (FA 4) of the research project International Monitoring is the placement and transfer of results. In the framework of half-yearly occurring international Expert Study Groups for the political, economical and scientific target groups, topic oriented Summer- and Winterschools for young professionals and researchers and four spin off Expert Study Groups, a continuous exchange and critical discourse with questions about the constitution of a modern working environment will be established. Since today three International Panels with respect to the German innovation research took place in Berlin and close to the City Aachen. The first congress of the Expert Study Group happened to be also in May 2008. Four national spin-offs could be founded, so the different Expert Groups will be working on the following topics “Intellectual Capital”, “Organisation oriented to Competence of Working Systems”, “Change of Work” and “Management of the Uncertainty”. The first Summerschool worked on the topic “Homo Sapiens vs. Homo Zappiens” in August 2008 in Aix-la-Chapelle. During the Winterschool (March 2009) the focus laid on “Generation Chamaelon Flexibility 2015”. The third School had the topic “The king is dead. Long live the customer! Open Innovation 2015” Simultaneously the results of the research project will be continuously, within the fields of action, validated and verified. Via constantly editing of the intermediate results and the deduction of recommendations based on the eight interdependend fields of action, International Monitoring has the objective to enhance, step by step, the basis for decision of actors in the field of politically and corporate constitutions in the perspective 2020. All the activities and results that are achieved during the IMO are available on www.internationalmonitoring.com.
5 Summary Because of turbulent markets, international processes, continuously made adjustments and further developments, companies in Germany are increasingly faced by global challenges. Those alterations require an increased flexibility and ability for
Innovation Rules 2030
25
innovation besides the companies and their employees in order to position themselves proactively in the international competition under the perspective of the future scenarios of 2020. Thus the outstanding position at the world market in many fields of business should be enlarged and preserved (e.g. mechanical engineering, plant engineering and construction, power engineering and electrical engineering). Here the owner managed companies play a significant role for the location Germany, which provide more than 75% of all working places in Germany. The development of competitive advantages through a company and sector specific innovation ability can only be achieved with the help of relevant strategies in the fields of action of working and learning but in the development of organization and competencies, looking at the background of a company. Precarious employment relationships and the demographic change, relearning and unlearning in the process of work as well as continuing fusions of technologies are, in this connection, only a few framework requirements of the future and already of the present, which companies and science will have to face. In the framework of the project International Monitoring these changes and processes will be analysed and realized by methods of resolution and recommendations. Sustainable monitoring instruments will identify Best-Practice of Innovation Processes on a worldwide basis. Thereby the research activities in the field of working, learning and competence development will be regularly adapted to the worldwide identified tendencies. This will help to enable Germany, also in the future, to stay one of the worldwide “innovation champions” in the long run.
References [BK07]
K. Brühl and I. Keicher. Creative Work. Business der Zukunft. Zukunftsinstitut GmbH. Kelkheim, 2007. [fMB06] Institut für Mittelstandsforschung Bonn, editor. Jahrbuch zur Mittelstandsforschung 2006/1. Wiesbaden, 2006. [FWK+ 03] T. Forzi, K. Winkelmann, S. Killich, C. Chwallek, and H. Luczak. Etablierung der Dienstleistung Wissensmanagement in vernetzten Organisationsstrukturen. In Kooperation und Arbeit in vernetzten Welten – Tagungsband der GFA Herbstkonferenz in Aachen, pages 261–265, Stuttgart, October 2003. [Gor00] A. Gorz. Arbeit zwischen Misere und Utopie. Frankfurt/Main, 2000. [GRRW06] R. Gleich, H. Rauen, P. Russo, and M. Wittenstein. Innovationsmanagement in der Investitionsgüterindustrie treffsicher voranbringen – Konzepte und Lösungen. Frankfurt/Main, 2006. [HI97] K. Henning and I. Isenhardt. Bildungstrends der zukünftigen Dienstleistungsgesellschaft, Aachen, 1997. [HL07] K. Henning and I. Leisten. Lernen und Arbeiten für Innovation: Lust auf Zukunft – zwölf Thesen. In D. Streich and D. Wahl, editors, Innovationsfähigkeit in einer modernen Arbeitswelt, pages 27–37. Frankfurt/Main, 2007. [HS06] K. Henning and R. Schmitt. Beteiligung im Veränderungsprozess. In Tagungsdokumentation „Arbeitsforschung als Innovationstreiber“, Dortmund, 2006. [Hun03] H. Hunecke. Produktionsfaktor Wissen – Untersuchung des Zusammenhangs zwischen Wissen und Standort von Unternehmen. Aachen, 2003. [LS04] H. Luczak and M. Stemann. Alternsadäquate und gesundheitsförderliche Gestaltung der Arbeitswelt: Arbeitswissenschaftliche Strategien. In Tagung des Gesundheitswe-
26
[Mic07] [NOOS07]
[Opa06] [Rea06]
[Rem07] [Sea04] [SUH+ 03]
Ursula Bach et al. sens der Volkswagen AG „Demographischer Wandel in der Arbeitswelt – Arbeitsund Beschäftigungsfähigkeit erhalten und fördern, pages 1–54, Wolfsburg, 2004. P. Micic. Das Zukunftsradar. Die wichtigsten Trends, Technologien und Themen für die Zukunft. Offenbach, 2007. H. Neuendorff, H. Oberquelle, B. Ott, and C. Schlick. Arbeitsgestaltung in der Netzwerkökonomie – Flexible Arbeit, Virtuelle Arbeit, Entgrenzte Arbeit. Baltmannsweiler, 2007. H . Opachowski. Deutschland 2020. Wie wir morgen leben – Prognosen der Wissenschaft, volume 2. erweiterte Auflage. Wiesbaden, 2006. U. Reuther and et al. Lernen im Prozess der Arbeit. Erforschen – Gestalten – Bewerten. In Arbeitsgemeinschaft betriebliche Weiterbildungsforschung (ABWF) e. V. (Hrsg.) Bulletin: Berufliche Kompetenzentwicklung, pages 1–9. QUEM, Berlin, 2006. H. Rempp. Produktion und Integration. Lecture to the 7th Framework Programme of the European Community, Brussels, 2007. J. Scharioth and et al. Horizons 2020. Ein Szenario zum Denkanstoßfür die Zukunft. TNS Infratest Wirtschaftsforschung, München, 2004. G. Strina, J. Uribe, K. Henning, R. Oertel, and I. Isenhardt. Innovationsmanagement – Stand der Forschung, Praxisbeispiele und Perspektiven. In Wissen – Innovation – Netzwerke. Wege zur Zukunftsfähigkeit. Berlin, 2003.
Gestaltungsansätze für ein systemisches Fakultätsmanagement Sabine Bischoff, Paul Flachskampf, Klaus Henning
Zusammenfassung „Wenn du Erfolg haben willst, beim Managen und beim Controlling anderer - lerne, dich selbst zu managen und zu kontrollieren“. (William J. H. Boetchker) Aufgrund gewachsener Anforderungen, Komplexität und Dynamik – auch im universitären Sektor – entstand an deutschen Hochschulen das recht neue Berufsbild des Fakultätsmanagers/der Fakultätsmanagerin. Der vorliegende Beitrag diskutiert, welche Elemente dieses Management besonders beachten soll, um das „‚System Fakultät“‘ auf der Basis der OSTO®-Systemdiagnose diagnostizieren und aufgrund der gewonnenen Diagnoseerkenntnisse steuern zu können. An einem Praxisbeispiel wird aufgezeigt, welche Möglichkeiten in der Kenntnis und Anwendung eines umfassenden Fakultätsmanagements liegen.
1 Einleitung Das zum 1. Januar 2007 für Hochschulen in NRW in Kraft getretene Hochschulfreiheitsgesetz (HFG) trug mit Dienstherreneigenschaft, rechtlicher Selbstständigkeit und Eigenständigkeit in der Wirtschaftsführung zur Autonomie der Hochschulen bei [NRW]. Diese Autonomie findet sich nicht nur in den Zentralen Hochschulverwaltungen, sondern auch innerhalb der Fakultätsverwaltungen, den Dekanaten, wieder. Gerade in den letzten Jahren lässt sich eine Entwicklung beobachten, die sich als Professionalisierung bezeichnen lässt: Statt des nebenamtlich geschäftsführenden Dekans gibt es vermehrt hauptamtliche Personen, die ihre Stelle in Richtung eines umfassenden Wissenschaftsmanagements weiterentwickeln [CHE09]. Charakterisiert ist jedoch auch der universitäre Sektor und somit das Fakultätsmanagement durch Komplexität und Dynamik. Das Ziel der „Professionalisierung“ ist noch in weiter Ferne, noch viele Schritte sind zu gehen. Was fehlt, ist eine systematische Herangehensweise zur Diagnose und Steuerung des Systems „Fakultät“. S. Bischoff (B) Institut für Unternehmenskybernetik e.V., Schurzelter Str. 25, 52074 Aachen, Germany E-Mail: [email protected]
Der vorliegende Artikel nimmt sich dieser Thematik an. Nach einer ersten Herausarbeitung von Rahmenbedingungen in Bezug auf ein systemisches Fakultätsmanagement (Abschnitt 2) erfolgt in Abschnitt 3 die Einführung der OSTO®Systemdiagnose [Hen03]. Diese stellt die theoretische Basis dar und gewährleistet, dass die „Organisation als lebendes System“ [HM00] ganzheitlich diagnostiziert wird. Im Fokus der Diagnose stehen acht Gestaltungselemente als Teilsysteme des Gesamtsystems. Teilsysteme sind lediglich Ausschnitte des Systems, die nur in ihrer ergänzenden Betrachtung das System vollständig beschreiben [Hen92] . Das Zusammenspiel von einzelnen Elementen kann unter dem Begriff Emergenz, einem weiteren kybernetischen Grundprinzip, zusammengefasst werden. Einzeln betrachtet werden: 1. 2. 3. 4. 5. 6. 7. 8.
Soziales Teilsystem Organisationssystem Technisches Teilsystem Informationssystem Aufgabenteilsystem Entscheidungssystem Belohnungs- und Kontrollsystem Entwicklungs- und Erneuerungssystem
Diese Elemente bilden den inhaltlichen Rahmen für die gestaltbaren Anteile einer Organisation und bilden somit Ansatzpunkte für ein systemisches Fakultätsmanagement. Ausgewählt für das Fallbeispiel wurde die Fakultät für Maschinenwesen der RWTH Aachen (Abschnitt 4), an der die Systemdiagnose durchgeführt wurde. Im Rahmen der Exzellenzinitiative/des Zukunftkonzeptes „RWTH 2020 – Meeting Global Challenges“ nimmt sich die RWTH Aachen der Stärkung der universitären Managementstrukturen an. Mit dem Projekt „Fakultätsmanagement“ kommt die Fakultät für Maschinenwesen der RWTH Aachen nicht nur der aktuellen Forderung nach professionellem, systemischem Management nach, sondern erfüllt zugleich eine Anforderung, um im internationalen Wettbewerb bestehen zu können.
2 Die Rahmenbedingungen Um Managementanforderungen an ein systemisches Fakultätsmanagement definieren zu können gilt es an dieser Stelle noch einmal einen Schritt zurückzugehen und sich über einige grundlegende Fragen Gedanken zu machen. Was sind überhaupt die Kennzeichen unserer Moderne? Und wie können Erkenntnisse nun konkret für die Managementforschung festgehalten werden? Ein plausibles Konzept, das auch solch neuere gesellschaftliche Entwicklungen einschließt, stammt von Klaus Henning und Heijo Rieckmann. Rieckmann veranschaulicht treffend, unter Rückgriff auf den gesellschaftlichen Kontext, einige
Gestaltungsansätze für Fakultätsmanagement
29
grundlegenden Kennzeichen unserer Gegenwart. So schreibt er: „Wir haben uns eine komplizierte, labile und nervöse Welt aufgebaut. Unsere Alltagserfahrungen führen uns das mittlerweile Tag für Tag vor Augen. Engmaschige Handels-, Finanz-, Kommunikations-, Energie- und Transportnetze haben inzwischen ganze Kulturen, Religionen, Völker und Institutionen bis auf wenige Flugstunden zusammengerückt“[Rie92]. So rutscht die Welt aufgrund der Möglichkeit ständig erreichbar zu sein immer enger zusammen. ähnlich argumentiert Henning, wenn er feststellt, dass wir in einer Welt wachsender Turbulenzen leben, in der wir tagtäglich mit Massen von Informationen versorgt werden, die wir entsprechend verarbeiten und einordnen müssen. „Immer weniger bleibt stabil – immer mehr wird instabil. Nur wer sich in der Instabilität und im zunehmenden Chaos zurechtfindet hat eine überlebenschance“ [Hen92]. Um auf diese Tendenzen reagieren zu können, müssen diese zunächst etwas schärfer analysiert werden. Henning schlägt einige mögliche Paradigmenwechsel vor: Darunter zählen etwa der Wandel vom monokausalen zum offenen, kybernetischen Denken oder der Wandel von objektiven Kriterien hin zu Beobachtermodellen der Wirklichkeit. Oder treffend: Der Wandel vom kurzfristigen Denken zum langfristigen Sinnhandeln [Hen92]. Zu einer wachsenden Vernetzung kommt eine steigende Dynamisierung der Lebens- und Arbeitsbereiche. Wie kann nun begrifflich mit diesen beiden prägenden Entwicklungen der Moderne umgegangen werden? Rieckmann kombiniert die für ihn prägenden Kernpunkte der Moderne: Dynamik und Komplexität fasst er unter dem Begriff Dynaxity zusammen und verweist somit explizit auch auf die steigende Verschränkung der beiden Begriffe. In seinen Worten: „DYNAXITY ist also das Resultat aus ‚dynamics‘ (Dynamik) und ‚complexity‘ (Komplexität) bei steigender Macht/Ohnmacht/Risiko-Relation“ [Rie92]. Mit steigender Komplexität und Dynamik wächst die Möglichkeit des Eintretens chaotischer Zustände. „Dynaxibility“ ist dementsprechend die Fähigkeit von Individuen und Organisationen, mit hohen Graden an Komplexität und Dynamik umgehen zu können und dabei selbstzerstörerisches Chaos zu vermeiden [HI98]. In Abbildung 1 sind über den Dimensionen Komplexität und Dynamik drei Zonen (statisch, dynamisch, turbulent) definiert, die einen wachsenden Grad an „Dynaxity“ aufweisen. Erst in einer „vierten“ Zone überwiegen chaotische Prozesse [HIZ99]. Mit Hilfe verschiedener qualitativer Daten und Indikatoren lassen sich Systeme (z.B. Unternehmen) und ihre Teilsysteme diesen Stadien zuordnen. In Bezug auf die Anforderungen an ein systemisches Fakultätsmanagement lässt sich festhalten, dass sich die Problemstellung, bspw. die Anzahl der Stakeholder (Lehrstühle, Professoren, Dekanat, Rektorat, Kommissionen etc.) und häufig unklaren Entscheidungsstrukturen, von einer Vielzahl von Wechselwirkungen und damit einer hohen Komplexität und Dynamik geprägt ist. Welche Auswirkungen hat also eine steigende Dynaxity und wie kann auf sie adäquat reagiert werden? Einige praktische Hinweise für Management und Führungspositionen lassen sich mit Hilfe des systemischen Denkens direkt in der Praxis umsetzen [HJM09]. Grundlegend sollte
30
Sabine Bischoff et al.
Abb. 1 Organisationen zwischen Komplexität und Dynamik
Dynaxity akzeptiert werden, da alle komplexen und dynamischen Systeme nicht monokausal erklärt werden können und nicht auf den ersten Blick direkt durchschaubar sind. Aus diesem Grund sollten Widersprüche zugelassen und ausbalanciert werden. „Bei komplexen Problemzusammenhängen sind außerdem Widersprüche an der Tagesordnung. Es ist wichtig diese dann nicht abzuwürgen und/oder zu verdrängen. Oft gilt es die Gleichzeitigkeit von ‚Sowohl-als-Auch‘ Situationen (statt Entweder-Oder) zu akzeptieren“ [HIZ99]. Nachdem nun die Rahmenbedingungen eines systemischen Fakultätsmanagement vorgestellt wurden, wird nun im Folgenden die OSTO®-Systemdiagnose eingeführt und die Anwendung in der Fallstudie (Abschnitt 4) an Hand der Fakultät für Maschinenwesen der RWTH Aachen demonstriert.
3 Die OSTO®-Systemdiagnose Da das OSTO®-System-Modell (OSM) besonders gut geeignet ist für die Durchführung von Systemdiagnosen innerhalb komplexer und dynamischer Organisationen sowie der damit einhergehenden Aufdeckung von Widerständen und Hemmnisse für Veränderungsprozesse [Mic06, Jan03, Ise94] wurde es für das hier vorliegende Beispiel als Denkmodell gewählt. Ebenfalls die durch das OSM stark unterstützte Zentrierung auf die involvierten Personen [Fla09] war in diesem Falle für die Auswahl entscheidend, da das universitäre Umfeld besonders stark durch die Befindlichkeiten von Einzelpersonen, die für das Gelingen eines Veränderungsprozesses entscheidend sein können, geprägt ist.
Gestaltungsansätze für Fakultätsmanagement
31
Die Vorgehensweise des OSTO®-Systemansatzes zur Organisationsentwicklung baut auf einem sechsschrittigen Verfahren auf [HM00]. Die weiteren Schritte des OSTO®-Systemansatzes umfassen eine Analyse, Untersuchung der Veränderbarkeit, die Neugestaltung, Synthese sowie Umsetzung und Erfolgskontrolle [HM00]. Im Folgenden wird aus Relevanzgründen ausschließlich der erste Schritt des Ansatzes - die OSTO®-Systemdiagnose - beschrieben. Dabei werden die einzelnen Schritte vorgestellt. Ihre Durchführung ist im Rahmen von Workshops in den entsprechenden Organisationen möglich.
3.1 Die Diagnoseschritte Mit Hilfe der OSTO®-Systemdiagnose werden durch tiefgehende Hinterfragungen die Wirkungszusammenhänge unterschiedlichster Elemente innerhalb einer Organisation herausgearbeitet. Grundlage der Diagnose sind unter anderem Beobachtungen, Gespräche und Befragungen von Mitarbeitern unterschiedlichster Hierarchiestufen. Abbildung 2 visualisiert die acht Schritte der Diagnose. In einem ersten Schritt findet die Systemdefinition statt. Hierunter fällt die Untersuchung der wichtigsten Beziehungen des Systems zu seiner Umwelt als übergeordnetes System [Mal86]. Die Abgrenzung geschieht über die Festlegung einer Systemgrenze der Organisation. Der Existenzgrund (reason for existing) beschreibt den Grund für die Existenz einer Organisation [Han88]. Er ist auf die Bedürfnisse von Kunden(gruppen) nach
Abb. 2 OSTO®-Systemdiagnose [HM00, RW90]
32
Sabine Bischoff et al.
einem Output, den die Organisation in einer gewissen Qualität erzeugt, zurückzuführen. Dieser Existenzgrund wird im zweiten Schritt der Diagnose bestimmt. Im dritten Schritt wird die Frage nach dem Output der Organisation gestellt. Daran anschließende steht im vierten Schritt die Frage nach dem spezifischen Organisationsverhalten, das für diesen Output sorgt, im Mittelpunkt der Diagnose. Organisationsspezifische Gestaltungskomponenten-/elemente charakterisieren eine Organisation in ihrer Gesamtheit. Einzelne Elemente bilden die inhaltlichen Cluster für die gestaltbaren Anteile einer Organisation und stellen dabei Teilsysteme des Gesamtsystems dar. Eine übersicht aller Gestaltungselemente kann Abbildung 3 entnommen werden. Nach Kenntnis der gestaltbaren Elemente der Organisation werden die dahinter liegenden Strategien und Ziele im sechsten und siebten Schritt der OSTO®-Systemdiagnose beschrieben. Die Strategien fragen dabei nach dem WIE? die Ziele nach dem WAS? Der Abschluss der Diagnose (Schritt 8) bildet die Rückführung im Sinne des kybernetischen Regelkreismodells. „Um zu wissen, ob man auf der richtigen Zielgeraden ist und ob die Outputs auch das bringen, was sie bringen sollen, nämlich Existenzgrundsicherung, Zielerreichung, Zukunftsfähigkeit etc., ist es notwendig, sich jederzeit diejenigen Systemauswirkungsdaten beschaffen zu können, die man braucht, um das System richtig steuern und entwickeln zu können.“ [Hen09]. Ergebnis der Diagnose ist somit eine vollständige Beschreibung des IstZustandes einer Organisation. Das bedeutet, dass durch die Diagnose wesentliche Eigenschaften sowie Stärken und Schwächen herausgearbeitet werden und bekannt sind. Das Management einer Organisation ist somit in der Lage, die diagnostizierten Schwächen als Ansatzpunkte für eine Neugestaltung im Sinne einer Verbesserung des Systems zu nutzen. Als Empfehlung gilt, dass die Stärken nicht nachteilig durch
Abb. 3 Gestaltungskomponenten nach dem OSTO®-Systemmodell [Ise94]
Gestaltungsansätze für Fakultätsmanagement
33
das Management beeinflusst werden sollten. Eine Identifizierung der Stärken und Schwächen ist innerhalb einer Vielzahl von Elementen, die ein System charakterisieren, möglich.
3.2 Die Gestaltungskomponenten Im Folgenden werden die acht Gestaltungskomponenten der Diagnose näher beschrieben: Das Gestaltungselement soziales Teilsystem (Mensch) stellt die menschlichen Rollen, Erwartungen und Bedürfnisse materieller und psychischer Art in den Mittelpunkt. „Darin eingeschlossen ist fernerhin das Geflecht der gefühls- und wertbedingten sozio-emotionalen Beziehungen und Interaktionsbedingungen („Klima“)“ (Rieckmann, 1982). Das Gestaltungselement technisches Teilsystem umfasst alle materiellen und räumlichen Gegebenheiten. Hierunter fallen bspw. die Maschinen, Betriebsmittel, Gebäude etc. Das Gestaltungselement Organisationsstruktur bezeichnet die Beziehungen zwischen den sozialen und technischen Komponenten der ersten beiden Gestaltungselemente. Gänzlich ist hierunter die Ablauf- und Aufbauorganisation zu fassen, die die funktionalen wie auch hierarchischen Verhältnisse in Zeit, Raum und Sache umfasst. Das Gestaltungselement Aufgaben erfasst die Aufgaben, die sich aus Kundenwünschen und –interessen des Systems ergeben. Oftmals widmen sich innerhalb einer Organisation unterschiedliche Abteilungen unterschiedlichen Aufgaben. Bei hoher Komplexität ist eine Differenzierung in Teilaufgaben und Arbeitspakete notwendig. Das Gestaltungselement Entscheidungssystem befasst sich mit den zugrunde liegenden Entscheidungsprozessen, die mit der Komplexität lebender Systeme immer umfassender werden [Gal77]. Wer, wann, wie, wo, mit welcher Hilfe Entscheidungen fällt, wird durch das Element des Entscheidungssystems festgelegt. Es handelt sich folglich um Spielregeln zur Steuerung der Entscheidungsprozesse. Das Gestaltungselement Informationssystem ist eng an das Entscheidungssystem gekoppelt. Im Informationssystem wird festgelegt, welche Person welchem Adressatenkreis über welchen Informationskanal und zu welcher Zeit die Informationen zugänglich macht. Das Gestaltungselement Belohnungs- und Kontrollsystem fokussiert die formelle und informelle Steuerung einer lebenden Organisation innerhalb des Belohnungs- und Kontrollsystems. Das Gestaltungselement Entwicklungs- und Erneuerungssystem legt die Grundlagen für die Flexibilität, die Leistungs- und die Anpassungsfähigkeit der Organisation. Dadurch können Verfahren entwickelt werden, die helfen, Rahmenbedingungen für Veränderungen in soziotechnischen Systemen zu installieren und zu festigen [HM00].
34
Sabine Bischoff et al.
Im Folgenden Abschnitt wird das hier dargestellte Verfahren der OSTO®Systemdiagnose auf die Fakultät für Maschinenwesen der RWTH Aachen angewendet. Die These, dass nur eine umfassende Diagnose und Kenntnis eines komplexen Systems in einer dynamischen Umwelt einem Management hilft, seine Führungsaufgaben vollständig zu bearbeiten, wird mittels des Fallbeispiels gestützt.
4 Fallbeispiel Seit 1870 ist die RWTH Aachen als Zentrum für Forschung und Lehre international renommiert und kann auf eine lange Tradition der technischen Expertise zurückblicken. Das Maschinenbaustudium an der RWTH Aachen ist so alt wie die Hochschule selbst. Die Systemgrenze der Fakultät für Maschinenwesen der RWTH Aachen umschreibt die Fakultät folgendermaßen: Die Fakultät für Maschinenwesen der RWTH Aachen forscht im Jahr 2009 mit fast 50 Instituten und Lehrstühlen im Bereich Maschinenbau und Verfahrenstechnik. Die Fakultät setzt sich zusammen aus ca. 800 wissenschaftlichen Mitarbeitern, ca. 550 Angestellten sowie 470 studentischen und wissenschaftlichen Hilfskräften. über 8.000 Studierenden steht – zusätzlich zu den 51 Professuren — die große Anzahl von 150 Dozenten und Lehrbeauftragten aus der Industrie gegenüber. Sie alle profitieren von der hervorragenden Beziehung zur Industrie. Etliche Public Private Partnership-Konstrukte ermöglichen bahnbrechende Erfolge im Bereich der Forschung (E.ON, RWE, Thyssen Krupp, VW, etc.). Der Gesamtumsatz der Fakultät für Maschinenwesen der RWTH Aachen betrug im Jahr 2008 rund 210,9 Mio. C (Zahlenspiegel RWTH Aachen, 2008). Auf eine Wiedergabe des Existenzgrundes der Fakultät für Maschinenwesen der RWTH Aachen muss, ebenso wie auf die fakultätsinternen Ziele und Strategien aufgrund der der Fakultät zugesicherten Vertraulichkeit an dieser Stelle verzichtet werden. Nur in einem ganzheitlichen Management sind alle beschriebenen Gestaltungselemente klar definiert und von allen Systembeteiligten akzeptiert. Daher gilt es in einem ersten Schritt eine große Bestandsaufnahme vorzunehmen und diese Elemente zu dokumentieren. Im sozialen Teilsystem geht es um die Menschen des Systems. Ein Fakultätsmanagement muss sich daher zunächst ein Bild über alle Menschen (Stakeholder) innerhalb und auch außerhalb der Fakultät inklusive deren Erwartungen machen. Dies sind zum einen die Studierenden in verschiedenen Studiengängen, die die Fakultät anbietet. Zum anderen die Instituts- und Lehrstuhlinhaberinnen/-inhaber (Professorinnen und Professoren), sowie wissenschaftliche und nichtwissenschaftliche Mitarbeiterinnen und Mitarbeiter einer Lehr- und/oder Forschungseinrichtung. Unter einer fakultätsangehörigen Lehr- und/oder Forschungseinheit wird eine Kostenstelle verstanden, die deren räumlichen, funktionalen, aufbauorganisatorischen und verrechnungstechnischen Verantwortungsbereich beschreibt. Ein Dekanat als zentrale Verwaltungseinheit einer Fakultät mit allen Dekanatsmitarbeiterinnen
Gestaltungsansätze für Fakultätsmanagement
35
und – mitarbeitern zählt ebenso zu dem sozialen Teilsystem wie bspw. Lehrbeauftragte und Gast- oder Privatdozentinnen/-dozenten, apl.-Professorinnen/Professoren, etc. Die Industrie wird an dieser Stelle ausgeklammert und der Umwelt der Organisation zugerechnet. Wie dieses Beispiel zeigt, muss das Fakultätsmanagement in der Lage sein, zunächst eine entsprechende Systemgrenze zu ziehen. Die Organisationsstruktur beschreibt die Aufbau und Ablauforganisation sowie die hierarchischen und funktionalen Beziehungen zwischen den sozialen und technischen Komponenten. Alle Menschen der Fakultät finden sich in unterschiedlichen Beziehungen zueinander wieder. Die Professorin/der Professor lehrt die Studierenden in den dafür vorgesehenen Räumlichkeiten (Hörsäle, Werkstätte, Labore). Sie/er leitet die Mitarbeiterinnen und Mitarbeiter der Kostenstelle und ist weiterhin Mitglied zahlreicher Fakultätsgremien. Auch Studierende und Mitarbeiter der Fakultät sind laut Fakultätsordnung in den verschiedenen Kommission und Ausschüssen vertreten, die es für das Fakultätsmanagement zu analysieren gilt. Ob Kommission für Finanzen, Struktur oder Lehre, Fakultätsrat, ältestenrat oder Prüfungsausschüsse – für jeden Bereich existieren bestimmte Zuordnungen und Zuständigkeiten. Abb. 4 zeigt beispielhaft die grobe Organisationsstruktur der Fakultät für Maschinenwesen der RWTH Aachen. Vollständigkeit ist durch diese Abbildung noch nicht erreicht, sie gibt vielmehr einen überblick über die Aufbauorganisation der Fakultät. Allein die Organisationseinheit „Dekanat“ ist durch zahlreiche Untereinheiten (Finanzen, Lehre, Struktur, etc.) charakterisiert. Im Rahmen der Ablauforganisation existiert eine Vielzahl von Prozessen: Beispielsweise gibt es neben den oben angesprochenen Gremien wöchentlich tagende
Abb. 4 Organigramm der Fakultät für Maschinenwesen RWTH Aachen (eigene Darstellung)
36
Sabine Bischoff et al.
Dekanatsrunden (Dekan, Prodekane und Geschäftsführung), Teamsitzungen (alle nicht-studentischen Dekanatsmitarbeiter ohne Dekan) und monatlich stattfindende Postbesprechungen (alle nicht-studentischen Dekanatsmitarbeiter mit Dekan) etc. Auf allen Ebenen der Zusammenarbeit lassen sich darüber hinaus diverse sozioemotionale Beziehungen und Interaktionsbedingungen identifizieren, die wiederum direkten Einfluss auf das Gelingen einer Organisation nehmen. Ein Fakultätsmanagement muss imstande sein, diese verschiedenen Menschen, Einrichtungen und Gremien seiner Fakultät wie auch die menschlichen, oftmals historisch gewachsene, Beziehungen zu (er-)kennen, um systemimmanente Interaktionen richtig deuten zu können. Als weiteres wesentliches Element eines systemischen Fakultätsmanagements steht das technische Teilsystem. Das Management einer Fakultät muss sich einen überblick über das Fakultätsinventar verschaffen. Aber auch die Gebäude, in denen die Menschen der Fakultät arbeiten, gehören zum technischen Teilsystem. Die Kenntnis der fakultätszugehörigen Fläche inklusive der Gebäude spielt bspw. bei der Ressourcenallokation und –kontrolle eine große Rolle. Denn auch Räume sind innerhalb von Fakultäten meist Mangelware; jede Kostenstelle möchte ein möglichst großes Stück vom Kuchen für sich in Anspruch nehmen. Gerade deshalb ist die Entwicklung und Einführung eines Ressourcenallokationskonzeptes von enormer Wichtigkeit. Von besonderer Relevanz ist an dieser Stelle die Kenntnis über die unterschiedlichen Bedürfnisse der Kostenstellen. Das Fakultätsmanagement einer geistes- oder sozialwissenschaftliche Fakultät hat es hier im Vergleich zu einer Fakultät für Maschinenwesen, die technisch geprägt und ausgestattet ist, wesentlich einfacher, da der oftmals sehr divergierende Bedarf an Fläche für Labore, Versuchsstände, etc. meist minimal ist. Die drei vorgenannten Gestaltungskomponenten zu Mensch, Organisation und Technik bilden die Grundlage einer jeden Systemdiagnose. Um einen möglichst hohen Detaillierungsgrad zu erreichen, empfiehlt es sich, auch die weiteren fünf Gestaltungselemente zu erfassen. Die Aufgaben einer Fakultät sind so vielfältig und komplex, das eine Darstellung aller Aufgaben der Beispielfakultät für Maschinenwesen der RWTH Aachen an dieser Stelle nicht zielführend erscheint. Neben Aufgaben in Forschung und Lehre existieren zahlreiche Verwaltungsaufgaben (Berufungsverfahren, Beratung, Personalführung, etc.), insbesondere im direkten Arbeitsumfeld des Fakultätsmanagers/der Fakultätsmanagerin. Es gilt, die Wünsche der Kunden (Studierende, Institutsleitungen, Zentrale Hochschulverwaltung, Mitarbeiter etc.) zu erkennen und die daraus resultierenden Arbeitsaufgaben möglichst effektiv und effizient zu bearbeiten. Nicht alle Aufgaben können und müssen selbst erledigt werden. Eine Orientierung an der Dringlichkeit und Wichtigkeit der Aufgabe ist zielführend. Aufgaben können demnach delegiert (dringlich, aber nicht wichtig) und terminiert (wichtig, aber nicht dringlich) oder eben sofort erledigt werden (dringlich und wichtig). Aufgrund der hohen Komplexität des Systems „Fakultät“ sind auch die zugrunde liegenden Entscheidungsprozesse komplex. Die oben dargestellte Organisationsstruktur (vgl. Abbildung 4) der Fakultät für Maschinenwesen der RWTH Aachen lässt erkennen, in welchen Organisationseinheiten Entscheidungen gefällt und
Gestaltungsansätze für Fakultätsmanagement
37
Spielregeln festgelegt werden. Da es sich bei der RWTH Aachen um eine Gruppenuniversität handelt, sind die Gremien der Beispielfakultät für Maschinenwesen der RWTH Aachen paritätisch aufgebaut. Der Fakultätsrat ist dabei das Gremium mit höchster Entscheidungsbefugnis. Vorbereitet werden die Themen in untergeordneten Kommissions- und Ausschusssitzungen, beratend steht der ältestenrat zur Seite. Nur der Dekan selbst hätte die Möglichkeit, Fragen des Haushalts, wie die Budgetverteilung, selbst zu entscheiden. An dieser Stelle ist es jedoch ratsam, vom Konjunktiv Gebrauch zu machen, denn eine alleinige Entscheidungsfällung würde nicht auf viel Gegenliebe bei allen Gruppen (Professorinnen und Professoren, wissenschaftliche Mitarbeiterinnen und Mitarbeiter, nicht-wissenschaftliche Mitarbeiterinnen und Mitarbeiter sowie Studierende) stoßen. Wie in Abschnitt 3 beschrieben, ist das Informationssystem eng an das Entscheidungssystem gekoppelt. Nach Fällung einer Entscheidung wird im Informationssystem festgelegt, wer wem was wann zugänglich macht. Hierbei ist die Fakultätsmanagerin/der Fakultätsmanager als Kopf der zentralen Verwaltungseinheit Dekanat mit Schnittstellenfunktion und –fähigkeit gefragt. Die Fakultät für Maschinenwesen der RWTH Aachen hat beispielsweise im letzten Jahr einen Server errichten lassen, auf dem sämtliche Protokolle, Rundschreiben, etc. für die Gremienmitglieder und für alle Gruppen tagesaktuell abrufbar sind. Mit einer sogenannten „Freitagsmail“ werden die Kostenstellen über die Ereignisse der Woche informiert. Hinzu kommen die Rubrik NEWS auf der Homepage, ein Veranstaltungskalender, ein Infoscreen, persönliche Sprechstunden- und Beratungszeiten, etc. Ein System zur Steuerung und Kontrolle einer lebenden Organisation wird innerhalb der Gestaltungskomponente Belohnungs- und Kontrollsystem entwickelt. An der Fakultät für Maschinenwesen haben sich vornehmlich in den letzten Jahren der Dekan sowie die Haushaltskommission der Entwicklung eines solchen Systems angenommen. Außerdem wurde im Rahmen der Exzellenzinitiative eine Stelle geschaffen, die das Konzept wissenschaftlich begleitet. Im Folgenden soll das Kernelement des Belohnungs- und Kontrollsystems der Fakultät für Maschinenwesen der RWTH Aachen kurz beschrieben werden: Ausgangsbasis bilden die der Fakultät zur Verfügung stehenden Ressourcen (Personal, Sachmittel, Fläche), die für ein Haushaltsjahr an die Kostenstellen weiterverteilt werden. Zunächst wurde die Entscheidung getroffen, das die Ressourcen - abzüglich eines gebundenen Budgets für strategische Verwendungszwecke und wiederkehrenden Bedarf wie bspw. zur Finanzierung des Dekanats und der Fachschaft – zu einem gewissen Prozentsatz leistungsabhängig an die Kostenstellen verteilt werden soll. Von den zur Verfügung stehenden Ressourcen wird daher zunächst ein prozentualer Anteil als sogenannter Sockelbetrag den Kostenstellen zugeteilt. Um den leistungsabhängigen Anteil verteilen zu können, muss zunächst festgelegt werden, welche Leistung erfasst werden soll. Die Fakultät für Maschinenwesen der RWTH Aachen hat sich darauf verständigt, die leistungsabhängige Ressourcenallokation zu 30% von der Forschungs- und zu 70% von der Lehrleistung abhängig zu machen. Diese Leistungen werden jährlich durch Abfrage der Prüfakten im Zentralen Prüfungsamt, der Drittmittelausgaben bei der Drittmittelstelle der RWTH Aachen sowie
38
Sabine Bischoff et al.
der Publikationsleistung der einzelnen Kostenstellen bei der Hochschulbibliothek der RWTH Aachen erfasst. Im Anschluss erfolgt die Umrechnung in sogenannte Personenstunden je Leistungsart. Die Division der Gesamtpersonenstunden mit den Gesamtplanstellen des wissenschaftlichen Personals ergibt den Mittelwert der Auslastung über alle Kostenstellen. Dieser Mittelwert wird auf 1,0 normiert. Somit lässt sich für jede Kostenstelle die Abweichung von der mittleren Auslastung nach oben oder unten ermitteln. Das leistungsabhängige Budget kann entsprechend der Leistung in Forschung und Lehre verteilt werden.1 Das Entwicklungs- und Erneuerungssystem sorgt dafür, dass die zuvor beschriebenen Gestaltungselemente stetiger Entwicklung, Erneuerung und Anpassung an die Umwelten, unterliegen. Das Fakultätsmanagement hat dabei die Aufgabe der Umweltbeobachtung und muss gleichzeitig dafür Sorge tragen, dass die Veränderungen im System widergespiegelt werden. Das bedeutet, dass an der richtigen Stelle die Entscheidungsträger zu informieren sind, Entscheidungen vorbereitet und im Nachgang umgesetzt werden müssen. Die Kenntnis der fakultätsspezifischen Ausprägungen der acht beschriebenen Gestaltungselemente unterstützt das Fakultätsmanagement, sich in der Welt wachsender Turbulenzen zu Recht zu finden und mit der Komplexität und Dynamik im universitären Sektor umzugehen. Sie befähigen die Fakultätsmanagerin/den Fakultätsmanager zu mehr Dynaxibility und helfen, eine gewisse Struktur in Strukturen und Prozesse zu bringen. Wie bereits erwähnt, wird in diesem Artikel auf eine Wiedergabe der Ziele und Strategien zum Erreichen derselben aus Vertraulichkeitsgründen verzichtet. Wie auch das Entwicklungs- und Erneuerungssystem schließt der Schritt der Rückführung den Regelkreis. Hierdurch wird die Sicherung des Existenzgrundes erreicht und der Zielerreichungsgrad überprüft. Strategien zur Erreichung der Ziele können gegebenenfalls angepasst werden; eine Steuerung des Systems ist möglich.
5 Fazit Der Beitrag hat erste Gestaltungsempfehlungen für eine systemische Herangehensweise im Bereich des Fakultätsmanagements gezeigt. An der Beispielorganisation der Fakultät für Maschinenwesen der RWTH Aachen wurde praxisnah die Herangehensweise für eine Diagnose nach dem OSTO®-Systemmodell dargestellt. Der Vorteil in der Analyse einer Organisation durch die acht Gestaltungselemente des OSTO®-Systemansatzes liegt in der Detailliertheit dieser Vorgehensweise. Neben Spielregeln, Normen, der Organisationskultur oder dem Organisationsklima lassen sich durch die Untersuchung des lebenden Systems die Gefühle und Einstellungen der Systemmitglieder abbilden. Somit wird die psycho-soziale Ebene in 1 Anmerkung: Ein Konzept zur leistungsorientierten Verteilung der Fläche befindet sich noch im Entwicklungsstadium. Hintergrund ist die Tatsache, dass eine jährliche Labor- und Werkstattflächenverteilung nicht praktizierbar ist und die spezifische Ausrichtung in Forschung und Lehre einer jeden Kostenstellen unterschiedliche Ausstattung bedarf.
Gestaltungsansätze für Fakultätsmanagement
39
die Vorgehensweise eingeflochten, die bei vielen anderen Ansätzen oftmals außen vor bleibt. Ergebnis ist eine umfassende Organisationsdiagnose, die die Stärken und Schwächen der Organisation herausarbeitet. Die Gestaltungselemente helfen dabei, Teilsysteme näher zu betrachten. Aussagen über notwendige und/oder mögliche Ansätze zur Neugestaltung des lebenden Systems sind Resultat der Diagnose und Auseinandersetzung mit den charakteristischen, gestalterischen Elementen einer Organisation. Ein Fakultätsmanagement, das sich der OSTO®-Systemdiagnose bedient, hat die Möglichkeit, die ihm obliegenden Führungsaufgaben vollständig, gewissenhaft und mittels einer systemischen Herangehensweise auszuführen. Dem Fakultätsmanagement ist somit ein Diagnosewerkzeug an die Hand gegeben, um in der Welt wachsender Turbulenzen nicht unterzugehen. Darüber hinaus werden mittels der Gestaltungselemente mögliche Stellschrauben aufgezeigt, um mit der ständig zunehmenden Komplexität und Dynamik umgehen zu können. Wie beschrieben sollte auch eine Fakultätsmanagerin/ein Fakultätsmanager nicht versuchen, die Komplexität und Dynamik zu reduzieren, sondern diese zuzulassen und auszubalancieren. Die in dem Beitrag entwickelten Gestaltungsempfehlungen für ein systemisches Fakultätsmanagement können anderen Universitäten als Hilfestellung für die Umgestaltung Ihrer Managementstrukturen dienen.
Literaturverzeichnis [CHE09] CHE. FakultätsManagement: über FakultätsManagement. http://www.fakultaets management.de, March 2009. [Fla09] Paul Flachskampf. 2009: A line of industry fights for survival - systemic strategy development and implementation using the example of a lead brokerage financial service provider. 2009. [Gal77] J. R. Galbraith. 1977: Organization Design, Reading, Mass. Addison-Wesley, 1977. [Han88] D.P. Hanna. Designing Organisations for High Performance, Reading, Mass. AddisonWesley, 1988. [Hen92] K. Henning. Zukunftsgestaltung in einer Welt wachsender Turbulenzen. In K. Henning and B. Harendt, editors, Methodik und Praxis der Komplexitätsbewältigung, pages 41–62. Duncker & Humboldt, Berlin, 1992. [Hen03] Renate Henning. Systemisches Management Seminar SYMA: Die OSTO Landkarte als Trainingsbasis für die Einführung von mehrjährigem Change-Management im Zeitraffer. Technical report, 2003. [Hen09] Renate Henning. Change Management - Eine Herausforderung für das Management im und am System. In Klaus Henning and C. Michulitz, editors, Unternehmenskybernetik 2020. Betriebswirtschaftliche und technische Aspekte von Geschäftsprozessen. Berlin, 2009. [HI98] Klaus Henning and Ingrid Isenhardt. Lernen trotz Chaos - Komplexität kreativ nutzen. Lernen im Chaos, Lernen für das Chaos, Heft 52:75–90, 1998. Berlin. [HIZ99] Klaus Henning, Ingrid Isenhardt, and S. Zweig. Zukunftsfähiges Wissensmanagement Sicherung der wirtschaftlichen Entwicklungsfähigkeit in einer ungewissen Zukunft. In Kompetenzentwicklung, editor, Arbeitsgemeinschaft Qualifikations-EntwicklungsManagement, pages 213–250. Münster, 1999. [HJM09] Frank Hees, Sebastian Jursch, and Colin Messerschmidt. Veränderungen strategisch planen wie die Großen: Ein internetbasierter Strategieplaner für KMU. In Klaus
40
[HM00] [Ise94]
[Jan03]
[Mal86] [Mic06]
[NRW] [Rie92]
[RW90]
Sabine Bischoff et al. Henning and Christiane Michulitz, editors, Unternehmenskybernetik 2020. Betriebswirtschaftliche und technische Aspekte von Geschäftsprozessen. Berlin, 2009. Klaus Henning and S. Marks. Kommunikations- und Organisationsentwicklung, volume 6. überarbeitete Auflage. Verlagsgruppe Mainz Aachen, Aachen, 2000. Ingrid Isenhardt. Komplexorientierte Gestaltungsprinzipien für Organisationen - dargestellt an Fallstudien zu Reorganisationsprozessen in einem Großkrankenhaus, volume 9 of Aachener Reihe Mensch und Technik. Wissenschaftsverlag Mainz, Aachen, 1994. C. Jansen. Scorecard für die Wissensmanagement-Performance in heterogenen Unternehmensnetzwerken, volume 1024 of Fortschritt-Bericht VDI Reihe 8: Meß-, Steuerungs- und Regelungstechnik. VDI Verlag, Düsseldorf, 2003. Fredmund Malik. Strategie des Managements komplexer Systeme. Ein Beitrag zur Management-Kybernetik evolutionärer Systeme. Bern/Stuttgart, 2 edition, 1986. C. Michulitz. Kommunikation in Organisationen verstehen - eine systemische Methode. SEM Radar, Zeitschrift für Systemdenken und Entscheidungsfindung im Management, 5. Jahrgang, pages 65–103, January 2006. Landesregierung NRW. Hochschulfreiheitsgesetz (HFG), 31. Oktober 2006. H. Rieckmann. Dynaxibility - oder wie systemisches Management in der Praxis funktionieren kann. In K. Henning and B. Harendt, editors, Methodik und Praxis der Komplexitätsbewältigung, pages 17–39. Duncker & Humblot, Berlin, 1992. H. Rieckmann and P. H. Weissengruber. Manageing the Unmanagable? - Oder: Lassen sich komplexe Systeme überhaupt noch steuern? – Offenes Systemmanagement mit dem OSTO-Systemansatz. In H. Kraus, N. Kailer, and K. Sandner, editors, Management Development im Wandel, pages 27–96. Wien, 1990.
Prävention und Innovation - Strategische Ausrichtung, aktuelle Fragen und Ausblick Klaus Henning, Ursula Bach
Zusammenfassung Die Innovationsfähigkeit des Arbeits- und Wirtschaftsstandortes Deutschland mit seinen Technologien, leistungs- und wettbewerbsfähigen Unternehmen sowie seinen kompetenten Menschen erfordert betriebliche Prävention und eine humane Arbeitsgestaltung, die nachhaltig in der Praxis verankert ist. Sie zielt auf die Erhaltung der Kreativität und Arbeitsfähigkeit der Menschen in einer Arbeitswelt, die durch dynamische, vernetzte Arbeitsformen im demografischen Wandel geprägt ist. Schlüsselwörter Prävention · Innovation · Unternehmensstragien · Arbeitsgestaltung
1 Einleitung Was verraten uns die Zeitungen, wenn wir sie in diesen Tagen aufschlagen: Die Demografie-Falle schnappt zu, denn es gibt zu wenige Kinder. Schulen müssen geschlossen werden, der Arbeitsmarkt wird immer globaler und Sozialpläne sind auf nationaler Ebene wenig wert. Wir lernen aber auch, dass in diesem Zusammenhang die Pflege der Stärken unseres Landes unzureichend ist. So waren z. B. mitten in der aktuellen weltweiten Finanzkrise 69.000 Ingenieurarbeitsplätze unbesetzt und auch der Bedarf an Berufstätigen im sozialen Bereich stieg weiter an [IAB09]. Bei dieser Entwicklung merken wir, dass wir den enormen Konsequenzen des demografischen Wandels proaktiv begegnen müssen und z. B. Migrationsbewegungen qualifizierter Arbeitskräfte in Deutschland brauchen.
Im folgenden Beitrag möchten wir folgender gedanklichen Linie folgen: Warum ist Deutschland eigentlich trotz der weltweiten Krise relativ gesehen so erfolgreich? Wie sieht in diesem Zusammenhang das Verhältnis zwischen Innovation und Prävention aus? Was ist eine erfolgreiche Innovationsstrategie und wie lässt sich daraus eine Präventionsstrategie ableiten? Und welche Konsequenzen lassen sich daraus ziehen?
2 Warum sind wir in Deutschland so erfolgreich? Deutschland ist schon immer ein Produktionsland der Teile und Komponenten gewesen, z. B. werden in Deutschland wesentliche Teile von Computern, Dieselmotoren und Fernsehgeräten hergestellt [Sim07]. Viele der großen technologischen Erfindungen, die in dem Massenartikelbereich aufgegangen sind, wurden in Deutschland erfunden, z. B. MP3- oder Faxgeräte (vgl. exemplarisch [BK08]). Trotz der Anlieferung verschiedener wichtiger Einzelkomponenten, ist das daraus erwachsende System selten unter deutscher Regie weiterentwickelt und in den Markt eingeführt worden. Dies ist aber letztlich nicht entscheidend, denn das jahrhundertealte Label „Made in Germany“ hat sich in den letzten Jahren zunehmend zu „Enabled by Germany“ weiterentwickelt [HHBH09]. Die beiden Labels beziehen sich auf die Kombination der drei Ebenen Handwerkskunst, Kaufmannskunst und Ingenieurskunst. Mit diesen drei Ebenen wird ein Verständnis von „Made in Germany“ und „Enabled by Germany“ geschaffen, das sich nicht auf die Profession der Ingenieure beschränkt, sondern auf die tatsächliche Kunst des „Engineering“. Warum sind diese drei Ebenen wichtig? Kaum ein anderes Land dieser Welt baut in solchem Umfang auf den historischen Traditionen der Handwerks- und Kaufmannsgilden. Im Mittelalter konnten durch die genossenschaftsartigen Zusammenschlüsse Einzigartigkeit gewährleistet werden. Sie verbanden solide Facharbeit mit gewerblicher Facharbeit und verbanden diese zu einem „europäischen Kapitalismus“, dem sogenannten Rheinischen Kapitalismus [Alb92]. Diese Form des Kapitalismus verfolgt nicht ausschließlich monetäre Ziele, sondern pflegt das humanistische Bildungsideal, wie z. B. Wertschätzung der Kunst, der Wissenschaft und der politischen Verantwortung [Kae01]. Das wirtschaftliche Rückgrat in Mitteleuropa ist die Kombination zwischen den oben genannten drei Ebenen. Mangel auf der Ebene der Ingenieurkunst könnte man langfristig verstärkt über internationale Kooperationen beheben. Allerdings ist unser Alleinstellungsmerkmal vor allem durch Mangel auf Ebene der Handwerkskunst, also den Facharbeitern mit handwerklicher oder industrieller Prägung, bedroht. Die Ausbildung zum Facharbeiter im mitteleuropäischen Raum ist alternativlos auf der Welt. Mit diesen Erfolgsfaktoren ist ein enorm hoher Spezialisierungsgrad verbunden, wie z. B. die Konstruktion und Herstellungsverfahren von Ventilteilen oder von Kolben für Motoren aller Größenordnungen. Dies erfordert gleichzeitig einen außerordentlich hohen Grad der Vernetzung- und Internationalisierung der deutschen Wirtschaft. Ein gutes Beispiel hierzu liefert die Firma Neumann und Esser in Alsdorf nördlich von Aachen. Sie stellt Kompressoren her und ist weltweit einer von
Prävention und Innovation
43
Abb. 1 Firma Neumann & Esser als Beispiel für Internationalisierung und Vernetzung in der Region Aachen
den wenigen Lieferanten für wartungsfreie Kompressoren und beliefert damit die ganze Welt (vgl. Abbildung 1). An diesem Unternehmen ist die Wirtschaftkrise fast unbemerkt vorbeigegangen. Es lebt davon, dass die handwerkliche Fertigung für Gussteile ins Erzgebirge ausgelagert wurde. Dies ist einer der wenigen Orte in der Welt, an dem qualitativ-hochwertige Großgussteile hergestellt werden können. Bei Neumann & Esser, werden diese Teile dann so weiterverarbeitet, dass sie als wartungsfrei gelten. Die Zubehörteile des Kompressors werden dabei oft in anderen Teilen der Welt „hinzugefügt“. Nur der Kern des Kompressors kann nach überzeugung der Eigentümer ausschließlich am Standort Deutschland in der erforderlichen Qualität gefertigt werden. Dies zeigt, dass ein so kleines Unternehmen in einem Spannungsfeld dramatischer internationaler Vernetzung lebt und damit gewinnt das Thema Prävention und Innovation notwendigerweise eine internationale Dimension. Dies ist eben nicht nur eine Frage der großen Konzerne, sondern betrifft die gesamten innovativen extrem exportabhängigen, inhabergeführten mittelständig geprägten deutschen Unternehmen, die in der Regel in einer Kombination aus Produktskernen und damit verbundenen umfangreichen Dienstleistungen bestehen. Deutschland wird vermutlich aus der derzeit weiter andauernden weltweiten Finanzkrise als eine der reichsten Nationen der Welt herausgehen, wenn auch ca. 20 % ärmer als 2009. Dies lässt sich u. a. durch die zugrunde liegende Wirtschaftstruktur belegen, die mit einer extremen Technologieorientierung einhergeht
44
Klaus Henning et al. Betriebsgröße MA-Zahl
Bis 19 20 bis 99 100 bis 249 250 bis 499 500 und mehr
Anteil der Betriebe
Anteil der Beschäftigten
89,9 % 8,2 % 1,1 % 0,6 % 0,2 %
Anteil der inhabergeführten Unternehmen
27,9 % 25,5 % 10,4 % 14,5 % 21,7 %
92,0 % 84,3 % 66,2 % 68,9 % 49,1 %
Arbeitsplätze durch inhabergeführte Unternehmen 25,6 % 21,5 % 6,9 % 10,0 % 10,6 % Summe 74,6 %
(vgl. Abbildung 2). 75 % aller sozialversicherungspflichtigen Arbeitsplätze in Deutschland werden durch inhabergeführte Unternehmen gestellt. In diesem Zusammenhang liegt besonderes Interesse auf den Unternehmen, die in den Kategorien 250 bis 499 bzw. 500 und mehr agieren. Sie sind Unternehmen, die mit mittelständischer Mentalität handeln und mit der internationalen Dynamik umgehen müssen. Um diesem Umstand gerecht zu werden, bedarf es einer neuen Kategorie der „zu groß geratenen“, mittelständisch geprägten, international agierenden, inhabergeführten Unternehmen. Ingesamt zeigt das Ergebnis, dass inhabergeführte Unternehmen der eigentliche Jobmotor Deutschlands ist. Denn Konzerne, im Gegensatz zu dieser Unternehmensgruppe stehen, reduzieren strukturell regelmäßig Arbeitsplätze wie zuletzt eindrucksvoll eine Umfrage des Manager Magazin bei 80 börsennotierten Unternehmen zeigte [kle09]. Das heißt der Fokus beim Thema Prävention und Innovation sollte sich verstärkt auf die Wirtschaftsdynamikstruktur der inhabergeführten Unternehmen konzentrieren. Das Zusammenspiel von Prävention und Innovation unterliegt hier zum Teil anderer Gesetzmäßigkeiten und internationale Verpflichtungen und bekommt damit eine andere strategische Bedeutung. Dies soll an drei Beispielen erläutert werden: Beispiel 1:Rosskopf und Partner Mit nur 120 Mitarbeitern, einem Exportanteil von 50%, Sitz im Erzgebirge fertigt Rosskopf und Partner u. a. aus Mineral- und Quarzwerkstoffen frei geformte Flächen für Bäder. Das ist ein weltweites Alleinstellungsmerkmal. Das Unternehmen ist zurzeit einzige Anbieter für z. B. Luxushotels, die ein Bad aus einem Formteil haben wollen. Abnehmer sind Hotels und Hotelketten weltweit von Madrid bis Dubai. Das Unternehmen legt dabei größten Wert darauf, dass die europaweit rekrutierten Montageteams unter den gleichen Arbeitsbedingungen arbeiten können wie die Stammbelegschaft des Unternehmens. Beispiel 2:Knorr Bremse Die Firma Knorr Bremse gegründet 1905 ist immer noch eigentümergeführt. 2009 lag der Jahresumsatz bei C 2,76 Mrd. Das Unternehmen ist für Eisenbahnund LKW-Bremssysteme Weltmarktführer und betreibt ca. 70 Standorte rund um den Globus.
Prävention und Innovation
45
Macht Knorr Bremse Prävention und wie machen sie Innovation? Welche Präventionsmaßnahmen werden an den über 20 Standorten in China getroffen? Knorr Bremse löst dies mit extremer Standardisierung von Prozessen im Bezug auf Arbeit, Sicherheit, IT-Unterstützung, die weltweit extrem überwacht werden. Dies ist eine mögliche Antwort auf das Thema Prävention im internationalen Kontext. Beispiel 3:Dürr Ecoclean Dürr Ecoclean ist ein Unternehmen südlich von Aachen. Es baut Waschanlagen für Motoren, die sich in der industriellen Fertigung befinden. Das Unternehmen hat den größten Weltmarktanteil für diese Waschanlagen. Das Werk bleibt bewusst an dem Standort mitten in der Eifel, obwohl es ein enormer Aufwand ist die recht großen Anlagen, die im Monschauer Tal produzierten werden, auf Schwerlastkraftwagen zu verladen und z. B. bis in die Urwaldregion von Brasilien zu transportieren. Das Unternehmen weiß, dass sich dieser Aufwand nur lohnt, weil sie auf eine solide Struktur von Facharbeitern und Ingenieuren, bauen kann, übrigens in einer Wirtschaftsregion mit weniger als 4% Arbeitslosigkeit. Hier kommt das Thema Prävention und Innovation ganz nah zusammen. Die dargestellten Stärken erhöhen die Attraktivität Deutschlands für ausländische Investoren. Eine Untersuchung von Ernst und Young (2007) beschreibt die Hauptfaktoren, warum ausländische Unternehmen in Deutschland investieren (vgl. Abbildung 3). Folgende Items sind ausschlaggebend: Produktivität, Arbeitseffizienz, Innovationsfähigkeit, Arbeitsumgebung, Organisationsstruktur, Umweltbewusstsein, Umweltschutz, Unternehmergeist, Fähigkeit zur Teamarbeit und nicht zuletzt ethische und soziale Responsibilität. Alle Faktoren, die in der Graphik durch Pfeile gekennzeichnet sind, sind unmittelbar oder mittelbar mit dem Thema Prävention verbunden und gestalten so den Standortvorteil Deutschlands mit.
Abb. 3 Items der Attraktivität deutscher Unternehmer, nach Ernst & Young (2007)
46
Klaus Henning et al.
Was bedeutet dieses auf einer strategischen Ebene? Prävention bewegt sich im Spannungsfeld von sozialen, politischen, ökologischen und ökonomischen Fragestellungen. Prävention hat auf politischer Ebene eine Dimension der Standortpolitik (vgl. Abbildung 4). Prävention hat ebenso mit Nachhaltigkeit zu tun. Der neuere Nachhaltigkeits-Begriff beinhaltet vier Faktoren ökologie, Politik, ökonomie und soziale Verträglichkeit gleichermaßen. Gerade dies wurde z. B. in der Initiative des Deutsch-Indischen Dialogs über Sustainable Solutions als Diskussionsgrundlage gewählt (BMBF´s International Dialogue on Sustainability Research) [bmb]. Prävention ist damit zentraler Bestandteil einer jeden nachhaltigen Entwicklungsstrategie „Enabled by Germany“. Wenn wir heute über den Export von Sustainability reden, ist Prävention ein essentieller und unverzichtbarer Anteil eines solchen Exportproduktes.
Abb. 4 Die vier Dimensionen der Nachhaltigkeit nach Henning 2008
3 Wie sieht das Verhältnis zwischen Innovation und Prävention aus? Die Rahmenbedingungen unter denen wir agieren, werden zunehmend komplexer und dynamischer. Das Phänomen wird durch das Kunstwort „Dynaxity“ bezeichnet, das die Begriffe „dynamics“ und „complexity“ verknüpft. [HI94] (vgl. Abbildung 5)
Prävention und Innovation
47
Abb. 5 Rahmenbedingung von Prävention unterliegen steigender Dynaxity
In der Zone 1 dominieren Kleinbetriebe in traditionellen Strukturen, die auch heute noch eine erhebliche Bedeutung haben. In der Zone 2 spielt sich das gewohnte wirtschaftliche Leben der zweiten Hälfte des 20. Jahrhunderts ab, dessen Dynamik unverändert eine sehr große Bedeutung hat. Diese Entwicklung wird jedoch seit mehreren Jahrzehnten von zunehmend globalisierten zur Turbulenz neigenden Prozessen überlagert. In vielen wirtschaftlichen Bereichen ist diese Zone 3 zum „Normalzustand“ geworden. Die damit verbundene wachsende Dynaxity stellt aber nicht nur ein Management-Problem dar, sondern führt zu sehr tiefgreifenden erhöhten Belastungsstrukturen für alle davon betroffenen Akteure und wird damit zu einem zentralen Thema von Prävention. In diesem Zusammenhang ist der Umgang mit Dynaxity, also die Dynaxability, von ausschlaggebender Bedeutung: • Menschen spielen zunehmend eine größere Rolle als Werkzeuge, • Laufende und funktionsfähige Prozesse spielen eine größere Rolle als eine vollständige Dokumentationen, • Die tatsächlichen Bedürfnisse des Kunden spielen eine größere Rolle als die mit ihm geschlossenen Verträge, • Eine Kultur der Veränderungsbereitschaft und die damit verbundene akzeptierte Flexibilität dominiert über die jeweils gültigen Planungen. Diese Werteabwägung hat die AIXCORE Group Aachen [aix] aus dem Manifest für agile Softwareentwicklung abgeleitet [man]: • Uns sind Individuen und Interaktionen wichtiger als Prozesse und Werkzeuge, • Uns sind lauffähige Prozesse wichtiger als umfangreiche Dokumentation,
48
Klaus Henning et al.
• Uns ist die Zusammenarbeit mit dem Kunden wichtiger als Vertragsverhandlungen, • Uns ist es wichtiger auf änderungen reagieren zu können, als einen Plan zu verfolgen, • Daher messen wir, obwohl die jeweils zweiten Dinge ihren Wert besitzen, den jeweils erstgenannten Dingen höheren Wert zu. Was bedeuten diese Rahmenbedingungen für Innovations- und Präventionsstrategien? In dem bisher diskutierten Kontext geht es bei einer Innovationsstrategie in Bezug auf den Standort Deutschland immer um die Erstellung von Einzigartigkeit [Vol07]. Einzigartigkeit verlangt einen kreativen Innovationsprozess, der beteiligungsorientierte Change Prozesse erfordert und der dann zum Innovationstreiber wird. Einzigartigkeit verlangt ein Verständnis von Innovation, dass immer auch Bruch mit dem Vergangenen ist, welches notwendigerweise liebgewordene Gewohnheiten und Verhaltensweisen verletzt. Innovation in dieser Form muss immer einen Human-Organization-Technology-Ansatz folgen (HOT-Approach), d. h. der Mensch steht im Fokus und seinen Bedürfnissen folgen Organisation und Technik [IB05]. Einzigartigkeit im Sinne des Innovationsprozesses muss berücksichtigen, dass die Technologieentwicklung durch dramatische Technologie-Fusionen beeinflusst wird [HH09]. Unterschiedliche Technologien, wie Nanotechnologie, Kunststofftechnik, Werkzeugtechnik, öl-Hydrauliktechnik, Strömungstechnik, werden auf engstem Raum in immer mehr Bauelementen integriert und unter massivem Einsatz von direkt angeschlossenen IT-Systemen, den sogenannten Embedded Systeme, verwendet. In diesem Kontext wird beteiligungsorientiertes Arbeiten und Lernen in global verteilten Teams zum Normalfall, weil diese Innovationsentwicklungen am Standort Deutschland sich auch auf Grund der demografischen Entwicklung, also des Mangels an verfügbaren Personal, zwangsläufig international oder mindestens im europäischen Kontext vernetzen muss. Nur einzigartige Organisationen in dieser Art sind auf Dauer wettbewerbsfähig, denn sie folgen einem Pfad den North durch seine Wissenstreppe zur Wettbewerbsfähigkeit angedeutet hat (vgl. Abbildung 6). Der Innovationsprozess muss sich konzentrieren auf den impliziten Anteil von Wissen ist und dessen Anwendungsbezug. Der nächste notwendige Schritt ist der Wille zur Anwendung. Eine nächste Stufe beschreibt, dass die Fähigkeit zu handeln, begleitet sein muss von der die richtigen Entscheidungen zu treffen. Diese erst führen mit den produkt- und verfahrensspezifischen Kompetenzen zu Einzigartigkeit. Mit dieser Strategie beim Umgang mit den impliziten Wissensbeständen hin zu einer unternehmerischen Einzigartigkeit, gilt es ein Prozessmanagement aufzubauen, das die drei wichtigsten Kernprozesse – den Aufgabenkernprozess, den sozialen Kernprozess und den individuellen Kernprozesse bündelt und zusammen bringt [HM01]. Bei diesem Innovationsverständnis ist zu fragen, wie eine hierzu passende Präventionsstrategie aussieht. Hierzu hat es in den letzten Jahren umfangreiche Forschungsinitiativen vom BMBF im Förderschwerpunkt Präventiver Arbeits- und Gesundheitsschutz gegeben [iP10]. Diese Forschungsaktivitäten wurden in dem sogenannten „Aachener Impuls“ (www.starg-online.de) zusammengefasst. Hier heißt es:
Prävention und Innovation
49
Abb. 6 Wissenstreppe zur Wettbewerbsfähigkeit [Nor]
„Die Innovationsfähigkeit des Arbeits- und Wirtschaftsstandortes Deutschland mit seinen Technologien, leistungs- und wettbewerbsfähigen Unternehmen sowie seinen kompetenten Menschen erfordert betriebliche Prävention und eine humane Arbeitsgestaltung, die nachhaltig in der Praxis verankert ist.“ (Aachener Impuls 2009)
Diese Formulierung fasst zusammen, was in der Einleitung dieser Ausführungen ausführlich erörtert wurde: Die Charakteristik des Arbeits- und Wirtschaftsstandortes Deutschland mit seinen Technologien, die Triangulation von Handwerks-, Kaufmanns- und Ingenieurskunst und den damit verbundenen speziellen Kompetenzen. Diese Aspekte sind nur dann langfristig Erfolgsgaranten und potentiell exportfähig, wenn betriebliche Prävention eine nachhaltige Dimension in der Praxis hat. Dieser Ansatz zielt auf die Erhaltung der Kreativität und Arbeitsfähigkeit in einer Arbeitswelt ab, die durch zunehmend turbulenter werdende Arbeitsformen im weltweiten Wettbewerb geprägt sind und zusätzlich durch den demografischen Wandel in Deutschland geprägt sind.
3.1 Welche Konsequenzen lassen sich aus dem „Aachener Impuls“ ziehen? Vier wesentliche Konsequenzen lassen sich aus dem Aachener Appell für eine zukünftige Präventionsstrategie ziehen. Sie betreffen • das Verhältnis von Prävention zu Partizipation und Empowerment, • die Einbeziehung des außerbetrieblichen Lebens- und Arbeitsfeldes in den Partizipationsansatz;
50
Klaus Henning et al.
• die Frage der Machbarkeit und Wirtschaftlichkeit präventiver Ansätze im internationalen Kontext und • die Präventionsmaßnahmen in Hinblick auf die Produktsicherheit. 1. Partizipation und Empowerment sind die strategischen Voraussetzungen für handlungsorientierte Prävention in turbulenter werdenden Umwelten. Empowerment bezieht sich hierbei vor allem auf Führungskräfte der mittleren und oberen Ebene einen solchen weltweiten Vernetzungsprozess überhaupt in den Griff zu bekommen. Beispiel 4: P3 Gruppe Insgesamt sind in der P3 Gruppe [p3?] 1.000 Beratern aus 49 Nationen angestellt. Der Umsatz der Firma wächst im Jahr um ca. 20–30 %. In einem Teilbereich der Gruppe sind zurzeit etwa 50 Mitarbeiter aus acht Nationen beschäftigt. Diese Teilgruppe expandiert in den Vereinigten Staaten und baut dort einen neuen Standort für das Testen der Leistungsfähigkeit von Mobilnetzen auf. Dies bedeutet für den Geschäftsführer, dass er trotz seiner Familienbedingungen alle 14 Tage in die USA muss. 2. Prävention muss über die Unternehmensgrenzen hinaus die Wechselwirkung zwischen betrieblichen- und außerbetrieblichen Arbeits- und Lebensfeldern proaktiv gestalten. Es ist absolut unabdingbar, dass sich die unternehmerische Praxis die Lebenswirklichkeit in den privaten Verhältnissen zum Teil zur Verantwortung macht. Bei diesen Neben-, Rück- und Fernwirkungen auf diesem Bereich unter den oben genannten Rahmenbedingungen wird dies zu einer Managementaufgabe. Beispiel 5: Bayer Die Firma Bayer [bay] baut in Shanghai eine der größten Chemieanlagen weltweit auf. In diesem Rahmen werden nicht einzelne Mitarbeiter versetzt, sondern auch deren Familien ziehen mit. Das Unternehmen übernimmt folgende Aufgaben: Arbeitsplatz- und Wohnungssuche für die Familienmitglieder und die Auswahl möglicher Schulen oder Kindergärten. Die Gestaltung der sozialen Rahmenbedingungen wird als Aufgabe des Unternehmens verstanden, damit der Standort in Shanghai betrieben werden kann. Treiber sind hier auch im Ansatz Präventionsaspekte, aber vor allen Dingen deshalb, weil sich die Mitarbeiter sonst weigern umzuziehen. 3. Prävention muss machbare und wirtschaftlich vertretbare Wege für international verteilte Unternehmensstandorte aufzeigen, die die verschiedenen kulturellen Implikationen berücksichtigt. Dies bedeutet: Die in unserem Land hochentwickelten Präventionsstandards lassen sich nicht ohne Adaption in Bezug auf Komplexität, finanzielle Aufwendung und Akzeptanz auf andere Länder übertragen. Es braucht geplante Schritte im Sinne einer Roadmap, um Prävention im internationalen Kontext und deren zugehörigen Normen und Standards im dortigen politischen Raum einzuführen – wie es bundesdeutsche Präventionsvereinigungen bereits tun. Damit wird aber auch normative Prävention zu einem
Prävention und Innovation
51
Exportartikel. So legt sich der Fokus viel stärker darauf, welche PräventionsDienstleistungen wir rund um diesen Globus exportieren. Hierzu ist die Nachfrage aus den entsprechenden Ländern überdimensional groß und kann eigentlich nicht befriedigt werden. 4. Prävention heißt auch, dass die Frage nach der Produktsicherheit nicht nur unter europäischen Einsatzbedingungen geprüft wird, sondern in dem jeweiligen Verwendungsraum des Produktes. Beispiel 6: Verwendung von Flexmaschinen in China Bei der Verwendung von Flexmaschinen zur Durchtrennung von Stahl ist es in China üblich sämtliche Sicherheitsvorrichtungen abzubauen, weil der chinesische Landarbeiter eine kulturellbedingt hohe Fähigkeit hat, Kreisradien manuell zu flexen. Der Radius wird mit dem Daumen festgelegt und per Hand geflext. Der Zusammenhang zwischen den kulturbedingten Kompetenzen und der Notwendigkeit zur Prävention wurde in diesem Zusammenhang bisher nicht untersucht. Das Wertesystem liegt dabei bei dem chinesischen Landarbeiter anders als bei einem deutschen Facharbeiter – hier gibt es ein großes Handlungsfeld für kulturell angepasste Prävention, das durch Dienstleistungsexport von Präventionsdienstleistungen weltweit durch Ansätze aus Deutschland besetzt werden kann. Die logische Schlussfolgerung hieraus lautet: Prävention und unternehmerische Praxis müssen sich begegnen. Die Konzentration muss auf der Frage liegen, welche Präventionsmaßnahmen dem konkreten unternehmerischen Handeln dienen. In der Praxis stimmen die Ziele des Unternehmens und deren Existenzgründe nicht mit denen durch die Steakholder des orientierten Arbeits- und Gesundheitsschutz formulierten Existenzgründe und Ziele nicht überein. (vgl. Abbildung 7) Daraus folgt, dass ein Unternehmen Ziele hat, bei denen Prävention ein Mittel zum Zweck ist, aber nicht das eigentliche unternehmerische Ziel. [HHBH09]. Prävention ist im Sinngrund eines Unternehmens verankert. Selten aber als genuiner Existenzgrund, denn Prävention verursacht dem Unternehmen zunächst in vielen Fällen nur
Abb. 7 Existenzgrund Unternehmen und Stakeholder des Präventiven Arbeits- und Gesundheitsschutz [Hen]
52
Klaus Henning et al.
zusätzliche Kosten. Also gilt es Schnittmengen zu diagnostizieren, die beschreiben, inwieweit sich Ansätze der Prävention in Ziele, Strategien, Gestaltungskomponenten und in den Qualitätssystemen eines Unternehmens integrieren lassen. Dies bedeutet, dass sich präventiver Arbeits- und Gesundheitsschutz auf die Unternehmen zu bewegen muss und nicht umgekehrt. Denn Unternehmensziele können nicht in Präventionsmaßnahmen integriert werden, sondern es geht darum Präventionsmaßnahmen in die Unternehmensziele zu integrieren [Hen]. Damit wird Prävention als Strategie zu einem „Enabler“ für qualitativ hochwertige Unternehmensziele. Zusammengefasst gilt für erfolgreiche Prävention, dass jede Organisation ihr eigener Patient ist. Menschen – egal, ob Führungskräfte oder Mitarbeiter – ändern, ihr Verhalten nicht ohne weiteres: Nur weil Geräuschschutz vorgeschrieben ist, steckt ein gelernter Facharbeiter noch lange keine Ohrstöpsel in seine Ohren. Es geht darum einen Prozess erfolgreicher Prävention „vorzuleben“, bevor man ihn mit formalen Strukturen verankert. Eine Anordnung hilft im Zweifelsfalle oft nicht.
4 Rahmenbedingung „Deutschland, deine nächsten 15 Jahre“ Die bisherigen Überlegungen zu Innovation und Prävention sollen nun in die konkrete Situation Deutschland in den nächsten 15 Jahren eingeordnet werden. Dazu werden zunächst einige wichtige Trends dargestellt, die sich aus den Dialogen im Bundeskanzleramt in den Jahren 2008 und 2009 ergeben haben. Die Dialoge standen unter dem Thema: Deutschland, deine nächsten 15 Jahre. Deutschland wird in diesem Zeitraum unter Berücksichtigung folgender Trends agieren müssen: 1. Mangelware „Junger Mensch“ 2. „Made in Germany“ wird zunehmend ergänzt und ersetzt durch „Enabled by Germany“ 3. Der „Homo Sapiens“ wird durch den „Homo Zappiens“ ergänzt. Mangelware „Junger Mensch“ Den demografischen Realitäten muss mit offenem Visier begegnet werden. Das macht Verhaltensänderungen notwendig. Die „Mangelware junger Mensch“ auf dem Arbeitsmarkt wird es dringend erforderlich machen, dass Menschen, die heute über 65 Jahre alt sind, länger arbeiten werden und können und müssen [Her03], (Rürup 2003, Nachhaltigkeit in der Finanzierung der sozialen Sicherungssysteme) und wir es uns nicht mehr leisten können, Menschen zwischen 50 und 65 nicht im Arbeitsleben zu halten. Was in Griechenland zurzeit passiert, ist ein Trendsetter für die ganze europäische Entwicklung: Weniger Einkommen, länger arbeiten. Es ist eine Illusion, dass wir von dieser Entwicklung bewahrt bleiben. Deutschland hat schon heute trotz Krise und Doppeljahrgängen des Abiturs einen so großen Fachkräftemangel, dass dies nur durch eine professionelle Migrationskultur und -politik gelöst werden kann. Es geht darum, mit den besten Migrationsländern dieser Welt auf Augenhöhe Migrationspolitik zu machen. Die Besten
Prävention und Innovation
53
dieser Welt sind zurzeit Kanada und Australien [Cam04]. Deutschland muss ein Land hervorragender Migrationskultur für Fachkräfte werden, um weiter erfolgreich im internationalen Wettbewerb bestehen zu können. Sonst wird es automatisch zum Export von Engineering-Dienstleistungen ins Ausland kommen. Und das hätte langfristig fatale Folgen für den Industrie und Dienstleistungsstandort Deutschland. In diesem Zusammenhang sei angemerkt, dass auch eine hoffentlich ansteigende Geburtenrate sich erst in 20 bis 25 Jahren auf dem Arbeitsmarkt positiv bemerkbar machen kann [BBLW02]. Von Made in Germany zu Enabled in Germany Der grundsätzliche Wandel der Typologie der Exportgüter aus Deutschland wurde bereits eingangs ausführlich erläutert. Ein weiterer Aspekt ist unser Image: Wir werden vom Ausland weltweit zweifelfrei als besser, geschickter, flexibler und innovativer angesehen als andere Nationen. Das bezieht sich auf den schulischen, gewerblichen und akademischen Bildungsprozess. Hier liegt ein riesiges Potential, das wir global vermarkten können [HHBH09]. Das bedeutet aber, dass Prävention und damit verbundene Prozesse nicht eine kostenlose Beigabe sein dürfen, sondern ein teures Exportprodukt. Weil wir in Fragen von Präventionsmaßnahmen und strategien „Best of Class“ im weltweiten Kontext sind. Wir müssen lernen uns diese bezahlen zu lassen sowie andere Dienstleistungen. Das gilt auch im übrigen für den Export universitärer Bildungsleistungen. Der „Homo Sapiens“ wird durch den „Homo Zappiens“ ergänzt. Der Begriff „Homo Zappiens“ beschreibt die gerade heranwachsende Generation, die durch eine digitale Welt und ihren Auswirkungen geprägt wird. Sie haben eine höhere Parallelitätskompetenz als viele Erwachsene [VV08](vgl. Abbildung 8).
Abb. 8 The homo zappiens learns differently [VV08]
54
Klaus Henning et al.
Damit im Zusammenhang steht aber, dass Arbeit und Lernen in weltweit verteilten virtuellen Arbeits- und Lebensumgebungen Normalität wird: „Work is what you do, not where you go“ (Vgl. Greiner 2009, Koper 2009). In enormer Geschwindigkeit entwickeln wir uns in einem doppelten Mobilitätsbegriff: Entweder bin ich „mobil“ in den Tätigkeitsarten und kann dann meinen Wohnsitz relativ konstant halten oder ich bin im weltweiten Kontext ortsflexibel und erhalte meine Spezialität. Ich „wandere“ dann dahin, wo es die Arbeit gibt [HH09]. Dabei darf nicht außer Acht gelassen werden, dass ein solches Projektnomadentum unter Gesichtspunkten einer präventiven Worklife-Balance nicht unproblematisch ist. Auf paradoxe Art und Weise gewinnt die fachliche und emotionale Kompetenz in relativ kleinräumigen regionalen Milieus weltweit an Bedeutung. Plötzlich wollen die Ingenieure – um ein bereits erwähntes Beispiel aufzugreifen, wieder in das regionale Milieu der Eifel zurück. Region, auch kleine Regionen, können so zu einem „Unique Selling Point“ im weltweiten Kontext in einer besonderen Qualität werden.
5 Was folgen hieraus für Konsequenzen für zukünftige Präventionsstrategien? In dem folgenden Abschnitt sollen nun aus den spezifischen Rahmenbedingungen der Entwicklung in Deutschland Konsequenzen für zukünftige Präventionsstrategien gezogen werden. Dabei erscheinen auf der Basis des Aachener Appells folgende Aspekte von zentraler Bedeutung: • Präventionsdienstleistungen aus Deutschland haben das Potential eines weltweiten Exportschlager. • Prävention wird unter turbulenten Marktbedingungen vor allem durch die intelligente Verknüpfung von Arbeit und Leben gelingen. • Präventive Ansätze in der Arbeits- und Unternehmenskultur sichern längerfristig die Wettbewerbsfähigkeit eines Unternehmens. • Präventionsmaßnahmen müssen sich auf Grund der demografischen Situation in besonderer Weise auf Menschen über 50 und auf die Migranten beziehen. • In den Präventionsstrategien bedarf es der Innovation in Bezug auf die veränderten gesellschaftlichen Wertesysteme. Im Einzelnen bedeutet dies: Export von Präventionsdienstleistungen Deutschland wird mehr und mehr zu einer Dienstleistungsgesellschaft, die ihren Kern in der Produktion hat. So können Produkte mit relevanten Dienstleistungen exportiert werden. Prävention als wissensintensive Dienstleistung kann in Zukunft noch mehr mit den Produktions- und Dienstleistungsprozessen der Unternehmen gekoppelt werden. Ein neues Geschäftsfeld für Akteure mit Präventions-Know-How
Prävention und Innovation
55
entsteht, das in der jeweiligen Adaption an die kulturellen Gegebenheiten an die Länder angepasst werden muss, in die exportiert wird. Prävention durch Verknüpfung von Arbeit und Leben Arbeit und Leben stehen zunehmend unter dem Einfluss des „Homo Zappiens“. Dazu gehört auch die Wiederentdeckung des spielerischen Zugangs zu Lern- und Arbeitsprozessen. Integration von Arbeiten, Leben, Lernen und Spielen ist dabei der Schlüssel für die fachliche und überfachliche Kompetenzentwicklung der jetzt heranwachsenden Generation. Dies sollte dramatische Auswirkungen auf den pädagogischen Prozess haben, wie dies ja in vielen Ansätzen der Reformpädagogik bereits angeregt, aber nur selten konsequent umgesetzt worden ist. So verwehrt sich z. B. in der Montessori-Pädagogik die Begründerin Maria Montessori gegen den Begriff „Kinder spielen“ und sagt „Kinder arbeiten“. In diesem Kontext müssen also pädagogische Konzepte nicht neu erfunden werden, sondern endlich in ihrer Relevanz für Bildung und Ausbildung vom Kindergarten bis zur Universität erkannt werden. Eine der Selbstverständlichkeiten in diesen Konzepten ist die erlebte Verknüpfung von Theorie und Praxis, wie sie sich z. B. in turbulenztauglichen Change Management-Ausbildungskonzepten niederschlägt [ost]. In diesem Zusammenhang ist insbesondere das duale Ausbildungssystem für die gewerblich-, technische- und handwerkliche Ausbildung einzuordnen. Dieses Konzept stellt einen weiteren Kandidaten für einen Export-Bestseller dar, wenn eine entsprechende weltweite Vermarktung erfolgen würde. Das heißt für alle Präventionssysteme, dass sie entsprechende Ansätze enthalten müssen. Insbesondere ist dadurch ein erfolgreicher Vermittlungsweg vorgezeichnet. Prävention in der Arbeits- und Unternehmenskultur Arbeits- und Unternehmenskultur ist durch einen Zuwachs der kulturellen Vielfalt gekennzeichnet. Eine solche Entwicklung stellt eine große kulturelle Bereicherung dar, birgt aber auch die Gefahr der Fragmentierung, des Zerrisses der Gesellschaft und der Polarisierung. Wir brauchen deshalb Präventionsmaßnahmen, die mit dieser kulturellen Vielfalt umgehen und diese Nutzen können. Eine Präventionsmaßnahme könnte sich z. B. darauf beziehen einen Produktionsbetrieb zu gestalten, in dem man neben nicht religiös gebundenen Mitarbeitern, christlich gebundene Mitarbeiter und islamisch gebundene Mitarbeiter hat. Wie gestaltet man nun den betrieblichen Alltag unter Berücksichtigung der muslimischen Gebetszeiten in einem Mehrschichtbetrieb? Das Beispiel soll deutlich machen, welche vielfältige Dimension eine positive Ausgestaltung der kulturellen Vielfalt in der Praxis des Unternehmens bedeutet. Moderne Arbeits- und Unternehmerkultur enthält aber als Folge der Projektstrukturen und des globalen Aktionsradius die Frage nach Präventionsstrategien in Hinblick auf den bereits ausgeführten doppelten Mobilitätsbegriff. Ein besonderes Augenmerk muss hier auf die Gilde der „Wanderarbeiter“, oder wie sie auch genannt werden „Projektnomaden“ [RSF06] gelegt werden. Wir leben in einer Zeit in
56
Klaus Henning et al.
der in effizienten Unternehmen Kernbelegschaften etwa die gleiche Größenordnung haben wie variable Projektmitarbeiter, die über Netzwerke realisiert werden, welche an verschiedenen Stellen tätig sind. Im Kontext dieses doppelten Mobilitätsbegriffes werden betriebliche Präventionsmaßnahmen zu einer überlebensfrage für Unternehmen [HH09] Beispiel: Das weithin bekannte „Burn-Out-Syndrom“ bei Lehrern, welches statistisch eine enorme Zuwachsrate hat, wird zunehmend in der Gruppe der Projektnomaden und vor allem der intellektuellen Wanderarbeiter ein Thema der frühen Lebensmitte, bei dem psychosomatische und psychiatrische Kliniken massive Zuwachsraten prognostizieren [Sch04]. Wie sehen die dazu gehörenden betrieblichen Präventionsmaßnahmen aus? Es kann ja nicht schlimmeres passieren, als dass ein Unternehmen KernkompetenzMitarbeiter durch Burn-out Prozesse verliert. Hier gilt es in den Präventionsstrategien die herkömmliche Methode des Supervision zu überwinden, weil sie in der Regel viel zu spät einsetzt. Sondern in der gesamten betrieblichen Weiterbildung der Management-Qualifizierung bei wachsender Dynaxity einen hohen Stellenwert zu geben. Prävention für die Alten und die Immigranten Wie erläutert führt die demografische Wende in Deutschland nicht nur zu einem enormen Fachkräftemangel. Gleichzeitig hat sich die durchschnittliche Lebenserwartung in den letzten 5 Dekaden um 10 Jahre erhöht. Diese Tendenz bleibt in den nächsten Jahren gleich. Dies muss zwangsläufig zu einer Erhöhung des Renteneintrittsalters führen – sowohl in Hinblick auf die länger vorhandene Arbeitsfähigkeit als auch in Hinblick auf die Finanzierbarkeit der Rentensysteme. Unabhängig davon werden viele Rentner aus zwei Gründen wieder arbeiten: Zum einen, weil ihre Fachkompetenz oft heute nicht mehr ausgebildet wird, wie z. B. Facharbeiter oder Ingenieure für Kernkraftwerke. Diese Gruppen werden reaktiviert, nicht etwa weil sie Geld bräuchten, sondern weil die Nachwuchskräfte fehlen. Im Fall der Kernkraftwerke ist es zusätzlich eine Frage des sicheren Betriebs geworden fehlende Fachkräfte durch entsprechend qualifizierte Rentner zu ersetzen. Zum anderen werden diejenigen Rentner wieder erwerbstätig, die auf Grund niedrig werdender Renten dazu gezwungen sind. Für beide Felder ist Prävention eine Schlüsselkompetenz erfolgreicher Integration und Reintegration der Alten. Als Beispiel sei ein berufstätiger Rentner aus dem Supermarkt genannt: Er ist 75 Jahre alt und arbeitet als freier, selbstständiger Mitarbeiter seit 15 Jahren in dem Unternehmen. Seine Frau ist vor zehn Jahren gestorben, er arbeitet jeden Tag von 11 bis 18 Uhr, wohnt in 150 km entfernt und fährt von dort jeden Tag 2 ½ Stunden zum Arbeitsort und zurück. Seine Arbeitszeit ist auf seine persönlichen Rahmenbedingungen angepasst. Er sagt, dass die Arbeit sein Lebensinhalt sei, ihm Spaß mache und er nicht ans Aufhören denke. Es geht also beim längeren Arbeiten im Alter nicht darum Lebensarbeitszeiten mit klassischen Arbeitsverhältnissen zu kopieren, sondern im Sinne von Work-Life-
Prävention und Innovation
57
Balance alters- und situationsangemessene Modelle zu entwickeln und auf ihr Zukunftspotenzial hin zu untersuchen. In ähnlicher Weise gilt es Präventionsstrategien für Migrationsströme nach Deutschland zu entwickeln. Dabei muss unterschieden werden zwischen häufig gering qualifizierten Migranten, die nach Deutschland „hineindrängen“ und den gut qualifizierten Migranten aller Berufsgruppen, die selten den Weg nach Deutschland finden, weil wir keine Migrationsstrategie für hoch qualifizierte Arbeitskräfte haben. Für diese Berufsgruppe werden präventive Maßnahmen zur schnellen und erfolgreichen Migration der zentrale Erfolgsfaktor, wobei diese Präventionsmaßmahmen systemisch neben den betrieblichen Aspekten den gesamten familiären Kontext von Wohnung, Kinderbetreuung, sprachliche Ausbildung, Schule, Nachbarschaftsintegration etc. enthalten muss. Innovation in der Prävention Als letztes sei auf eine generelle Werteverschiebung hingewiesen, die eine Innovation in der Prävention auslösen sollte. Laut Opaschowski gelten für die jüngere Generation die Zielwerte Vertrauen, Veränderung und Verlässlichkeit. Wohlstand und Freiheit sind keine dominanten Werte mehr, nicht, weil sie nicht wichtig sind, sondern weil sie als selbstverständlich vorausgesetzt werden. Vor diesem Hintergrund geht es um das Einverständnis zwischen den Generationen und um den Wunsch nach bleibenden Werten, wie Familie, Freundschaft, Partnerschaft. Für Präventionsstrategien bedeutet das, dass Vertrauen, Veränderung und Verlässlichkeit immer zentrale Elemente sein müssen. Erst dann wird eine Präventionsmaßnahme in Hinblick auf die nächste Generation nachhaltig. Präventionsmaßnahmen haben als Kernkompetenz die Verstetigung von Präventionselementen in einer Organisation oder Umgebung. Es nützt nichts eine Präventionsmaßnahme zu machen, die nach zwei Jahren wieder aufhört. Prävention verdient diesen Namen also nur, wenn sie den Charakter von Verlässlichkeit, Vertrauenswürdigkeit hat und damit auch nachhaltig Veränderungen schafft.
6 Was ist zu tun? Zunächst seien die wichtigsten Handlungsfelder zukünftiger Prävention zusammengefasst: • Es gibt noch keine ausreichenden „Arbeiten im Alter-Modelle“, • Es gibt kein ausreichendes Diversitymanagement als Migrationsstrategie mit denen sich Ausländer in kürzerer Zeit inklusive Familien, Kindern und Arbeitsplatz in Deutschland wohl und aufgenommen fühlen. • Es fehlt an Erfahrung in Partizipation und Beteiligungsqualifizierungen im weltweit verteilten multinationalen Team.
58
Klaus Henning et al.
• Es fehlt ein pädagogisches Konzept für die Arbeits- und Prozessgestaltung für den „Homo Zappiens.“ • Es fehlen Geschäftsmodelle für den Export von Bildung- und Präventionsdienstleistungen. Zum Abschluss seien ein paar Beispiele angefügt, die zu denken anregen sollen: • Prävention für Lkw-Fahrer: Ist es nicht dringend nötig, dass Lkw automatisiert auf Autobahnen fahren können? Ist der Arbeitsplatz eines Lkw-Fahrers nicht ein menschenunwürdig, wenn eine 55-Stunden-Woche als fortschrittlich gilt. Sind hier einschlägige Kriterien von Prävention und Arbeitsschutz, wie wir ihn innerbetrieblich gewohnt sind, umgesetzt? • Prävention für Omnibusfahrer: Ist es nicht dringend nötig, die Fahrerassistenzsysteme für mit 30 bis 40 Personen beladene Busse zwangsweise mit allen verfügbaren Technologien zu versehen? So könnten zum Beispiel automatische Notbremsen ausgelöst, Spurverlassens-Systeme mit Vollbremssystemen verknüpft werden und zur Unfallvermeidung von fast 50% der tödlichen Unfälle beitragen. • Ist es nicht ein Präventionsthema, dass unsere Kinder in ihren Kleidungen eingewebte Computersysteme haben sollten? So könnten Kinder auf Schulwegen virtuell begleitet werden, um sie gegen alle Formen von Missbrauch zu schützen. • Ist es nicht notwendig, dass wir psychosomatische Kliniken im Rahmen von Präventionsstrategien vermehrt strukturell als betriebliche Partner integrieren? Damit wäre der übergang von der entsprechenden medizinischen Behandlung in den Betrieb ein weniger schwerer Schritt. Und ein letztes Beispiel: Die Firma Vita Needle hat auf besondere Art Arbeitsfelder für Alte geschaffen [sta]. Ein Unternehmen in der Nähe von Boston existiert in der dritten Generation. Es wurde in der Wirtschaftkrise der 1920er gegründet, das reguläre Mindesteintrittsalter ist 75 Jahre. Es werden im Unternehmen bei ca. 5 Millionen Jahresumsatz Spezialnadeln für den medizinischen Einsatz gefertigt. Das Unternehmen ist inhabergeführt. Man darf dort solange arbeiten, solange die zweite Etage noch selbstständig zu Fuß erreichen kann. Viele, die dort arbeiten, haben sehr unterschiedliche Arbeitsverhältnisse und arbeiten zum Teil bis wenige Wochen vor ihrem Tod. Ein mögliches Modell für Prävention im Alter? Wir wollen damit deutlich machen, dass unser Eindruck ist, dass es nicht an Best Practice Beispielen für den präventiven Umweltschutz im weltweiten Kontext fehlt, sondern an Strategien der Nachhaltigkeit, enabled by Germany.
Literaturverzeichnis [aix] [Alb92] [bay] [BBLW02]
Aixcore Group. http://www.aixcore.de [Zugegriffen 2010-06-09]. M. Albert. Kapitalismus contra Kapitalismus. Frankfurt/Main, New York, 1992. Bayer - Konzern-Homepage. http://www.bayer.de/de/homepage.aspx [Zugegriffen Juni 9, 2010]. Barbara Berkel, Axel Börsch-Supan, Alexander Ludwig, and Joachim Winter. Sind die Probleme der Bevölkerungsalterung durch eine höhere Geburtenrate lösbar?
Prävention und Innovation
[BK08]
[bmb] [Cam04] [Hen]
[Her03] [HH09]
[HHBH09]
[HI94]
[HM01] [IAB09] [IB05]
[ifm] [iP10] [Kae01]
[kle09] [man] [Nor] [ost] [p3?] [R¨] [RSF06]
59
Mannheim Researchinstitute for the Economics of Aging (MEA), Universität Mannheim, Mannheim, September 2002. Hanno Ballhausen and Ute Kleinlümern. Die wichtigsten Erfindungen der Menschheit geniale Ideen, die die Welt veränderten. Chronik-Verl., Gütersloh, München, 2008. BMBF’s International Dialogue on Sustainability Research. E. Cameron. Multiculturalism and Immigration in Canada. An introductory reader. Michigan, 2004. Klaus Henning. Präventionsforschung und unternehmerische Praxis: Zwei Seiten einer Medaille. In Klaus Henning, Ingo Leisten, and Frank Hees, editors, Innovationsfähigkeit stärken - Wettbewerbsfähigkeit erhalten. Präventiver Arbeits- und Gesundheitsschutz als Treiber, Aachener Reihe Mensch und Technik, pages 12–30. Roman Herzog. Soziale Sicherheit zur Reform der sozialen Sicherungssysteme. BMBF. Berlin, 2003. Klaus Henning and Andrea Huson. Innovation Champions - Or how to achieve (global) Competitiveness! In Christopher Schlick, editor, Industrial engineering and ergonomics visions, concepts, methods and tools. Festschrift in honour of Professor Holger Luczak, pages 31–66. Springer, Berlin, Heidelberg, 2009. Klaus Henning, Frank Hees, Ursula Bach, and Alan Hansen. Yes, we can! Warum Deutschland den Kopf nicht in den Sand stecken sollte. In Dieter Spath, editor, Arbeits- und Dienstlesitungsforschung als Innovationstreiber. Bilanzen, Herausforderungen, Zukünfte., pages 51–59. Stuttgart, 2009. Klaus Henning and Ingrid Isenhardt. Kybernetische Organisationsentwicklung - Gestaltungsprinzipien für komplexe, soziotechnische Systeme. In Bernd Schiemenz, editor, Interaktion : Modellierung, Kommunikation und Lenkung in komplexen Organisationen, pages 103–108. Duncker & Humblot, Berlin, 1994. Klaus Henning and Simon Marks. Kommunikations- und Organisationsentwicklung. Aachen, 6 edition, 2001. IAB. IAB 2009: Offene Stellen im IV. Quartal 2008: Einbruch in der Industrie Soziale Berufe legen zu. IAB-Kurzbericht 11/2009. Technical report, 2009. Ingrid Isenhardt and Dietrich Brandt. Der Mensch in der Kommunikation mit der Technik. In Ingrid Isenhardt, editor, Der Mensch in der Kommunikation mit der Technik, pages 1–45. Wiss.-Verl. Mainz, Aachen, 1 edition, 2005. IfM Bonn, Statistisches Bundesamt, 2004. Technical report. DLR im PT. 2010. Jahresbericht 2008/2009 Forschung und Entwicklung Innovative Arbeitsgestaltung und Dienstleistungen. Technical report, Bonn, 2010. Hartmut Kaelble. Europäer über Europa : die Entstehung des europäischen Selbstverständnisses im 19. und 20. Jahrhundert. Campus, Frankfurt/Main, New York, 2001. Kleiner, jünger, wachstumsstärker. Studie zu Familienunternehmen. Manager Magazin, December 2009. Manifest für agile Softwareentwicklung. http://scrum-fibel.de/agiles-manifest.html [zugegriffen 2010-06-07]. Klaus North. Wissensorientierte Unternehmensführung. Wertschöpfung durch Wissen. Wiesbaden. OSTO Systemberatung GmbH -Home. http://osto.de/index.php [zugegriffen 201006-07]. P3 Group: Home. http://www.p3-group.com/ingenieurgesellschaft/de/home.html [zugegriffen 2010-06-07]. Bert Rürup. Nachhaltigkeit in der Finanzierung der sozialen Sicherungssysteme. Bericht der Kommission. BMAS. Berlin. Jutta Rump, Thomas Sattelberger, and Heinz Fischer, editors. Employability Management. Grundlagen, Konzepte, Perspektiven. Betriebswirtschaftlicher Verlag Dr. Th. Gabler/GWV Fachverlage GmbH Wiesbaden (GWV), New York, 2006.
60 [Sch04] [Sim07] [sta] [Vol07]
[VV08]
Klaus Henning et al. U. Schaarschmidt. Halbtagsjobber? : psychische Gesundheit im Lehrerberuf - Analyse eines veränderungsbedürftigen Zustandes. Weinheim, 2004. Hermann Simon. Hidden Champions des 21. Jahrhunderts die Erfolgsstrategien unbekannter Weltmarktführer. Campus-Verlag, Frankfurt/Main, New York, 2007. Stainless Steel Tubing - Medical Needles - Luers & Fabrication - Vita Needle Co. http://www.vitaneedle.com [zugegriffen 2010-06-07]. Volker Volkholz. Capability for Innovation. In Joachim Ludwig, editor, Arbeitsforschung und Innovationsfähigkeit in Deutschland, pages 41–49. Hampp, München, 1 edition, 2007. Wim Veen and Ben Vraking. Homo Zappiens and its Consequences for Learning, Working and Social Life. Aachen, 2008.
Yes, we can! Warum Deutschland den Kopf nicht in den Sand stecken sollte Klaus Henning, Frank Hees, Ursula Bach, Alan Hansen
Zusammenfassung Die aktuelle Stimmung in der Bundesrepublik und den europäischen Staaten ist durch die Finanz- und Wirtschaftskrise mehr als getrübt. Dabei gibt es gar keinen Grund, den Kopf in den Sand zu stecken, denn Deutschland hat in den vergangenen Jahren seine Hausaufgaben gemacht, um sich erfolgreich auf die Anforderungen der modernen, digitalen und flexiblen Welt einzustellen. Die Trends sind gelegt und die erfolgversprechenden Strategien müssen weitergetrieben werden. So können Deutschland und Europa als Gewinner aus den Krisen hervorgehen. Schlüsselwörter Demografischer Wandel · Trends · Wirtschaftskrisen · Enabled by Germany · Homo Zappiens
Wie wird es uns gelingen, die Chancen, die aus dem demografischen Wandel erwachsen, zu erkennen und für uns nutzbar zu machen? Wird es uns gelingen die neuen Technologien, wie z. B. moderne Energietechnik, Nanotechnik, Robotik und Automation, in einem Dienstleistungspaket in aller Welt zu vermarkten und die Anwender zu befähigen, diese neue Technologien auch zu betreiben? Wird es uns gelingen das Internet und all seine Chancen, die es bietet, gut und verantwortungsvoll zu nutzen? Was bedeutet all dies für den Standort Deutschland und seine europäischen Nachbarn hinsichtlich einer Entwicklungsperspektive für die nächsten 20 Jahre?
2 Drei Trends und ihre Chancen für Deutschland und Europa 2.1 Mangelware „Junger Mensch“ Wenn eine positive Tendenz bei der Zunahme von Geburten einsetzt, wird sich der Knick in der Geburtenrate erst in ca. 20 Jahren am Arbeitsmarkt auswirken können [BBLW02]. Bis dahin müssen wir mit den gegebenen demografischen Gesellschaftsbedingungen leben: Immer weniger Menschen im jungen und mittleren Alter stehen einer immer größer werdenden Anzahl von Menschen im Rentenalter gegenüber [BBLW02], die eine immer längere Lebenserwartung genießen können [Bun06]. Daraus ergeben hauptsächlich sich zwei Konsequenzen: ¨ 1. Ältere Menschen werden tendenziell länger arbeiten können und müssen [R03], [Her03]. Hier reicht allein das Konzept „Rente mit 67“ nicht aus, sondern es bedarf neuer Ansätze für eine alterns- und altersgerechte Arbeitspolitik [Kis06], die die Komponenten Altern, Gesundheit und Kompetenzentwicklung berücksichtigt [Hen]. 2. Deutschland muss ein Land mit einer hervorragenden Immigrationskultur werden. Vorbild dafür sind klassische Einwanderungsländer wie Kanada und Australien [Cam04]. Um den Standort Deutschland für ausländische Hochqualifizierte interessant zu machen, muss dem guten ersten Schritt ein „Land der Ideen“ [LdIMfD08] zu etablieren, der zweite Schritt zu einem „Land der Wertschöpfung“ [Cla06] folgen. Außerdem muss es denjenigen, die sich entschlossen haben nach Deutschland zu kommen, einfacher gemacht werden, ein soziales Netzwerk aufzubauen [BdVdZ09]. Einen Ansatz in die richtige Richtung stellt die RWTH Aachen mit ihren „Starter Kits für neuberufene Professoren“ vor. Hiermit soll es den neuen Angestellten der RWTH Aachen schneller gelingen, ihre Rolle in der Hochschule wahrzunehmen und ein soziales Umfeld aufzubauen. Diese Maßnahmen werden durch finanzielle Mittel der Exzellenzinitiative gefördert.
Yes, we can!
63
2.2 „Enabled by Germany“ durch Deutschlands Hidden Champions Deutschland wird den Titel „Exportweltmeister“ nicht verteidigen können. Deutschland hat dadurch die Chance in zunehmendem Maß ein „Enabler“ zu werden, der anderen Ländern hilft, ihre eigenen Produktions- und Dienstleistungsprozesse zu gestalten. Durch dieses Umdenken kann Deutschland sich erstens neue Märkte und Abnehmer erschließen und zweitens einer neuen Art von Entwicklungshilfe Hilfestellung leisten. Denn das Beste für die Entwicklungsländer ist es, wenn sie ihren Eigenbedarf an Produkten und Dienstleistungen eigenständig herstellen und produzieren können. Für die Entwicklungsländer bestünde so die Möglichkeit die eigene Versorgung unabhängig von den Industriestaaten sicherzustellen. Deutschland sollte demnach diesen kombinierten Produktions- und Dienstleistungssektor im Sinne seiner zukünftigen wirtschaftlichen Entwicklung für sich entdecken [BD06] und sich als globaler Dienstleister engagieren. Zukünftig werden sich die Sektoren Produktion und Dienstleistungen immer schwerer voneinander trennen lassen [Bry09]. Zwei Bereiche können im Fokus dieser Anstrengungen stehen: 1. Vermarktung von Bildung, Aus- und Fortbildung „Made in Germany“, 2. Internationaler Partner für kombinierte Produktions- und Dienstleistungsprozesse. Hier könnte der Begriff „Made in Germany“ ersetzt werden durch den Slogan „Enabled by Germany“. Die Voraussetzung für ein erfolgreiches Auftreten auf dem Markt der Dienstleistungen ist, dass deutsche Unternehmer mit ihren Mitarbeitern mobil sind und auf den Kunden im Ausland zugehen [BM05]. Das duale Ausbildungssystem, gewerblich-technische und Handwerksausbildungen sind prädestiniert, um weltweite Bestseller zu werden. Ein Beispiel für die Nachfrage nach deutschen Dienstleistungen ist die Gründung von 1000 Schulen in Indien mit Hilfe des Bundesministeriums. Der zweite Fokus „Enabled by Germany“ bezieht sich im besonderen Maße auf die Bereiche Forschung, Entwicklung und Design. In vielen Nischen der Wirtschaft haben sich deutsche Unternehmen einen Platz als Weltmarktführer erarbeitet. Denn eigentümergeführte Unternehmen, die in Deutschland 75 % aller Arbeitsplätze stellen [fMB06], agieren heute schon in großem Maß erfolgreich in globalen Strukturen [Hun03] und treiben Innovationsprozesse voran [Hen]. Von dieser Art Unternehmen gibt es eine große Anzahl. Allein 75 deutsche Zulieferer produzieren technologisch innovative Teilkomponenten für das Luft- und Raumfahrtunternehmen Boeing [Boe] – Deutschland ist voll von weltweit besten TechnologieKomponenten [Sim07]. Was heißt dies in letzter Konsequenz? Wir müssen unser Privileg, zu einem der reichsten Länder der Erde zu gehören, durch Fleiß und Mehrarbeit an den Stellen weiterentwickeln, an denen die anderen Länder (noch) nicht so weit sind, dass sie es selbst machen könnten. Nur da, wo wir wirklich besser, geschickter, flexibler und innovativer als andere sind, sollten wir die zugehörigen Arbeitsplätze in Deutschland behalten.
64
Klaus Henning et al.
2.3 Der global-regionale Homo Zappiens Die fortschreitende Digitalisierung und die kontinuierliche Verbreitung von Internetanschlüssen [Sch07] gebiert eine globale Intelligenz. Wer heute geboren wird, wird als Homo Zappiens geboren. Für Mitglieder der digitalisierten Generation ist es bereits nach wenigen Lebensjahren selbstverständlich, sich im Internet zu bewegen, dort einen großen Teil seiner Lebenszeit zu verbringen und Freundschaften in Form virtueller sozialer Netzwerke zu pflegen. So entwickelt sich eine neue Art von Lebensqualität [VV08]. Seit dem das Internet untrennbar mit dem alltäglichen Leben verbunden ist, wurden von den Nutzern neue Strategien entwickelt, wie sie mit der Vielfalt an Informationen und Wissen effizient und ergiebig umgehen können [Hen]. Die Welt rückt durch den technischen Fortschritt zusammen. Tägliches Kommunizieren mit Freunden über weite Distanzen hinweg, der Austausch von Ideen und die gemeinsame Arbeit an einem Projekt über die Grenzen von Kontingenten hinweg, wie z. B. der Entwurf und die Weiterentwicklung von Open-Source Software, [HK09] stellen heutzutage kein Problem mehr dar Blogs und Wikis gehören zum Standard der alltäglichen Kommunikation. Im Rahmen dieses Digitalisierungsprozesses wird das Bedürfnis nach räumlicher Geborgenheit in regionalen Milieus paradoxerweise extrem zunehmen [Mos96]. Die einzelnen Regionen wie beispielsweise Oberschwaben oder das Inntaldreieck, die Euregio Aachen oder Zwickau, das Vogtland oder die Lausitz werden für die Identität des Menschen wieder an Bedeutung gewinnen – angesichts der „Verlorenheit“ im globalen Raum der zappenden Internet-Welt.
3 Wie sind wir vorbereitet? Um Chancen und Vorteile der Trends in der Zukunft adäquat nutzen zu können, wurde in der Vergangenheit eine umsichtige und vorausschauende Arbeitsund Lernforschung durch das BMBF und den Projektträger im DLR (PT im DLR) gefordert und gefördert. Dies sei exemplarisch an einigen Projekten gezeigt, die am ZLW/IMA der RWTH Aachen bewilligt und erfolgreich im Rahmen der Forschungs- und Förderprogramme „Arbeit und Technik“ (1989–2000) „Lernkultur Kompetenzentwicklung“ (2001–2006), „Innovative Arbeitsgestaltung“ (2001– 2006) und „Arbeiten, Lernen, Kompetenzen entwickeln. Innovationsfähigkeit in einer modernen Arbeitswelt.“ (2007–2011) durchgeführt wurden. Im Mittelpunkt standen Innovationen der Arbeitsgestaltung im ganzheitlichen Sinne: Die Faktoren Mensch, Organisation und Technik müssen gleichermaßen beachtet und durchdacht sein [Har05]. Um den oben genannten Trends begegnen zu können, wurden verschiedene Projekte unter den Schlagwörtern, Beteiligungsqualifizierung [Bit91], Partizipation & Empowerment, Gestaltung von Dienstleistungsprozessen und die aktive Mitgestaltung technischer Innovationsprozesse durchgeführt. An drei Beispielen soll dies verdeutlicht werden.
Yes, we can!
65
3.1 Mitarbeiterqualifizierung durch Partizipation und Empowerment Als „vordringliche Maßnahme im Rahmenkonzept ‚Innovative Arbeitsgestaltung‘ des BMBF“ wurde das Projekt Partizipation und Empowerment (P & E) beschrieben. Das Projekt hatte zur Aufgabe das „beobachtete Phänomen eines integrierten Konzeptes von Partizipation & Empowerment“ [IHH03] auf individueller und organisationeller Ebene zu betrachten. In aller Kürze umschreibt, Partizipation und Empowerment „Maßnahmen aus der Personal- und Organisationsentwicklung, die selbständiges und eigenverantwortliches Handeln aller Mitarbeiter fördern“ [San05]. Ziel des Projektes war die Beobachtung von P & E-Vorgängen, um anschließend mögliche Gestaltungsmöglichkeiten zu identifizieren. Um dies zu erreichen war eine enge und kooperative Zusammenarbeit mit der unternehmerischen Praxis notwendig. So konnten realitätsnah Ableitungen und Einflussgrößen beschrieben werden. Mit diesem Ansatz im Rahmen des Projektes wurden wissenschaftliche und genuin praxisbezogene Ergebnisse generiert [IHH03]. Im Rahmen dieses Vorhabens wurden durch Integration bestehender Ansätze und den Abgleich mit aktuellen Gegebenheiten in deutschen Unternehmen die Forschungsfelder von P & E miteinander verknüpft und im Sinne praxisnaher Handlungsempfehlungen und Szenarien erschlossen. Bestimmende Faktoren waren hierbei tatsächliche Bedarfe von Mitarbeitern, Führungskräften und Unternehmen sowie die Entwicklung der Märkte für Güter, Dienstleistungen und Arbeitskraft. Darüber hinaus gilt es, geeignete innerbetriebliche Organisations- und Personalentwicklungsprozesse für die sich verändernden Rahmenbedingungen wirtschaftlichen Handelns aufzuzeigen (Henning 2000).Von besonderem Interesse sind hierbei die Art und der Einfluss des Führungsverhaltens. Darüber hinaus war ein übergeordnetes Ziel des Projektes, gemeinsam mit Unternehmen und im Austausch mit (internationalen) Experten Szenarien, zukünftige Unternehmensentwicklungen im Kontext von P&E zu beschreiben und entsprechende Handlungsempfehlungen zu entwerfen.
3.2 Servicezentrum: Dienstleistungsprozesse kundenorientiert gestalten Innovative Märkte werden von Dienstleistern erschlossen, deshalb hat das BMBF mit dem PT im DLR 1995 die Initiative „Dienstleistungen für das 21. Jahrhundert“ ins Leben gerufen [HP03]. In diesem Rahmen hatte das Projekt „Prozessorientierte Gestaltung und Absicherung von Dienstleistungsprozessen am Beispiel Servicezentrum“ sich zum Ziel gesetzt Unternehmen zu befähigen, ihre unternehmensspezifische Dienstleistungsorientierung neu zu gestalten. Ziel des Vorhabens war die Entwicklung von „Servicezentren“ eine nachhaltige Absicherung der Dienstleistungsqualität gewährleistet.
66
Klaus Henning et al.
Abb. 1 Projektziel Partizipation und Empowerment
Denn nicht nur klassische Dienstleistungsunternehmen, sondern auch zunehmend industrielle Anbieter, die ihre Produkte unter den verschärften Bedingungen heutiger Märkte anbieten, müssen ihre Produkte mit Dienstleistungen koppeln, um den gestiegenen Kundenanforderungen und -wünschen gerecht zu werden. Dienstleistungsqualität wird hier zur entscheidenden Kernkompetenz. Die Entwicklung und Umsetzung entsprechender Qualitätsmanagementinstrumentarien sowie einer lernenden Dienstleistungskultur bilden dabei eine wesentliche Grundlage für die erfolgreiche Gestaltung von unternehmensinternen und -externen Dienstleistungsbeziehungen. Ingesamt waren die Entwicklung und die Umsetzung einer lernenden Dienstleistungskultur eine wesentliche Grundlage für die erfolgreiche Gestaltung von Dienstleistungsprozessen, z. B. durch IT-gestützte Kundenschnittstellen und dienstleistungsorientiertes Qualitätsmanagement [HP03].
3.3 SInn: Smarte Innovation Der Anlagen- und Maschinenbau ist eine der innovativsten Branchen Deutschlands. Damit dies auch so bleibt, muss an der Innovationsfähigkeit der Branche gearbeitet werden. Innovationen und Innovationsfähigkeit bestehen nicht nur aus der technischen Komponente, sondern müssen viele Einflussgrößen betrachten, wie z. B. wirtschaftliche und gesellschaftliche Entwicklungen, Trends auf dem Weltmarkt oder Veränderungen auf organisationeller oder individueller Ebene. Die Fragen „Wie kann Personal- und Kompetenzentwicklung Innovationsprozesse unterstützen?“ oder „Wie können Mitarbeiter an der Entwicklung von Innovationen beteiligt werden?“ bedürfen der Beantwortung. Innovation muss immer wieder neu erfunden werden und wird zukünftig noch mehr als bisher zu einer permanenten Herausforderung.
Yes, we can!
67
Abb. 2 Projektstruktur und Aufbau des Servicezentrums
3.4 Daraus resultierende Forschungsfragen für die Zukunft Die Forschungsergebnisse, die hier nur exemplarisch für viele Projekte aus den Programmen und Initiativen des BMBF stehen, müssen nun an die Bedingungen einer globalisierten, sich stetig wandelnden (Arbeits-)Welt angepasst werden. Sie müssen u. a. auf die Anforderungen der Digitalisierung, der Globalisierung und des demografischen Wandels gespiegelt und weiterentwickelt werden: • Wie kann ein Arbeiten-im-Alter-Modell aussehen? • Bei notwendiger und gewünschter Steigerung der Immigranten wie wollen wir Diversitymanagement verstehen und anwenden? • Wie gestaltet sich Mitarbeiterqualifizierung in Teams, die über die Welt verstreut sind und aus den unterschiedlichsten Kulturen und sozialen Bedingungen kommen? • Wie sieht Partizipation und Beteiligungsqualifizierung aus, wenn das Internet als hauptsächliches Kommunikationsmedium verwendet wird? Wie werden in diesem Zusammenhang Blogs, Foren und Wiki die Unternehmenskulturen nachhaltig verändern können? • Wie sind die Kompetenzen des Homo Zappiens zu bewerten? Wie können diese weiterentwickelt werden? Welche sozialen Spannungen werden dadurch neu entstehen und wird diese Entwicklung die Arbeits- und Verhaltensstrukturen in den Betrieben beeinflussen?
68
Klaus Henning et al.
Abb. 3 Projektstruktur Sinn – Smarte Innovationen
4 „Ja, aber. . . “ Unsere Zeit ist also spannend und Nerven aufreibend, wohl spannender als manche Zeiten davor. Der eine oder andere Zeigefinger mag bei diesen Ausführungen jetzt hochgehen mit dem typisch deutschen „Aber. . . !“ ...Aber die Krise, Aber der Terrorismus, Aber die Umweltverschmutzung, Aber die Kriege, Aber die Ausbeutung, Aber die Politiker, Aber die Unternehmen, Aber die Börse, .... Es gibt aber eigentlich keinen Grund für ein Aber, denn wir sind gut gewappnet die Herausforderungen erfolgreich zu bestehen. Die turbulenten Märkte, die steigende Internationalisierung von Prozessen, der Finanz- und Mentalitätskrise sowie die noch nicht immer in aller Gänze abschätzbaren Folgen des demografischen Wandels können deutsche Unternehmen im Zusammenarbeit mit der Wissenschaft erfolgreich bewältigt und sollten als Chance genutzt werden [BdVdZ09]. Flexibilität und Innovationsfähigkeit wird verlangt, um im internationalen Wettbewerb auch in mittel- und langfristiger Zukunft mithalten zu können und um die hervorragende Stellung am Weltmarkt in den Bereichen Maschinenbau, Anlagenbau, Energietechnik zu erhalten. Flexibilität und Innovationsfähigkeit wird ebenso verlangt wenn Unternehmer und Wissenschaftler Ansätze für ein ganzheitliches und für viele Menschengruppen Diversitymanagement zu entwickeln. Flexibilität und Innovationsfähigkeit wird ebenso verlangt, wenn die Errungenschaften von Partizipation und Empowerment nicht aufgeben wollen. Innovationsfähigkeit und Flexibilität sind aber die entscheidenden Schlüsselqualifikationen um den erfolgreichen Weg in die Zukunft zu beginnen.
Yes, we can!
69
Literaturverzeichnis [BBLW02]
Barbara Berkel, Axel Börsch-Supan, Alexander Ludwig, and Joachim Winter. Sind die Probleme der Bevölkerungsalterung durch eine höhere Geburtenrate lösbar? Mannheim Researchinstitute for the Economics of Aging (MEA), Universität Mannheim, Mannheim, September 2002. http://www.mea.uni-mannheim.de. [BD06] E. Bartsch and B. Diekmann. Deutschlands Chancen im Handel mit Dienstleistungen. Wirtschaftsdienst 2006, 1:53–61, 2006. [BdVdZ09] Bericht des Vorsitzenden der Zukunftskommission. Zukunftskommission Nordrhein-Westfalen (2009). Innovation und Solidarität., 2009. [Bit91] Arno Bitzer. Beteiligungsqualifizierung zur Gestaltung von technischen und organisatorischen Innovationen. Doctoral thesis, ZLW/IMA, RWTH Aachen University, Aachen, Germany, 1991. VDI-Verlag Düsseldorf. [BM05] Manfred Bruhn and Heribert Meffert. Dienstleistungsmarketing. Grundlagen - Konzepte - Methoden. Mit Fallstudien. Gabler, Wiesbaden, 5 edition, 2005. [Boe] Boenig. Partnerschaft mit Deutschland. [Bry09] John Bryson. Hybrid manufactoring Systems & hybrid Products. Trendstudie im Rahmen des BMBF-Hausvorhabens Internationales Monitoring, Birmingham, 2009. [Bun06] Statistisches Bundesamt. Bevölkerung Deutschlands bis 2050, 11. koordinierte Bevölkerungsvorausberechnung, 2006. [Cam04] Elspeth Cameron. Multiculturalism and Immigration in Canada: An Introductory Reader. Canadian Scholars Press, 2004. [Cla06] U. Claassen. Wissen ist Macht. Innovationsindikator Deutschland 2006, page 26, 2006. [fMB06] Institut für Mittelstandsforschung Bonn. Jahrbuch 2006/1, 2006. [Har05] Ernst-Andreas Hartmann. Arbeitssysteme und Arbeitsprozesse, volume 39. Vdf Hochschulverlag, Zürich, 2005. [Hen] Klaus Henning. Innovation Champions. In Christopher Schlick, editor, Methods and Tools of Industrial Engineering and Ergonomics. Heidelberg edition. [Her03] Roman Herzog. “Soziale Sicherheit” zur Reform der sozialen Sicherungssysteme, 2003. [HK09] Max Haberstroh and Peter Kochalski. Innovation im Netz - Produktentwicklung am Beispiel von Crystal Space. In GWS Tagungsband 2007 Unternehmenskybernetik 2020 - betriebswirtschaftliche und technische Aspekte von Geschäftsprozessen, Aachen, 2009. [HP03] Klaus Henning and T. Pfeifer. Dienstleistungsprozesse kundenorientiert gestalten. Schlussbericht des Projektes “Prozessorientierte Gestaltung und Absicherung von Dienstleistungsprozessen am Beispiel Servicezentrum ”, 2003. [Hun03] Heike Hunecke. Produktionsfaktor Wissen - Untersuchung des Zusammenhangs zwischen Wissen und Standort von Unternehmen. Wissenschaftsverlag Mainz, Aachen, 2003. [IHH03] I. Isenhardt, F. Hees, and K. Henning. Partizipation & Empowerment, vordringliche Maßnahme im Rahmenkonzept Innovative Arbeitsgestaltung des BMBF. Aachen, 2003. [Kis06] Ernst Kistler. Die Methusalem-Lüge: Wie mit demographischen Mythen Politik gemacht wird. Hanser Wirtschaft, München, 2006. [LdIMfD08] Land der Ideen Marketing für Deutschland. 365 Ideen aus Deutschland, die in die Zukunft tragen, 2008. [Mos96] Rosabeth Moss-Kanter. Weltklasse. Im globalen Wettbewerb lokal triumphieren. Ueberreuter Verl., Wien, 1996. ¨ [R03] B. Rürup. Nachhaltigkeit in der Finanzierung der sozialen Sicherungssysteme, 2003. [San05] E.-M. Sanders. Total Quality Management in kleinen und mittelständischen Unternehmen - der Beitrag des Konzepts Partizipation & Empowerment, 2005.
70 [Sch07] [Sim07] [VV08]
Klaus Henning et al. C. Schachtner. Virtualität, Identität, Gemeinschaft. Reisende im Netz. In H. Willems, editor, Weltweite Welten, pages 103–117. Wiesbaden, 2007. Hermann Simon. Hidden Champions des 21. Jahrhunderts: Die Erfolgsstrategien unbekannter Weltmarktführer. Campus Verlag, Frankfurt, 2007. W. Veen and B. Vrakking. Homo Zappiens and his consequences for learning, working and social life. Trendstudie im Rahmen des BMBF-Hausvorhabens Internationales Monitoring. Delft, 2008.
Management and Optimal Distribution of Large Student Numbers Sabina Jeschke, Gerald Lach, Robert Luce, Olivier Pfeiffer, Erhard Zorn
Abstract Timetabling problems appear at every university. The degree of difficulty increases with an increasing number of students and courses for which the scheduling shall be carried out. From the mathematical point of view this is a “hard” problem, since the runtime on a computer cannot be estimated by a simple law (i.e. by a polynomial law) in the number of parameters. These kinds of problems are called “NP hard”. There are different aspects of the timetabling problem at universities and all specified problems are important for room management at universities, for the realization of courses that can be studied according to curricula, and for the satisfaction of students and teachers. These problems are related to the optimization of room management and personnel costs (e.g. by a uniform distribution of students). Thus, the solution of these problems is related to the optimization of “real” costs, a more and more important economic factor at (German) universities. Since 2003 for the solution of the post enrollment based course timetabling problem at the Technische Universität Berlin we are using an algorithm that has been realized by members of our team. Moreover, administration of homework and exams needs to be done. Thus, the Moses (Mobile Services for Students)-Account is being developed and used since 2004. This web-based software allows students to enroll in tutorials, with a list of preferences for given dates. A special algorithm, providing a globally optimized solution, processes all registrations. Keywords University Timetabling · Academic Administration · Integer Programming · NP-completeness
1 Introduction In principle, timetabling problems appear at every school or university. The degree of difficulty increases, however, very strongly with an increasing number of students and courses for which the time scheduling shall be carried out. From the mathematical point of view this is a “hard” problem, since the runtime on a computer cannot be estimated by a simple law (i.e. by a polynomial law) in the number of parameters. These kinds of problems are called NP hard, for details on NP hard problems see [GJ90], for the complexity of timetabling problems [CK96]. In the following we present four important aspects of the timetabling problem at universities.
1.1 Curriculum based course timetabling In each semester lectures for several courses have to be scheduled within the given rooms without conflicts for students according to the curricula for each study plan. For each teacher time conflicts have to be taken into account, i.e. one teacher can only give one course at a time. There are further constraints given by the required capacity of rooms for a course (based on estimated number of students in each course). There may be further constraints, e.g. by the requested room equipment. This is the most important university timetabling problem that has to be solved at any university.
1.2 Post enrollment based course timetabling For each course, e.g. big lectures with several hundreds of students, there may be additional small classes (tutorials). These tutorials are scheduled after enrollment of the students for the courses. Typically, each student has to be scheduled for several tutorials. Since the tutorials are small classes and each teacher (tutor) has to give several tutorials and there is only a limited number of small rooms there will be several parallel tutorials during the whole week. Thus, this timetabling problem cannot be solved together with the curriculum based course timetabling problem.
1.3 Examination timetabling In this setting the dates for the written examinations have to be determined such that students can attend all examinations as defined by their curriculum. As an additional boundary condition an adequate space of time shall lie between each examination each individual student has to attend. In contrast to the problem from section 1.1 room-equipment is generally not taken into account. In exchange generally more than one room has to be assigned for one examination in order to provide enough seats.
Management and Optimal Distribution of Large Student Numbers
73
1.4 Timetabling Student Assistants For each of TU Berlin’s approximately 600 teaching student assistants it has to be determined which lecture or course is suitable for his or her needs and abilities. This assignment was done on a paper and ink basis until winter semester 08/09: standard forms had to be handed out to all student assistants who then entered three lectures they would like to assist in the forthcoming semester. After collecting those forms (approximately 70 student assistants at the Institute of Mathematics) all the data raised was written into a huge piece of paper and the distribution of the student assistants was finally conducted on that basis. To overcome this time consuming obstacle we have recently integrated timetabling for student assistants into the MosesKonto. Similar to the procedure for the students the student assistants can access the web-interface of the MosesKonto and enter their wishes for their next semester’s assignment. As the MosesKonto also holds the student assistants’ course of studies and study progress the abilities of the student assistants are directly at hand. We also know immediately how many tutorials each student assistant has to teach during the semester. Depending on the student assistant’s respective type of contract 40–80 hours per month this results in 2–4 tutorials (including homework revisions) per week. The actual demand for tutorials for each of the lectures integrated in the MosesKonto is calculated on the basis of the student registrations for a specific lecture divided by the maximum population for this lecture. For mathematics service lectures this maximum population ranges from 15–25 students. All specified problems are important for room management at universities, for the realization of courses that can be studied according to curricula, and for the satisfaction of students and teachers. These problems are related to the optimization of room management and personnel costs (e.g. by a uniform distribution of students). Thus, the solution of these problems is related to the optimization of “real” costs, a more and more important economic factor at (German) universities. Introduction of the two-tiered Bachelor and Master courses has raised awareness for these problems at German universities: due to the multitude of new courses the timetables, which have been established and stood the test of time, cannot be used any longer. Moreover classes tend to be more structured and school-like; attendance is compulsory and dependencies between modules are depending on the feasibility of the curricula. This feasibility is also evaluated while accrediting new study courses. Since 2003 for the solution of the post enrollment based course timetabling problem at the TU Berlin we are using an algorithm that has been realized in our team [GJL+ 06]. For the third problem, the examination timetabling problem (sec. 1.3), an algorithm has been developed within our team, that is currently implemented to solve the real data problem for a large set of courses [Lac08].
2 Organization of Classes at TU Berlin One of the major challenges facing universities is the organization of the study supporting processes, especially in freshmen courses [JLPZ07]. With student numbers of about 30.000 (cf. Fig. 1) and about 6.000 freshmen students (cf. Fig. 2), TU Berlin
74
Sabina Jeschke et al.
Fig. 1 Student Numbers at TU Berlin from winter semester 99 until winter semester 08, translated from [rec]
Fig. 2 Medium term development of number of freshmen students, translated from [rec]
is one of the largest universities of technology in Germany. As a service for other faculties, the institute of mathematics is in charge of the mathematical education of most students, independent of their actual course of studies, making it the biggest “service provider” of the university. Students from more than 20 different programs attend one or more courses of the nine mathematics for engineering modules: • • • •
Calculus I-III, Ordinary Differential Equations, Integral Transformations and Partial Differential Equations, Linear Algebra,
Management and Optimal Distribution of Large Student Numbers
75
• Numerical Mathematics, • Mathematics for Economics I & II. The freshmen course Calculus is the largest module with a total count of more than 2.200 students per semester. In order to guarantee lecture class sizes of less than 250 attendees the modules are organized as follows: multiple lectures are held in parallel and the lecturers of the individual lecture classes take care of presenting the same mathematical material to all attendees. In addition to lecture classes, all students have to solve the same exercises as homework to be admitted to the written examination at the end of the semester - in case of a sufficiently high homework score. Students also are eligible to attend small-sized exercise/recitation classes (tutorials) consisting of 15 up to 30 students. The administrational duties for performing mathematics service are summarized in the following list: 1. Assign all math-service students to exercise classes 2. Manage homework scores for admission to final examinations and course credits 3. Management of student registration for examinations 4. Inform students about examination scores 5. Collect and submit examination scores to the central office of examination. Although all of the above tasks are typical administrational duties for student management at universities, the large course sizes make them very time-consuming and labor-intensive. This is especially true for the assignment into exercise groups for every module: All students have to be assigned to small groups such that assignments do neither conflict with each other nor conflict with the individual schedules of the students while respecting certain capacity restrictions. However, the large course sizes not only present difficulties; they imply a great opportunity of rationalization in the student-administration. Against this background the development of MosesKonto ([GJL+ 06], cf. Figs. 3,4,6 for screenshots) began in 2002 and was initially deployed and used since 2003 for all courses within the math-service modules described above. Since the winter term 2005/6, the assignment into exercise classes and management of exams has been extended to cover further courses, even across different schools. In the winter term 2007/08, 6 out of 7 schools at TU Berlin used the MosesKonto for the post-enrollment into their tutorials, i.e. approximately 50% of all students are distributed into their tutorials by this system. In winter term 2008/2009 15.142 tutorial places were distributed to 5.498 students (thereof 1.928 freshmen). In 2008 we allocated a total of 29.138 places. Please see appendix 1 for the complete results for the winter term.
3 Student-Tutorial Assignment Every student attending a course with additional tutorials is eligible to attend an exercise class. The number of attendees for a particular course is not known until the semester starts, as the enrollment process does not end until then. The students cannot know their complete schedule until this time either, and the variety (with
76
Sabina Jeschke et al.
Fig. 3 Web-administration-interface of the MosesKonto, displaying the menu functions for “persons”, “tutorials”, “examinations”, “administration” (left to right)
respect to course of study) of students attending different courses implies a great variety in individual schedules. These two details make it impossible to fix a set of dates for exercise classes in advance. Instead, one is forced to find dates for these tutorials in the first week of a new semester, by taking into account all students’ timetables. Until winter term 2002/3, the assignment of students into exercise classes was performed independently for every math-service module (and for each of the other modules). This procedure required all students of a particular module (up to 1.000 at that time) to gather in the main lecture hall in order to receive their exercise class dates. Students were numbered and subsequently drawn by lot to choose their most preferred date from the remaining dates in the global pool of exercise classes. Of course, this procedure could not guarantee every student receiving his most preferred date, but it did ensure that the assignment of a whole course is feasible within 90 min. Integration of additional courses (thus, increasing the number of students to be administrated by math-service) invalidated the above assumption about the feasibility of the assignment method. In fact, the last time the above method was applied; it took several hours and forced more than 2.000 students to be physically present in a completely overcrowded main lecture hall offering seats for 1.200 students.
Management and Optimal Distribution of Large Student Numbers
77
Fig. 4 Web-administration-interface of the MosesKonto, setting the compulsory modules
Since this development in the number of attendees was foreseeable, the development of a web-based registration procedure began in the spring of 2002; the new method has been used since the summer term of 2003. The major goals of the new procedure were: 1. Gather early information on the expected number of students for every module 2. Take into account the student’s individual schedules by collecting their “wishes” for dates 3. Distribute students as even as possible among all exercise classes of every module 4. Assign students such that no date conflict arises for students that attend tutorials in more than one course. For various reasons it was impossible to access the personal data of the students from the central enrollment office. Thus, to realize a procedure that respects the above requirements, students would need to register at MosesKonto and create a personal account. But, as a side effect, the same personal data can be used to manage the final exams. The latter is a great improvement over the established way of registering for an exam: students had to register at the central exam-office with a paper form, from which a press copy of their registration form is sent to the department of mathematics, where these copies are entered in an excel sheet. Of course, numerous transcription errors are produced that way.
78
Sabina Jeschke et al.
All of the above requirements could be easily met by means of an assignment procedure operating on a first-come-first-served basis. However, it would favor those students over others, who receive their certificate of enrollment earlier. Moreover, from a technical point of view, it would concentrate the registration traffic on a very short time span, which means that the system probably would not scale well for increasing numbers of students. This motivated our decision for a global optimization method: Over a given time span, say, two weeks, students have the opportunity to register for the tutorials. When we started in 2002 the system had only been used for the math-service tutorials. The students could choose personal priorities from a range of possible dates for every math-service module. Since the system is used for many other lectures with tutorials, the students can globally choose priorities for different time slots. All students will choose their favorite dates using their MosesKonto account and they can revise their choices at any time. After the registration time span, the collected data is used to compute an assignment that is optimal in respect to all students’ priorized wishes. Often, students have organized themselves into groups of two or three to work collaboratively on their homework, for example. It is desirable, although not essential, to let all group members attend the same exercise class. MosesKonto offers the opportunity to register for tutorials as a group (different students can form different groups for tutorials of different lectures), which implies that all group members automatically choose the same priorities for the available tutorial dates. The functionality to mange groups (change groups, step back, etc.) is provided through MosesKonto as well. These groups are not formulated as constraints in the global optimization problem, but some care is taken to respect these wishes for groups if possible.
4 Optimized Assignment of Student Assistants to Tutorials The optimization step involves the computation of all assignments of student assistants to their tutorials, so that room-capacities and maximum tutorial population are respected. The computed solution is optimal in respect to the tutorials required and to the wishes and capabilities of the student assistants. The problem admits the following formulation as a constrained minimum-cost-flow network problem. [AMO93, CSLR01, Löb00, NT03, FS03]
min s.t.
pn (t j , ci )xt j ci xt j ci = 1∀t j
ci ∈courses(t j )
supply(t j , ci )xt j ci ≥ demand(ci )∀ci
xt j , ci ∈ {0, 1}
Management and Optimal Distribution of Large Student Numbers
79
Fig. 5 Tutor-assignment network
An example of this network is depicted in Fig. 5. A detailed description of the analogously implemented network flow problem for distributing the students to tutorials can be found in [Luc03]. The data from MosesKonto is accessed and preprocessed by the software TUTOP [Luc03], which also formulates the integer program that is then to be solved by the commercial software CPLEX [ilo].
5 Administration of Examinations In most courses using the MosesKonto to distribute their students into the tutorials, each student has to pass a written final examination (consisting of up to three separate written tests) for each module. As a result, up to 2.500 written exams have to be handled in each course, creating a substantial administrative overhead. The efficient organization of such large exams requires punctual registration on the part of the participating students. The results have to be published and forwarded to the central office of examinations. Finally, the results have to be processed statistically to provide the responsible deans with information concerning the success of the courses. To satisfy the regulations imposed by the different courses of studies it is necessary to hold up to three different (though identical in content) exams. As a result, the students feel insecure for which exam they have to register. To alleviate this
80
Sabina Jeschke et al.
problem, the students are only presented with the appropriate exam for their particular course of studies when registering for the final exam for a given module. Registration for final exams for compulsory optional subjects and exams governed by older or more “exotic” examination regulations has to be done in person at the service center, which also has access to the examination administration of the system. When “creating” an exam in the database, the registration and deregistration periods have to be specified (cf. Fig. 5). Students have to (de-)register during that specified period only, deregistering for (officially certified) medical reasons being the only exception being handled exclusively by the central office of examinations. Information for each exam is available for download by the service center or other authorized staff in the form of a zip-file containing the following: (1) A list of all students registered for this exam, including their personal data and the results of the exam (in csv-format) (2) Separate lists for each course of studies, containing the names and personal data of each student registered for the exam to be forwarded to the central office of examinations (LATEX-Format) (3) A complete list of all registered students including personal data for proof of identity during the exam (LATEX-Format) (4) Separate lists for each course of studies, containing the names, personal data and exam score of each student registered for the exam; these lists can be directly imported into the database of the central examination office (Excel-format) (5) A complete statistic including a graphical representation of the results (LATEXFormat)
Fig. 6 MosesKonto web-administrator-interface displaying an overview of tutors and their desired courses
Management and Optimal Distribution of Large Student Numbers
81
The lists of registered students forwarded to the central office of examinations replace the more traditional registration forms handed in by the students in person. In the near future, this procedure will be replaced by an automatic export into the database of the central examination office. Until winter term 2006/07, the lists containing the result, once signed by the professor, replaced the minutes of the examination previously required for each student separately. Since 2007 the results are imported directly into the database of the central examination office. The complete csv-format list of participating students is used as the basis for assigning the students to the rooms available for the examination and to determine the needed capacity and thus required rooms in the first place (cf. Fig. 6).
6 Administration of Homework Most courses require mandatory homework as a prerequisite for admission to the exams for most courses of studies. The homework-related criteria that are to be met for admission to the exams are currently dependent on both the module and the professor teaching it. In consequence, it is necessary to store the criteria for each module and each semester separately. Authorized staff can access the list of course participants at the end of the semester to add if the student met the homeworkrelated criteria or not. The system does not check if the homework-related criteria are met during the online registration for the exam as the last homework assignments are regularly neither handed in nor graded by the time the registration period has expired. Staff members can afterwards easily get a list of the students who registered for the final examinations, but do not fulfill the homework criteria.
7 Cost/Performance A single user license of the optimization software used by us, CPLEX, for commercial purposes is approximately C 15.000 with additional annual maintenance costs of 18%. For a period of 10 years this totals in costs of C 42.000 or C 4.200 yearly costs. In the year 2008 the system allocated more than 29.000 tutorial places. Thus, the cost per tutorial place allocated is less than C 0.15. The personnel expenditures for doing the distribution by hand are difficult to estimate, yet we try to give a comparison: the cost per year roughly equal a scientific staff’s monthly wage. For simplicity reasons we assume that this scientific staff works one whole month (we should actually consider two times two weeks, as the distribution has to be done twice a year) on the tutorial distribution, then in order to finish the job on time for the allocation of every single tutorial place less than 22.5 seconds are available, whereas the system needs less than one minute to provide an optimized solution for all places. This simple thought experiment shows that in terms of cost and velocity our system clearly outperforms a manual distribution. Moreover, every teaching staff who is confronted with schedule difficulties from students and colleagues knows that
82
Sabina Jeschke et al.
manual solutions, in the majority of cases, are far from optimal. The development costs of our system would have to be added to the above cost, however they could be neutralized as this work was done as a regular research activity. Since CPLEX universally is usable therefore can be used for all mentioned distribution problems (assistant-, tutorial-, room- and exam distribution), the costs go down correspondingly. An exact estimate is difficult because the previous costs are based on personnel expenditures and are not on hand. Furthermore, “costs” like administrative-staff, teaching-staff and student satisfaction cannot be estimated. At TU Berlin all participants involved are extremely satisfied; especially the teaching colleagues who are exempt from these organizational burdens at the beginning of each semester. Another advantage arises from our modus operandi: the assignment of student assistants can be planned within whole departments, e.g. mathematics or mechanics beyond specific lectures; i.e. assistants can be allocated to the specific courses “on demand”, depending on how many students have registered for a particular course. Moreover, human resources are not only optimally employed; personnel costs are saved by avoiding offering tutorials for sparse auditoria. Table 1 Registrations for written examinations in the system 2005-2009 Year
#Registrations to written exams
2005 2006 2007 2008 2009 so far Overall
8.475 11.012 14.848 19.473 25.561 116.732
8 Résumé and Perspective One desirable future feature of the system would be a direct, automatic exchange of data, in particular, with the room management system of the TU Berlin. This exchange could be realized by using web services. Preliminary efforts of coordination have shown the general willingness of the central facility management to participate. However, certain adaptations to their software are required and are not yet implemented. For efficiency-reasons and student-contentedness, of course in the medium-term it is planned to manage all examinations and optimize all tutorials at TU Berlin using the MosesKonto within two years. Anyhow, up to now MosesKonto is a service, offered by the Center for Multimedia in Education and Research (MuLF) and every department is invited to take part on a voluntary basis. We have shown that the timetabling problem of post-enrollment can successfully be solved with mathematical techniques of discrete optimization, see [22] as
Management and Optimal Distribution of Large Student Numbers
83
a general reference. For the future our group at MuLF is developing a solution for the university timetabling problem (for all lectures at the TU Berlin) and for the examination timetabling problem. The final aim is a system for the solution of these most important administrative problems at universities.
References [AMO93] [CK96]
[CSLR01]
[FS03]
[GJ90] [GJL+ 06]
[ilo] [JLPZ07]
[Lac08]
[Löb00] [Luc03] [NT03]
[rec]
Ravindra K. Ahuja, Thomas L. Magnanti, and James B. Orlin. Network Flows: Theory, Algorithms, and Applications. Prentice Hall, New Jersey, USA, 3 edition, 1993. Tim Cooper and Jeffrey Kingston. The complexity of timetable construction problems. In Practice and Theory of Automated Timetabling, First International Conference, Edinburgh, UK, August 29 - September 1, 1995, volume 1153/1996 of Lecture Notes in Computer Science (LNCS), pages 281–295. Springer Berlin / Heidelberg, 1996. Thomas H. Cormen, Clifford Stein, Charles E. Leiserson, and Robert L. Rivest. Introduction to Algorithms. MIT Press and McGraw-Hill, New York, USA, 2 edition, 2001. H. Salehi Fathabadi and G.H. Shridel. AN O(nm2) TIME ALGORITHM FOR SOLVING MINIMAL COST NETWORK FLOW PROBLEMS. Asia-Pacific Journal of Operational Research, 20:161–175, 2003. Michael R. Garey and David S. Johnson. Computers and Intractability; A Guide to the Theory of NP-Completeness. W. H. Freeman & Co., 1990. Sven Grottke, Sabina Jeschke, Gerald Lach, Robert Luce, Olivier Pfeiffer, Jan Sablatnig, and Erhard Zorn. MosesKonto: Optimiertes Verteilungsverfahren für Tutorien und Studierendenverwaltung an der TU Berlin. In DeLFI 2006: 4. e-Learning Fachtagung Informatik der Gesellschaft für Informatik e.V. (GI). Lecture Notes in Informatics (LNI) - Proceedings (P-87, pages 385–386. Gesellschaft für Informatik, Bonn, Germany, 2006. ILOG 2008 High-performance software for mathematical programming and optimization. Sabina Jeschke, Robert Luce, Olivier Pfeiffer, and Erhard Zorn. Optimized Allocation of Exercise Classes and Study Management at TU Berlin. In 6th Annual ASEE Global Colloquium on Engineering Education, Istanbul, Turkey, October 2007. M. Lach. Ein Verfahren zur Optimierung der Klausurterminplanung an der TU Berlin. Masters thesis, Technische Universität Berlin, Institut für Mathematik, 2008. In German. A. Löbel. MCF – A network simplex Implementation, 2000. R. Luce. TUTOP: Tutorienplätze optimal verteilen. Seminarpaper, Technische Universität Berlin, Institut für Mathematik, Germany, 2003. V. Nguyen and Y. Tan. Minimum convex cost flow problem. In Proceedings of the 2003 Joint Conference of the Fourth International Conf. on Information, Communications and Signal Processing, 2003 and the 4th Pacific Rim Conf. on Multimedia, 2003, volume 2, pages 1248–1252, 2003. Rechenschaftsbericht des Präsidenten der Technischen Universität Berlin, Teil 2. Technical report.
84
Sabina Jeschke et al.
Appendix: Winter-Semester 2008/09 tutorial allocation data
course
Table 2 Statistical data from summer-semester 2009 tutorial allocation average ununpriority 1 2 3 4 5 6 7 matched wished
Ana1
1.33
897
277
51
11
0
0
0
0
0
Ana2
1.25
1070
249
43
3
0
0
0
0
0
Ana3
1.70
81
47
17
5
1
0
1
0
0
AVWL2
1.42
219
72
20
7
0
0
0
0
0
BauInf
1.40
68
27
5
1
0
0
0
0
0
CAD-KL1
2.26
335
89
83
32
43
51
4
0
0
DGL
1.69
184
95
28
14
7
1
0
0
0
EL-NW
1.70
141
66
28
13
4
0
0
0
0
ETGLServ
1.42
341
99
29
14
1
0
0
0
0
Info2-Wirt
1.61
150
71
30
9
0
0
0
0
0
InfoRules1
1.13
70
10
0
0
0
0
0
0
0
ITPDG
1.55
78
43
12
1
1
0
0
0
0
K1
2.25
382
161
73
58
51
36
18
0
0
K2-A
1.42
126
55
9
1
1
0
0
0
0
K2-B
1.40
97
36
9
1
0
0
0
0
0
KW
1.86
165
74
32
17
11
3
2
0
0
LinA
1.48
754
261
80
29
7
1
0
0
0
Mech1
1.63
362
153
53
26
6
3
1
0
0
Mech2
1.41
386
135
45
3
0
0
0
0
0
Mech3
1.69
46
23
10
2
2
0
0
0
0
Mech3-EM
1.92
42
36
20
3
1
0
1
0
0
Mech3-KM
1.95
20
8
10
2
0
1
0
0
0
MechE
1.54
285
95
55
12
0
0
0
0
0
MMPhy
1.17
156
26
3
0
0
0
0
0
0
MPGI2
1.52
178
106
26
1
0
0
0
0
0
NumI
1.97
54
24
13
8
3
2
1
0
0
Ph-Ing2
1.46
458
171
46
16
0
1
0
0
0
PhysET2
1.35
128
35
9
3
0
0
0
0
0
Stoch-Inf
1.16
137
22
2
0
0
0
0
0
0
TechGI2
1.59
193
56
29
19
1
0
0
0
0
TechGI2-P
1.52
44
42
2
0
0
0
0
0
0
TechGI4
1.48
106
38
19
1
0
0
0
0
0
TheGI2
1.41
135
48
14
2
0
0
0
0
0
TheGI4
1.62
78
31
20
4
0
0
0
0
0
WT1
1.91
434
177
78
47
38
16
3
0
0
all
1.58
8400
2958
1003
365
178
115
31
0
0
Spirit University of Stuttgart’s Life-Cycle-Based Gender-Mainstreaming-Concept Sabina Jeschke, Barbara Burr, Peter Göhner, Wolfram Ressel, Wolfgang Schlicht
Abstract In spite of social and political efforts to achieve equal opportunities, women remain a minority in natural sciences, technical and related fields. We present the gender concept of the University of Stuttgart. First, the steps for promotion of female students within natural sciences and technical fields are developed. Keywords gender concept · female academic education · diversity studies · women in natural sciences and engineering
by men willing to pursue an academic education [Win04]. The existing loss of interest by men in technical and engineering subjects intensifies the challenge [KW03]. Moreover, demographic development in Germany and global competition exaggerate these challenges. • Technical disciplines graduates are filling important jobs in our society. These are characterized by fields of responsibility and extensive influence. Technical authorities are a key factor of shaping our society [Wäc98]. The under-representation of women in these fields is a central drawback to an equal participation of women. In fact, there are lots of single initiatives aimed at “breaking down the gender-gap” in German universities. Each engaged by different single persons that include offices for equal opportunity but also individual professors, centres, or initiatives in single institutes. Consistency, continuity and transparency are missing due to the lack of a comprehensive strategy. This leads often to a “twofer” for some target groups while other arrangements and initiatives are missing. Communication and information campaigns aimed at reaching a broad group. Creating a common understanding of challenges and requirements are an exception. The objective of universities should be concentrate and coordinate existing and successful measures. All arrangements result from one coordinated gender master plan, • that comprises the full life-cycle from kindergarten to professorship with leadership function, • whose single steps make a seamless, coordinated transition between different measures possible, • that incorporates all areas (education, research and organization), • and that includes a concept of family-friendly policies for all members of university. The female professor program is a characteristic component of the gender master plan of the University of Stuttgart. The under-representation of women in technological fields becomes particular visible in predominantly scientifically and technologically aligned universities like the University of Stuttgart. Contemporaneously, a large percentage of future scientists and engineers are trained here. These universities are in charge to overcome the gender gap in the technological disciplines. This is justified broad and early commitment of gender management. Additionally, women are nominated as top position holders of technical disciplines in order to increase diversity in research and education presenting female role models to new generations.
2 Situation-(Deficit-) Analysis Girls and women are still under-represented in technology-oriented programs. The University of Stuttgart is well aware of these challenges being a long term driving force to raise the percentage of females. The structure and development plan SEPUS
Spirit
87
of the University of Stuttgart V defines measures. Each of the listed measures is successfully proven. However, the University of Stuttgart neglects a master plan that connects all of single measures and sustainable networks that build on and support each other. The concept presented here pursues the goal of integration and sustainability of measures. The University of Stuttgart is a fully profiled university oriented towards technological and scientific programs and the concept of equal opportunity is aligned with this profile. Therefore, the focus is on the recruitment of women into technical disciplines. They are initially acquired for scientific and technical degrees, then promoted depending on their talents, and supported on their “job ladder” inside or outside university. The objective is to boost the percentage of female students in scientific and technical disciplines from currently 20 % to 30 % (within the entire university from 34 % to 40 %) and the rate of female young scientists from 18 % to 30 % (within the entire university from 21 % to 30 %). The target in these disciplines is set for an increase of female’s professors from 6 % to 10 % (same within the entire university) within the next five years. Furthermore, 30 % of the professorships have to be newly appointed until 2012. Therefore, a considerable increase is possible. Baden-Württemberg is striving for a long term increase of the quota of female professors to 30 % (defined by the head of the Office of Scientific and Technological Affairs, Frankenberg).
3 Objectives concerning different target groups Chapter 4.1 describes the different measures to be implemented within the framework of the gender master plan of the University of Stuttgart in detail. This chapter summarizes the most important foci from the catalogue of measures concerning addressing the different target groups: • Increase percentage of women in top scientific positions The incompatibility of a scientific career and family is still a major reason for women to distance themselves from the former . To address this challenge, the University of Stuttgart is working towards improving the surrounding circumstance to unite a scientific career with family life. The intention of sufficient child care offers, a family friendly housing situation, the adjustment of study and examination regulations and development of alternative work schedules are supporting this goal. Another important point considering the encouragement of women in leading scientific positions is the inclusion of the career planning of their partners (Dual Career). The University of Stuttgart is aiming to increase the ratio of female professors to 10 % according to their structure and development plan. To create the long-term framework for this goal, the University of Stuttgart is developing a package of measures. Especially qualified female students and young female scientist are provided with individual sponsorship and a structural framework to improve the compatibility of scientific career and family. The financial support through scholarships is seen as of similar importance as the personal support through formal and casual networks
88
Sabina Jeschke et al.
and the acquisition of career-furthering qualifications. Versatile cooperations with regional companies put the University of Stuttgart in the unique position to offer attractive choices within the industry and economy to partners of female researchers. • Career and personal development for young female scientists Next to the scholarships for talented young female scientists mentioned in the previous paragraph there is the phase between academic studies and the begin of the scientific career as well as the Post-Doc-phase. It is important for future scientists that they are presented with career-options and develop career-strategies during these times. They need to gain access to networks and acquire leadership and management qualifications. Community building between students and young scientists is fostered by employing comprehensive measures within the single institutes and creating positive role model initiates a regular exchange of experience between successful young scientists and female entrepreneurs. It is important for raising female students and young scientists to create a supportive living environment and to present attractive employment opportunities to their partners, in short, offering Dual Career Options as early as the qualification phase. To achieve this goal the University of Stuttgart has set out to earn the certificate of “family friendly university” awarded by the Hertie-Trust. • Acquisition of female students in the natural sciences and technological disciplines A central condition is the gain of enthusiasm for natural sciences and technology to foster the election of a scientific or technologically oriented field of study later in life. The University of Stuttgart has build a package of measures that incorporates the education of teaching staff as well as considering research of gender sensitive deployment of toys and the creation of an academy for younger children. Furthermore, the University is compiling effective public programs - female students are given an age-based understanding of natural and technical questions. For this existing programs like the Girl’s Day are expanded and new choices like the Technology-Camp are developed. Female students can use their vacation time to research exciting scientific and technological topics. To ease the decision of girls to choose a technological or scientific program of studies there is the offer of a trial course as well as a program to support them from the last year of high school to the first semester at the university. Building on the program “Try the University” girls in the last year of high school are being offered the full curriculum of available bachelor degrees and are enabled to collect credits towards their future studies.
4 Implementation of the concept SPIRIT The concept is following a life-cycle model. The target groups are girls and women which are accompanied from kindergarten to professorship by multifaceted package of programs: The single activities are weighted differently depending on the phase
Spirit
89
of life, build on each other and are interlocked. This ensures that measures exist for each state of the life cycle that aim towards a goal and support the progress towards the next phase. Special weight will be put on the systematic advancement of female migrants due to the fact that their requirements are defined by a different process of socialization compared to women without a migration background.
4.1 Measures for Each Scope of the Life-Cycle: 4.1.1 Gender Sensitivity Training Courses for Educators Measures have to be taken already in kindergarten to ensure the best possible equal upbringing of girls and boys. To spark interest in natural and technical questions kids requires careful planning of all educational activities. To improve this concept the University of Stuttgart designs, prepares and implements gender- and diversitysensitive scientific and technical professional training for educators. Evaluation and counselling is provided by the Institute for Social Sciences. 4.1.2 TechToy Gender and diversity sensitive engineering toys that are accepted and appreciated by both sexes are necessary to awaken the enthusiasm for the natural sciences and engineering early and across both genders. The University of Stuttgart will apply a main research focus on the development of gender sensitive engineering toys. A co-operation of different chairs of engineering, the Institute of educational science and psychology, and the institutes for social sciences, technology and environmental sociology [ZR] will carry this research focus. This concept is built on the preparations for the initiative “Wissensfabrik” (knowledge factory). The University of Stuttgart will shortly join this initiative as one of the first Universities. A pedagogic concept is being developed to allow for a meaningful use of the toys in pre-schools, schools and families. Such a concept could look similar to – the extremely successful – Roberta- teaching materials of the Fraunhofer IAIS for LEGO Mindstorms robot building (a Roberta Regional Centre is currently being build up and developed at the Institute for IT Service Technologies). 4.1.3 Exhibition Team University of Stuttgart (ETUS) A professional Team is build at the University of Stuttgart whose focus will be public relations with specific target groups. The field of responsibility for this team also includes the development of effective publicity programs for events at the University of Stuttgart including the Day of Science (“Children’s-Campus – Program for young researchers”), as well as Technology exhibitions like the “Land of Ideas” or the “Cebit”. During this process programs for the different target groups of preschool age, elementary school age, junior high school, secondary school age, and for teachers are prepared. This project can build on the experience of the Student Counselling Centre and of the marketing office.
90
Sabina Jeschke et al.
Fig. 1 Life-cycles from kindergarten to professorship
4.1.4 Rent-a-Scientist & SchoolgirlUni Pre-schools and schools are being offered possibilities to “rent” (female) professors or (female) research staff of the University of Stuttgart. They visit the educational institution and answer questions from the fields of science and engineering with age group specific lectures and demonstrations. In reverse, the University of Stuttgart offers (female) students the possibility to come to the University to get to know the University “from the inside” (student/schoolgirlUni ). To ensure the success of these events a pedagogical concept is drawn up and staff members are being trained (pedagogical mentoring by the Institute of educational sciences and psychology). The concept also addresses culture-based conflicts of interests – especially relevant for pre-schools and schools with a high ratio of children from migrant families. The measures 4–6 and 8–10 are part of the University of Stuttgart Young Academy which is currently being implemented.
4.1.5 Girls’ Day The University of Stuttgart has been participating in the mono educational Girls’ Day for several years. Girls are invited to the University of Stuttgart to gain a short inside look into the different scientific and technical areas. Specialized subjects from
Spirit
91
different departments are being prepared to spark the girls’ interest. In this regard the girls have the possibility to try the experiments also on their own. The program will be further developed in the coming years and expanded inside of the university – through the integration of more faculties and considering culture-specific differences in interests and approaches. 4.1.6 Partner Schools Direct connections to partner high schools improve the cooperation between schools and universities. Common activities give female students a chance to gain some early insights into research and education at the universities. The choice of study and the ease of transition from high school to university should benefit from these measures. They should further benefit from the close cooperation with the successful “Mine-Mint”-Network. The continuing initiatives involving the Institute for Hydraulic Engineering, the Department of Physics, the Department of Chemistry, as well as the Department of Computer Science, Electrical Engineering and Information Technology and the freshly initiated cooperation with the “Landesgymnasium für Hochbegabte in Schwäbisch-Gmünd” will be continued respectively further expanded. 4.1.7 Gender Sensitivity Training for Teachers This training course is designed to enhance the skills of teachers in gender and diversity sensitive didactics for teaching technological or scientific courses. As the result of both formal and family education, girls are interested in different questions than boys. The course develops guidelines based on concrete examples to help teachers create a sense of fascination and adventure concerning science and technology in students. These courses are supported by the competency of the Institute for the Social Sciences. The University of Stuttgart takes advantage of the unique situation created by the restructuring and modularization of the academic education of teachers in Baden-Württemberg to enhance the competencies of teachers in training for gender sensitivity. Current plans allot 6 credit points for this field. 4.1.8 Technology-Camps Technically-oriented summer camps teach modern science in a hands-on approach. These “Technology Camps” present selected current topics from the fields of research and education at the University of Stuttgart in high school level lectures and experimental labs. High school and university students research interesting topics from the natural and engineering sciences in close cooperation with professors. These summer camps expand upon the already existing summer break programs offered by the University of Stuttgart (Stuttgarter Forschungsferien). The University of Stuttgart can build on experiences gathered in the (Nanocamp, Faculty for Civil- and Environmental Engineering). Future summer camps will be
92
Sabina Jeschke et al.
offered as both mono or co-educational. Mono-educational summer camps will make participation easier for female students from families with a migration background. The SPIRIT program plans to offer stipends for students from financially weaker families through the integrated foundation.
4.1.9 TechnoClub “Test your University” The TechnoClub offers single workshops and one-semester courses for female high school students, held at the university. Lectures and labs held by members of the university are designed to familiarize high school students with the daily life at the university. These courses include a lab for Robo-Rescue and courses for electronics and soldering. Special attention will be given to culture-specific variations in interest and approaches in support of the main focuses 4, 5, and 8. The courses offered in the TechnoClubs are created by a representative cross section of all institutes within the university and are centrally coordinated. This mono-educational concept is based on a long running and successful program, “Try the University” (“Probiert die Uni aus”).
4.1.10 Studium Experimentale The “Studium Experimentale” presents several different approaches to familiarize students with the University of Stuttgart and the different programs offered. First of all, high school students are offered university level courses in a more compact form, including the final exam. This idea is based on the “SchülerStudium” at the FU Berlin. If a participating student should decide to pursue a compatible course of study at the University of Stuttgart, they will receive priority in the acceptance process and will be credited with the corresponding credit points. As a result, ties between students and the University of Stuttgart are created early and the transition from high school to university is eased. Secondly, since female students often find it hard to decide on one single technological or scientific course of study, the University of Stuttgart offers one-year courses with access to all available bachelor courses without forcing the student to enrol with any one particular program. The credit earned during this period is fully applicable to any future course of compatible studies. The University of Stuttgart is currently engaged in defining the necessary legal framework (crediting, BAFöG).
4.1.11 Mentoring “Transition from High School to University” The Campus Mentoring Program matches selected university students with high school students to mentor them at the end of their high school and the beginning (first few semesters) of their university education. The mentors help the students adapt to the daily routine at university. The students will be gradually transferred to an advanced mentoring program supervised by professors as they enter university.
Spirit
93
4.1.12 Gender Sensitivity in Technologically-Oriented Courses of Study Technological-oriented courses of studies and careers are not unattractive to women per se. However, motivation, specific interests, style and strategies of learning, goals and expectations can differ significantly from those of young men. The University of Stuttgart, under the leadership of the vice rector for academic affairs, will create concepts and guidelines for gender sensitivity in all courses of studies, taking advantage of the opportunity offered by the restructuring of all courses of studies, including their content, to the new, modular bachelor/master system. 4.1.13 New Technologically-Oriented Programs Quite often female students fail to feel attracted towards the existing curriculum of engineering and natural science programs. The main reasons are that women do not see the social relevance and/or that they are worried whether their interests and talents qualify them to pursue studies in these fields. The Bologna-Process offers a unique opportunity to create new and reorganize existing programs. The University of Stuttgart is planning to implement number of new programs specifically addressing the specific interests of female students. These programs will build on the experience from the Galilea-Project at the TU-Berlin (www.galilea.tu-berlin.de, already existing cooperation [DJTW06], [DJSW07], [EJN+ 07], [EJN+ 08], [DEJ+ 08a], [EJNS08], [DEJ+ 08]. The first two programs are already being designed (Renewable Energy and Medical Technology). The Galilea-Programs are characterized by a high degree of flexibility in the early choice of courses, as well as socially relevant, interdisciplinary foci. 4.1.14 Diversity Studies & Technology Management It is necessary to understand the role of female engineers and scientists, including their special contributions and unique potential, to attract more women to studies and careers in the technological fields. To address these requirements, the University of Stuttgart is planning to create a program for diversity studies in the engineering sciences with a focus on technology management. The courses offered in the program will be credited as minors or mandatory elective courses in the engineering, economics, scientific and social studies programs, strengthening the impact of gender and diversity studies in the existing programs. 4.1.15 Program for Stipends: “Opportunity” Stipends provide a means for direct promotion of talented students. As such, it is a tool well suited for supporting students from financially weaker families. In addition, they create social networks between the recipients and bring them into contact with prospective employers (often former recipients or even founders of the stipends). The University of Stuttgart plans to create specific stipends for female students of the natural sciences and engineering in close cooperation with
94
Sabina Jeschke et al.
successful regional companies and research institutions and with the European Social Fond, ESF. This increase in cooperation will further strengthen the regional economy. The “Opportunity” program complements the existing stipends (stipends for PhD students under the “Landesgraduiertenförderungsgesetz”, Schlieben-Lange und Margret von Wrangell Programm). 4.1.16 Femtec Region Stuttgart The low ratio of women in positions of leadership within the economy and the resulting perception of a lack of career opportunities has a direct, negative impact on the motivation of female students. The University of Stuttgart is planning to expand the existing cooperation with the successful Femtec program and its associated career building program. Femtec is a unique network of leading technical universities and successful international companies. The University of Stuttgart has been a member of the Femtec network since 2005. The aim of the program is on strengthening the cooperation with the regional economy. The associated career building program teaches leadership qualifications easing the transition to a successful, professional career. The University of Stuttgart is already represented on the advisory board of Femtec. 4.1.17 Entrepreneur Initiative SPIRIT Based on the experience gathered in the joint project “Erfolgreich ist weiblich!” (in cooperation with the TTI GmbH, started in February 2007) the University of Stuttgart is creating an informational and consulting offer for students and post graduates as well as the alumni of the University of Stuttgart that wish to found their own start-up. This network will provide gender specific support within and beyond the boundaries of the university itself. 4.1.18 Mentoring for Women in Science and Research In addition to good qualifications and grades it is often the contact and support resulting from access to informal networks that is decisive in professional success. The University of Stuttgart is planning to expand existing and successful mentoring programs for women in science and research to support even more female students in implementing their career strategies. The mentoring program provides support for highly qualified graduates, post graduates and post docs. An additional focus of the program is on qualifying women for leadership positions. 4.1.19 Mentoring Program “ProFiL BW” The University of Stuttgart is planning to create a program to support excellence in scientific careers, following the example of the successful elite mentoring program “ProFiL” (http://www.profil-programm.de/, existing cooperation) of the three universities in Berlin. Future expansions of the program will include cooperations with
Spirit
95
other universities in Baden-Württemberg. The (female) participants of this program will receive extensive support in planning and furthering their careers, including training leadership and management skills required of a full professor. Prerequisite for joining this program is an excellent PhD thesis, a one or two year post doc phase and an interview with the applicants. 4.1.20 “Ladies’ Community” Female students often feel isolated during their studies due to their low ratio in most engineering programs. Support programs like girls-only exercise courses are faced with a low acceptance as they are perceived as further reaffirming this outsider role. The lack of social infrastructure on-campus will grow even worse in the near future as first semester students are getting younger due to the shortened high school education. This development necessitates a pervasive community building among female students beyond the boundaries of the single study programs. The University of Stuttgart has created a work group to design a concept for realizing such a community. Some measures already planned include regular seminars with successful female entrepreneurs and researchers, regular social gatherings and the creation of a special web portal for the female community. The community to be created should ideally include all female students and alumni of the university. 4.1.21 Colloquium “Women in Leadership Positions” Female students are often missing role models, women who have mastered the obstacles and challenges of a program and have reached a leading position in research or business. This results in low motivation and an increase in the drop-out rate. To counter this effect, the University of Stuttgart is planning to hold regular seminars where successful women from research and business share their experience and advice with female students. Future synergies with the “ladies’ community” and the alumni program will support this measure. 4.1.22 Family Friendly University It is necessary to provide a family and children friendly environment for young families or single students with children to attract more women to study or pursue a career at the university. The University of Stuttgart has already implemented a number of measures to this end. Among these are child care for the children of students (Kinderbetreuung für Kinder von Studierenden der Universität Stuttgart STUPS e. V., emergency child care, semester break child care) and the Stuttgarter Forschungsferien, an attractive child care program for school kids, organized by the Konzept-e GmbH and five different Fraunhofer institutes. Going beyond these measures, the University of Stuttgart is aiming at the creation of on-campus dorms with integrated child care for students with children and on-campus housing for young families of employees with comfortable access to nearby child care centres. Getting the students themselves to donate their time to
96
Sabina Jeschke et al.
help support the child care centres will decrease the financial burden and help with community building among female students. Additional important measures should address the specific needs created by the burden of raising children (part time studies, postponed exams due to a child falling ill, virtual courses). Researchers with children will profit mainly from dual career models, family friendly work processes, flexible time, and specific support of researchers during the first years of parenthood and the possibility of working at home. The University of Stuttgart is aiming to be the first university in Baden-Württemberg to gain the certificate as a “family friendly university” awarded by the audit council of the Hertie-Stiftung. 4.1.23 Re-Entry Program “SPIRIT returns” The knowledge and skills of many qualified female researchers and engineers is lost to the market due to the massive obstacles in returning to a job after a prolonged absence due to pregnancy and child birth. The University of Stuttgart is searching funding from the ESF based on the BMBF-Program “Wiedereinstieg für Ingenieurinnen leicht gemacht”. Engineers are given additional scientific training to re-qualify them for their re-entry into the job market. 4.1.24 Dual Career Service Program “DuCaSUS” As accepting the appointment for a professorship usually involves moving to another city, the career chances of the partner are often a decisive factor in decision making process. A similar situation exists during the post-grad and post-doc studies. Dual Career Services can advance the career chances of researcher couples. Currently, the University of Stuttgart is expanding the dual career supporting policies already common in appointment negotiations to include post-grads and post-docs in the dual career service program “DuCaSUS”. This program will involve the creation of a support network including regional businesses and public and private research companies to offer a larger basis of possible career opportunities for the partners outside the University of Stuttgart itself.
4.2 Comprehensive Measures for the whole Life-Cycle: 4.2.1 Chair for “Diversity Studies & Technology Management” The Institute of Construction, Production and Vehicle Technology will create a new chair focusing on gender and diversity, based on the example of the “Chair for Gender Studies and Information Technology” at the TU Munich. The main aim is the strengthening and localization of the gender concept of the University of Stuttgart. The academic focus of the chair will be on interdisciplinary courses in the field of gender studies, in particular for the new program “Diversity Studies and Technology Management” as well as interdisciplinary courses offered to students of other programs at the University of Stuttgart.
Spirit
97
4.2.2 Gender Components in New Appointments The University of Stuttgart is trying to appoint scientist as professors whose research focus integrates an explicit gender and diversity focus. Examples for such new appointments would be “Diversity in Usability Studies”, “Diversity in Design and Construction” (for a chair in civil engineering), “Gender Aspects in Biomedical Technology” etc.
4.2.3 Media Offensive “Fascination Technology” One of the central challenges for engineering and science is the public image. While being accepted as important for technological advances and therefore economic growth, they are also perceived as dry and boring subjects. Technologically oriented programs are often painting an unattractive image of technology as they focus on scientific theory and implementation but neglect to demonstrate social relevance. The result is a negative image of “soulless technology” as a simple means to an end, not as a field with its own aesthetic and fascination. This might be less of a problem for young men whose studies of a given field are often the result of a love affair with technology. In contrast, young women are more interested in social relevance than technology per se. The planned media offensive focuses on presenting mathematics, science and engineering as a vital and fascinating part of our modern culture. The methodology could be based on the experiences gathered in the popularization of mathematics at the MATHEON centre of research.
4.2.4 Media Offensive “SPIRIT” A program can only be as good as its general acceptance and public interest. This is one major challenge for most current gender programs and measures to overcome. Gender programs have to incorporate and implement a pervasive PR strategy. This includes extensive online presentations aimed at addressing students living in more rural areas outside of Stuttgart itself.
4.2.5 eCRM “SPIRIT” One major factor in the success of all measures described above is the continuous contact to the participating high school and university students as well as researchers. Our gender concept can only be successful if we manage to address these “customers” directly and tie them to the University of Stuttgart. To this end, an eCRM (electronic customer relationship management) will register and record all transactions between the University of Stuttgart and the participating students and researchers. As a result, they can be addressed directly and the life cycle process can be specifically tailored to each participant using intelligent agent systems (under the strictest observance of applying data privacy laws).
98
Sabina Jeschke et al.
4.2.6 Brain Storming “SPIRIT” The University of Stuttgart performs an annual brain storming on gender topics and diversity to define, select and implement further measures, with the participation of all interested institutes. The aim is a higher degree of flexibility and a stronger bound to the rest of the university.
4.3 Measures to Ensure Structural Implementation and Sustainability: To ensure the implementation and sustainability of gender concepts the following catalogue of measures are implemented: 4.3.1 Pervasive Implementation throughout the University: The single measures are placed within the jurisdiction of existing institutions (e. g. the office for equal opportunity), organizational centres (e. g. career or academic counselling), and the chairs of the institutes. Concrete tasks, their implementation and adaptation are fixed in target agreements with the president of the university. Additional measures focusing on strengthening the gender competency are the object of future appointment negotiations and target agreements. Internal projects reinforce the acceptance and broad basis for the concept. a) Collaboration with Existing Equal Opportunity Programs: Plans of equal opportunity in the academic [Thö02] and non-academic fields define an catalogue of measures aimed at achieving equal opportunity for women, based on detailed analysis of the current state. The structural and organizational preconditions for these measures already exist. b) Collaboration with the Female Professors Program: The program to increase the ratio of female professors becomes a vital part of the gender concept. Individual professors assume the operative responsibility for particular measures. These measures have already been made to increase the public visibility of the program. c) Strengthening the Scientific Gender Competency: The chair of “gender in the engineering sciences” is a component for the success of the gender concept. It strengthens the scientific background and provides the expertise required in the evaluation and long term quality control. The chair provides new input for additional projects and necessary adaptations. The competency of the chair is enhanced by the support of the office for equal opportunity and related chairs.
Spirit
99
4.3.2 Financing: Financing for the SPIRIT program is based on a three-pillar model: a) Budget: The present shift to success-oriented funding allows for funding further or extended measures. Target agreements with the institutes will initiate additional measures. Measures focusing on students directly can also be funded from tuition fees. b) Third Party Funding: Third party funding is an important component in the development and test of new measures and concepts. The EU, the ESF and the National Ministry of Science and Education as well as industrial partners have been increasingly supportive of gender and diversity projects. c) Endowment: It is planned to found a SPIRIT-Foundation to raise additional funds for the gender concept of the University of Stuttgart. This foundation will approach the industrial partners already engaged in the Femtec project, regional SMEs, alumni of the university, and private donors that recognize the need to support gender in the regional economy. The financial goal is an endowment providing a total annual funding of 500.000 C. 4.3.3 Responsibility – A priority for the President! The implementation of a pervasive and coherent gender concept is seen as a vital challenge for the future success of a technologically oriented university. The coordination, further development and quality control of the program will be the responsibility of the president of the university. The program will be managed by the president. Further developments will be coordinated with the commissioner for equal opportunity under the leadership of the president.
5 Summary The University of Stuttgart takes the female professor program of the state of BadenWürttemberg as an inducement to re-evaluate the currently used measures and proceedings in the area of equal opportunity and equal treatment. Where necessary these measures should be amended, enhanced and optimized. The major need for action does not affect the – numerous and successful – measures of the last years but does affect the gender culture at the University of Stuttgart. The measures should not only be carried by the equal opportunity department and the commissioners
100
Sabina Jeschke et al.
for equal opportunity. It is essential to further anchor the measures in research and teaching at the University of Stuttgart. At the same time patency and continuity have to be raised by developing a complete gender concept that views women and girls along a life-cycles model. It should group the different target groups and measures with regard to content and organization. Furthermore, a natural science and technology-oriented university has to take special responsibility for the participation of women and girls in the evolution of technology in our society: Women are still underrepresented in almost all natural science and technological areas. This under representation has serious consequences – for women and for society. The demand for adjustment for women is not just a demand for equality of opportunity – in a country like Germany with few resources economic success is tightly connected to advancements in technology. The future requirement for skilled personnel and managers in this field cannot be met by the current number of male graduates and the number of female graduates is nowhere close to enough to build the necessary specialized “mixed teams” which are expected to provide an important part of economic success. The University of Stuttgart sees the sustainable implementation of gender-justice as an important challenge. The realization of a profound gender concept based on the observation of the complete life cycle with measures for different target groups – schoolgirls, female students and female scientists – will become an attractor for the University of Stuttgart to survive the fight for the best minds – on national and international levels. Within the scope of the overall process the university administration coordinates the existing activities and combines and expands on them. A numerous additional innovative measures are currently being prepared and will be sustained by a financial package containing budgeting, third-party funds, and endowments. The professorships within the female professor program serve as seeds for new impulses and will develop institutional and thus sustainable responsibility for important building blocks of the master plan. The gender mission statement is being created under the leadership of the president including a road map with measures for quality control and operative realization – henceforth gender is given top priority!
References [DEJ+ 08a] N. Dahlmann, M. Elsner, S. Jeschke, N. Natho, O. Pfeiffer, and C. Schröder. Challenge Diversity: New Curricula for Natural and Computer Sciences and Engineering. 9th Nordic Research Symposium on Science Education. Reykjavik/ Iceland, June 2008. [DEJ+ 08] N. Dahlmann, M. Elsner, S. Jeschke, N. Natho, and C. Schröder. Gender Gap in Technological Disciplines: Societal Causes and Consequences. In 2008 IEEE International Symposium on Technology and Society (ISTAS 08), volume DOI: 10.1109/ISTAS.2008.4559761, Fredericton, NB, Canada„ June 2008. [DJSW07] N. Dahlmann, S. Jeschke, C. Schröder, and L. Wilke. Challenge Diversity: New Curricula in Natural Sciences, Computer Science and Engineering. In FIE 2007 The 2007 Frontiers in Education Conference (IEEE), Milwaukee/ Wisconsin/USA, October 2007.
Spirit
101
[DJTW06] N. Dahlmann, S. Jeschke, C. Thomsen, and M. Wilke. Overcoming the Gender Gap: New Concepts of Study in Technological Areas. In 2006 ASEE Annual Conference Proceedings, Chicago/USA, June 2006. [EJN+ 07] M. Elsner, S. Jeschke, N. Natho, O. Pfeiffer, and C. Schröder. Attractive Universities: New Curricula in Natural Sciences and Engineering. In Meeting the Growing Demand For Engineers and Their Educators 2010 - 2020 International Summit (IEEE), Munich/Germany, November 2007. [EJN+ 08] M. Elsner, S. Jeschke, N. Natho, O. Pfeiffer, and C. Schröder. Attractive Sciences Recruiting and Retention activities for women in academic CSET education. In 2008 ASEE annual conference, Pittsburgh/PA/USA, 2008. [EJNS08] M. Elsner, S. Jeschke, N. Natho, and C. Schröder. The Galilea Program - New Curricula for Engineering and Natural Sciences. In SEFI Annual Conference, Aalborg/ Denmark, 2008. [EK99] Europäische Kommission. Frauen und Wissenschaft - Mobilisierung der Frauen im Interesse der europäischen Forschung. Technical Report KOM(99) 76, 1999. [Kom01] Europäische Kommission. Wissenschaftspolitik in der Europäischen Union, Förderung herausragender wissenschaftlicher Leistungen durch Gender Mainstreaming. Bericht der ETAN-Expertinnengruppe “Frauen und Wissenschaft”, Brussels, 2001. [KW03] G. Koch and G. Winker. Genderforschung im geschlechterdifferenten Feld der Technik - Perspektiven für die Gewinnung von Gestaltungskompetenz. Stuttgarter Beiträge zur Medienwirtschaft, 8:31–40, 2003. [Sch03] B. Schwarze. Wer ist wirklich drin? Gender in der Informationsgesellschaft. In Analyse mehrerer Studien und darauf aufbauende Handlungsempfehlungen. 2003. [Thö02] K. Thöne. Frauenförderplan für den wissenschaftlichen Bereich. www.uni-stuttgart. de/gleichstellung/pdf/frauenfoerderplan, Last retrieved 6.11.2009, 2002. [Wäc98] C. Wächter. Frauen in der Technik - Pionierinnen in Technopatria. In C. Wächter and et. al., editors, Technik Gestalten, Interdisziplinäre Beiträge zur Technikforschung und Technologiepolitik. Kluwer Academic Publishers, München & Wien, 1998. [Win04] G. Winker. Informationstechnik und Geschlechterhierarchie - eine bewegende Beziehung. Technikfolgenabschätzung. Theorie und Praxis, 2:70–78, 2004. [ZR] M. Zwick and O. Renn. Attraktivität von technischen und ingenieurwissenschaftlichen Fächern bei der Studien- und Berufswahl jungerFrauen und Männer. Arbeitsbericht 219, Akademie für Technikfolgenabschätzung.
Going diverse in the two Clusters of Excellence “Integrative Production Technology for High-wage Countries” and “Tailor-Made Fuels from Biomass” at RWTH Aachen University Claudia Jooß, René Vossen, Anja Richert, Ingrid Isenhardt
and methods have been developed and with 50 national and international partners field-tested. After Sydow’s categorisation of networks, CoE can be interpreted as project networks (cp. Sydow 2001). To foster interdisciplinary networking among researchers of the Clusters the promotion of the tasks Scientific Cooperation, Education and Lifelong-Learning, Equal Opportunities and Diversity Management as well as Knowledge and Technology Transfer is necessary. On different levels of clusterinternal scientific cooperation (eg. such as Knowledge Organisation, Research Organisation, Communication, and Knowledge Output) networking and knowledge management measures on the field of action in this context is the fostering of Gender Strategy and Development. The CoE aim at achieving corresponding quotas of 10-30% females among its newly employed staff. The existing activities of the RWTH regarding research and measures of diversity management will therefore be extended in the context of the Clusters of Excellence. In order to attract females to engineering and natural science studies, different measures will be introduced which provide insights into topics from diverse areas of the CoE by means of age-appropriate lectures and workshops. Co operations in the Assessment Programme tasteMINT already started. tasteMINT aims at HighSchool Graduates having the chance to explore their strengths in engineering and natural science and get in contact with female students of appropriate courses of studies. Individual feedback helps them to make a more established decision on the end of the assessment centre. The existing programmes “Kinderferienzeit“, “Girls’ Day“, “Schnupperstudium für Schülerinnen“, “Do-Ing in Aachen“ will be expanded to increase the proportion of females among students and research assistants. Especially, it is planned to build up a showcase of diverse RWTH career paths of successful scientists into industry. Therefore a series of presentations with the title “I did it my way – careerpaths from RWTH Aachen University to industry” will start in October 2009. With the intention to illustrate social diversity as a factor of success the motivation of all scientists shall be increased to reach a sensitisation for the broad field of diverse career paths. In addition to that role-models and best-practices are demonstrated. During the runtime of both Clusters of Excellence the Center for Learning and Knowledge Management and Department of Information Management in Mechanical Engineering arranges periodically different workshops to match all iteratively implemets the presented measures of the toolbox. After each implementation cycle the results are reflected and thereby a model of “A Model of Use for Arrangements and Instruments of the Cross Sectional Processes in the skill-intensive Organisation Cluster of Excellence” (ASPO) is build up. The intention throughout the efficient networking and knowledge management measures applied in “Supplementary cluster activities” and “Cross Sectional Processes” is to create a model which can be transferred to other complex science-based clusters and networks.
Going diverse in the two Clusters of Excellence
Appendix
Fig. 1 Poster
105
Strategic Transfer Communication in Prevention Research as a Contribution to the Innovative and Competive Ability of Enterprises Ingo Leisten, Frank Hees
Abstract The innovation pressure as well as the huge dynamism of the working environment demand a high flexibility, autonomy and interconnection, not only of enterprise but also of employees. The meaning of the physical and mental health of employees becomes an important factor of the innovation ability, according to the concept “only a healthy employee is an innovative employee”. Indeed, a sustainable integration of occupational security and health in operational acting can only be created by an active participation of enterprises in the development of new prevention approaches. In order to make cooperation between prevention actors (especially those of the academical field) and enterprises succeed in yielding mutual benefits a fundamental change in the comprehension of the role of entrepreneurial practice in the research process is necessary: No more research FOR practice, but research together WITH practice. That means: a strategic transfer communication concept by the use of a cooperative research design is needed. This article shows a concept that picks up on impulses especially from engineering to develop strategic transfer communication in prevention research as a contribution to innovative and competitive ability of enterprises. Keywords Knowledge Management · Research on Innovational Aspects of Society · Experiences in University-Industry Cooperation
1 Introduction Facing the challenges of the ongoing developments in global economics, the constantly increasing globalization, the increasing complexity and dynamisation of (business) processes [HL08], once more the question arises, how well enterprises
cope with constant competition? The German occupational scientist Volkholz [Vol07] answers this question: by being unique. Applied to Occupational Safety and Health that means: it “shall and has to be measured by the question if (it) contributes to the enterprise’s development and ability to undergo changes” [Vol07]. In times in which innovations, corporate responsibility and sustainable solutions are decisive factors in macroeconomic competition the interface between science and entrepreneurial practice is of increasing importance. The transfer of innovative science from research into entrepreneurial practice (a transfer suiting target groups and delivering results quickly) is becoming more and more the focus of scientific considerations [Lud07]. Numerous research disciplines are dealing with the developments of innovative as well as long-lasting solutions and procedures for business practice [AW05]. Enterprises interested in the latest results of science and research to augment their competitive abilities are also numerous. Thereby the current promotional focus of the German Federal Ministry of Education and Research (BMBF) “Preventive Occupational Health and Safety” offers an interesting field of application for the design of transfer since prevention is considered a central element of Human Resources and thus as a roadmark for innovation and sustainability of enterprises which currently has often not been put into entrepreneurial practice yet. On the one hand, research results on Preventive Occupational Health and Safety must not only be successfully transferred into enterprises but also have to be sustainably integrated there. Inversely, many research questions have to be derived from practice. However, this exchange process between scientists and enterprises does not always succeed. This is also the case with prevention research, aggravated by the fact that the needs of Occupational Health and Safety often are not – or insufficiently – reconciled with the primary contentual and financial goals of the enterprises [HL08]. Isolated research can not increase competitiveness, only a cooperation of research and enterprises can expand the potential of occupational research. The target groups of the current projects in Preventive Occupational Health and Safety are as diverse as the projects: they range from the creative sector to health care, crafts and trade to tax officers. Global Players as well as microentrepreneurs are integrated withtin the field of research. Labour unions, professional and industrial associations as well as chambers of industry and health insurances are involved. The challenge is: complex personnel and institutional structures need a systematic and strategic transfer communication.
2 Strategic Transfer within Occupational Safety and Health Within the domain of occupational research the objective “funding of competitiveness and innovation skills” is not a new one. However, the transfer problem is not completely solved within the new BMBF funding programme “Working, Learning, Developing Skills. Potential for Innovation in a Modern Working Environment”. This was emphasized by statements made in a survey conducted by Project StArG [sta], asking researchers dealing with the research focus “Preventive Occupational
Strategic Transfer Communication in Prevention Research
109
Safety and Health” about the transfer of their research results [HLR07]. Two different types of reasons have been differentiated: On the one hand it is possible that the target systems of Occupational Safety and Health fundamentally contrast with those of the enterprises. On the other hand it is possible that the enterprises have different research problems, yet they are not able to phrase them precisely. To identify these is the necessary condition for a successful research process in the sense of a sustainable change within an enterprise in the field of Preventive Occupational Safety and Health. Within the context of a constant increase of national and international research funding, researchers increasingly focus on problems concerning the successful transfer of research knowledge into practice, helping to enhance the effectiveness and efficiency of investments made by research sponsors. But not only the transfer into the entrepreneurial practice within the framework of research and innovation transfer has to be considered. Actors of different recursion levels like individual assistants or the employees of the enterprise, lobbies and networks, politics and society have to be integrated into the transfer process, in a way that suits the relevant target groups. Especially within the domain of occupational research such a complex personnel and institutional structure can be determined. But it is already a success if the transfer processes can be approached on different levels [HLBH09]: The direct, operative transfer of research results into entrepreneurial practice (Interactive, Deep Transfer): the transfer is part of the research process meaning it is a test field for research results and the resulting research questions, which again can be of strategic importance. The strategic transfer into the general public (Public Transfer): Because they have a normative effect and can therefore in the long run enforce new international standards, it is important to publish the results of research projects and catch the attention of the general public. If one applies these assumptions to transfer communication within Preventive Occupational Safety and Health, the combination of Interactive, Deep Transfer and Public Transfer can help to optimise difficulties concerning transfer: the explicit forms of knowledge are therefore with the help of Public Transfer publicly presented and communicated, whereas for the development of implicit knowledge different methods of the Interactive, Deep transfer seem to be more suitable. Intermediary, politics and science as well as enterprises, which are not directly integrated into the research, are the main target groups. These target groups are to be reached by applying Public Transfer. By applying Interactive, Deep transfer in contrast to Public Transfer a great amount of implicit knowledge can be imparted. The objective here is not primarily to pass on information, but by using the transfer information to generate relevant knowledge for the enterprise, allowing the enterprise developing specific Know-How. After this it has to be considered whether the different involved want to apply this Know-How. Only if they do, the specific solution makes sense. For the field of Preventive Occupational Safety and Health this means to promote the competitive and innovative capacity as well as the unique integration of prevention within the enterprise.
110
Ingo Leisten et al.
3 Interactive, Deep Transfer Through the Integration of Entrepreneurial Practice into the Research Process Within the last years an obvious increase concerning the speed of innovation within all innovative industries has be observed [Koh08]. Furthermore it becomes clear that the innovation process is changing from a traditional push approach to a demand and lead market orientated process, with the innovation activities increasingly oriented downstream. For rapid innovations an early interaction with the customer and a close connection to the fields of innovation is of importance [Ger04]. A promising transfer strategy is the exchange between enterprises and science before the actual research starts and to agree on a common research focus in advance. Especially within the early stage of an innovation process there is often a tendency to act unstructured and dynamic [HV03]. This shows the necessity of a close and interactive cooperation between research and practice prior to the start of a research project. It allows to compare research gaps with the actual need for research results and both researchers as well as entrepreneurs contribute their ideas to the project. Arrangements concerning Preventive Occupational Safety and Health can only become an integral part of an enterprise if there is an active cooperation, for both transfer partners successes become apparent and the business culture is further developed in terms of Preventive Occupational Safety and Health. How can this be achieved? Every research project and enterprise is acting to fulfil its specific purpose [Han88]. The purpose displays the vital exchanges between a project or an enterprise and its environment, i.a. each project or enterprise needs to adjust its actions to the need of the respective client [Rie97]. Only if they can fulfil the needs of their clients the purpose is fulfilled and the project or rather the enterprise is acting successfully [HM00]. Before an active cooperation between enterprises and researchers can start, these purpose and therefore the objectives of the cooperation, have to be determined. The purpose of research and the purpose of enterprises are of an essential different kind because the projects concerning Occupational Safety and Health do not hit the enterprises’ primary objectives. Concerning subject matters, different interest may collide – regarding to the process and the result of value performance. This gap has to be reduced and common interfaces have to be found. Within these interfaces subgoals of occupational safety and health should help to strengthen the core objectives of the enterprises so that the needs of the stakeholders of Preventive Occupational Safety and Health and those of the clients do better consist. Because of the different conceptions and expectations a common focus of cooperation and interaction has to be determined. Research and entrepreneurial practice should coordinate their actions and secure that they work on a common research focus [Dun04]. This is the fundament for a cooperative, interactive research process in which the enterprises play the role of co-producers. This means a changed role: from a passive object of observation to an independent innovator. The collaboration starts when research and enterprises identify the common interfaces of the particular purpose and determine these interfaces as one common purpose of the coorparation. As in this interface it is recorded which clients and needs the reasearch project and enterprise can satisfy
Strategic Transfer Communication in Prevention Research
111
together, this interface is the basis of the cooperation. Therefore it is important to focus on this interface, which can be extended during the project concerning Preventive Occupational Safety and Health. Only if Preventive Occupational Safety and Health is deliberately established in this way it can be viewed as an integral part of an enterprise. It has a sustainable impact because it is of direct importance for the success of the enterprise.
4 Successful Public Transfer to the Stakeholders of Preventive Occupational Safety and Health Public Transfer is in view of the fact an important factor that it places the stakeholders of Preventive Occupational Safety and Health in the centre of transfer. The aim and purpose of Public Transfer is to guide the public interest and point out the necessity to integrate Preventive Occupational Safety and Health into business practice. According to Freeman [Fre84] the term stakeholder means “any group or individual who can affect or is affected by the achievement of a corporation’s purpose”. The influence of the individual stakeholders is strongly dependent of the general conditions of the enterprise/project and of the interaction between the individual stakeholders [Bal99]. Stakeholders of Preventive Occupational Safety and Health are for example branch-specific or non specific organizations and associations, public and political institutions, health insurances or trade unions. All stakeholders have different requirements which the enterprises and their measures for Preventive Occupational Safety and Health have to satisfy. For the formation of a target-oriented Public Transfer the individual stakeholders, who are of relevance for the enterprises in this project, have to be identified. Especially intermediaries have to be integrated into the transfer process to use their potentials for a long-term integration of preventive thinking. As mentioned before, contradictions between more on the business purpose orientated requirements (like those of the sales and capital market) and more value-oriented requirements (for example those of different lobbies) are predictable. Every enterprise itself has to evaluate if this is of limited, mid-level or of a long term importance. Concerning this evaluation the orientation and vehemence of the requirements of the stakeholders is of great importance. Successful public transfer within Preventive Occupational Safety and Health responds to the stakeholders at that point. This may support expanding a lobby of Preventive Occupational Safety and Health influencing entrepreneurial practice.
5 Transfer Engineering as a Concept for the Strategic Design of Transfer Communication The starting point for the analysis of the chosen field of research within Preventive Occupational Safety and Health is the observation that the transfer of knowledge from research into practice works better in some projects than in others. Hence the
112
Ingo Leisten et al.
question arises how the transfer activities of the current BMBF projects dealing with a certain research focus differ from each other and which factors of success can be deduced. Because of the already described importance of communication and information within the innovation process the focus of the analysis concerns the extend of integration of business practice into the research process. Thereby it is assumed that information pathologies are an essential reason for the failed transfer of research results into practice. Following Scholl [Sch04] information pathologies are information relevant for decisions, which are (although they are in principle accessible) not correctly produced, passed on, gathered or processed. Following Bruhn [Bru03], the proximity to customers becomes apparent both within the customer orientation of service offers (meaning through the usability of research results) and within the customer orientation in the interactive behaviour. The projects concerning the current BMBF research focus “Preventive Occupational Safety and Health” are analysed and systematises following Kunz and Mangold’s [KM03] segmentation model for customer integration. This analysis focuses the role of the entrepreneurial practice within the research process as well as the integration degree. This classification follows various empiric data, based on surveys and interviews as well as on a qualitative content analysis of documents and media of the respective projects. The first step towards customer orientation, which has already been realized by many research projects, is to collect information, especially information concerning current and future needs within business practice and corresponding concepts for project work [DB09]. A few well-known and widespread approaches for the integration of customers are mainly dealing with the first stage of the innovation process. Therefore interviews with customers at the beginning of a research project are common practice. Thereby business practice takes up the role of a passive observation object. Traditionally and enforced by the political promotion transfer is an essential task in the last stage of research projects. Within this stage the results and solutions will be presented to business practice. The role of business practice can be characterised depending on the applied type of media as a heteronomous (push media) or self-determined (pull media) dialogue partner. The analysis shows as well, that some projects view business practice as an equal interaction partner: for example with the help of regular workshops taking place at different stages of project work, research results are compared with the requirements of business practice and therefore the research design can be changed where it is necessary. The analysis, however, shows that some projects go beyond the described basic approaches: they realise a cooperative, interactive research process for the purpose of customer integration, so that the innovation activities are marked by common system and problem solving capacities between enterprises and customers [BDMS09]. Entrepreneurial practice is seen analogue to the understanding of Open Innovation within the branch of production and service industry “not only as a source of information concerning the customer’s requirements, but also as a source of information about possible solutions.” [RP05] The more intensive thereby the interaction and the need for cooperation with the customer are (in this case business practice within research processes) the better are the conditions for the
Strategic Transfer Communication in Prevention Research
113
acceptance of new perspectives and information [BDMS09] -this applies within research as well as entrepreneurial practice. Approaches for the design as well as for a solution to the described tasks (concerning the transfer of research results) can be found within the fields of linguistics, communication science, psychology, pedagogic as well as business studies. Especially within the field of engineering trend-setting ideas can be found, which have not been sufficiently considered within the scientific debate. Especially within engineering science impressive examples for a successful transfer from research into practice can be found. In these examples the addressee of transfer communication has no longer the role of a research consumer but of a research producer. Engineering offers concepts for the development of solution strategies by use of proven methods for example from the field of Service Engineering. The concept of Service Engineering supports enterprises “to organise services in a way that they can be offered to the market in a desired quality and efficiency” [BS03], by constantly including the customers and their respective needs into the provision of services. Thereby the researcher’s motivation is combined with the motivation of the addressee, whereby the addressee not only becomes part of the transfer process, but rather becomes a co-producer of research results. Within the framework of interaction between the transfer partners the potentials from research and (entrepreneurial) practice are used for the realisation of effects following research transfer. For the integration of the addressee into the research process -and therefore for a successful research transfer -a concept concerning Transfer Engineering is designed, describing a systematic approach for the design of transfer communication. By understanding the addressee as an expert and therefore as a constitutive part of the research process, Transfer Engineering enables research to work consequently for the needs of business practice. With the help of research designs from the field of engineering, approaches of research transfer which have already realised a successful integration of the addressee can be demonstrated. Examples can be found amongst others within the Agile Software Development, in models for process synchronisation within the automobile industry (e.g. Methology for Engineering Process Synchronisation) or realising Open Innovation product and service engineering.
6 Perspective Within fields dealing with a complex personnel structure, as in Preventive Occupational Safety and Health, an extensive participation of the addressee within the research process seldom takes place. Projects concerning the research focus “Preventive Occupational Safety and Health” are therefore checked with regard to the kind and extent of integration with which the addressee is integrated into the research process as well as checked for successful instruments for the realisation of this integration. Based on the classification of these methods into the concept of Transfer Engineering a procedure is shown that transfer communication can be successfully shaped by the consequent integration of the addressee into the research process. In Summary: due to the present results of reflection, two steps are needed
114
Ingo Leisten et al.
for the improvement of the transfer of research projects and therefore for a successful cooperation between research and enterprises: 1. For the improvement of Interactive/Deep Transfer and therefore for the promotion of competitiveness and innovation skills of enterprises through Preventive Occupational Safety and Health, business practice has to be integrated into the research process. 2. With the help of Public Transfer intermediary, for social discourse relevant stakeholders and lobbies have to be approached in addition to business practice and to be sensitised for the concerns of Preventive Occupational Safety and Health. Prevention research needs the active participation of enterprises to produce successful research results. This participation should already start with the foundation of the research projects so that they can respond to the requirements of entrepreneurial practice and approach problems. Thereby transfer partnerships are formed, in which research and practice cooperate as experts to increase the research success or rather to achieve an increase of innovation skills and competitiveness. For this purpose, the suggested approach of Transfer Engineering offers methods used within the field of engineering. To integrate a strategic transfer of research results and innovative knowledge into business practice the standard “research for practice” has to be changed in “research with practice”.
References [AW05]
G. Antos and S. Wichter. Wissenstransfer durch Sprache als gesellschaftliches Problem. Frankfurt a. M, 2005. [Bal99] B. Balkenhol. Ein unternehmenskybernetischer Ansatz zu Integration von Umweltmanagementsystemen. Aachen, 1999. [BDMS09] E. Bamberg, J. Dettmers, C. Marggraf-Micheel, and S. Stremming. Innovationen in Organisationen – der Kunde als König? Huber, Bern, 2009. [Bru03] Bruhn. Kundenorientierung. Bausteine für ein exzellentes Customer Relationship Management. DTV-Beck, München, 2003. [BS03] H.J. Bullinger and A.-W. Scheer. Service Engineering – Entwicklung und Gestaltung innovativer Dienstleistungen. Springer, Berlin, 2003. [DB09] J. Dettmers and E. Bamberg. Innovationen durch Kundenorientierung? In E. Bamberg, editor, Innovationen in Organisationen – der Kunde als König? Huber, Bern, 2009. [Dun04] W. Dunkel. Arbeit am Kunden. Herausforderungen und Zukunftschancen für das personenbezogene Handwerk. In R. Kreibich and B. Oertel, editors, Erfolg mit Dienstleistungen. Innovationen, Märkte, Kunden, Arbeit. Schaeffer-Poeschel, Stuttgart, 2004. [Fre84] R. Freeman. Strategic Management. A Stakeholder Approach. Marshfield, 1984. [Ger04] A. Gerybadze. Knowledge Management, Cognitve Coherence and Equivocality in Distributed Innovation Processes in MNCs. Management International Review, 44(Special Issue 3):103–128, 2004. [Han88] D. P. Hanna. Designing Organizations for High Performance. Reading, Mass, 1988. [HL08] K. Henning and I. Leisten. Lernen und Arbeiten für Innovationen: Lust auf Zukunft – zwölf Thesen. In D. Streich and D. Abel, editors, Innovationsfähigkeit in einer
Strategic Transfer Communication in Prevention Research
115
modernen Arbeitswelt, Personalentwicklung, Organisationsentwicklung, Kompetenzentwicklung. Frankfurt am Main, 2008. [HLBH09] K. Henning, I. Leisten, U. Bach, and F. Hees. Präventionsforschung und unternehmerische Praxis: Zwei Seiten einer Medaille. In K. Henning, I. Leisten, and F. Hees, editors, Innovationsfähigkeit stärken -Wettbewerbsfähigkeit erhalten, ARMT, pages 12–30. Mainz Verlag, Aachen, 2009. [HLR07] F. Hees, I. Leisten, and A. Richert. Unveröffentlichter Akzeptanzbericht des Metaprojektes StArG. Aachen, 2007. [HM00] K. Henning and S. Marks. Kommunikations-und Organisationsentwicklung. Aachen, 6 edition, 2000. [HV03] C. Herstatt and B Verworn. Bedeutung und Charakteristika der frühen Phasen des Innovationsprozesses. In Management der frühen Innovationsphasen: Grundlagen Methoden. Neue Ansätze. Gabler, Wiesbaden, 2003. [KM03] W. Kunz and M. Mangold. Segmentierungsmodell für die Kundenintegration in Dienstleistungs-Innovationsprozesse. Eine Anreiz-Beitrags-theoretische Analyse. www.Win-serv.de, 2003. [Koh08] J. Kohler. Wissenstransfer bei hoher Produkt-und Prozesskomplexität. Gabler, Wiesbaden, 2008. [Lud07] J. See Ludwig. Wissenschaftstransfer, Wissenstransfer und neue Veränderungskulturen. In J. Ludwig and M. Moldaschl, editors, Arbeitsforschung und Innovationsfähigkeit in Deutschland, pages 238–247. München, 2007. [Rie97] H. Riekmann. Managen und Führen am Rande des 3. Jahrtausends: Praktisches, Theoretisches, Bedenkliches. Frankfurt a.M., 1997. [RP05] R. Reichswald and F. Piller. Open Innovation. Kunden als Partner im Innovationsprozess. www.Impulse.de/downloads/open_innovation.pdf, 2005. [Sch04] W. Scholl. Innovation und Information. Wie in Unternehmen neues Wissen produziert wird. Hogrefe, Göttingen, 2004. [sta] StArG. www.starg-online.de. for more information. [Vol07] V. Volkholz. Capability for Innovation. In J. Ludwig and M. Moldaschl, editors, Arbeitsforschung und Innovationsfähigkeit in Deutschland, pages 41–49. München, 2007.
A Methodology to Reduce Technical Risk in the Development of Telematic Rescue Assistance Systems Matthias Müller, Michael Protogerakis, Klaus Henning
Abstract In Germany demand for Emergency Medical Services (EMS) physicians is outstripping supply. Telematic Rescue Assistance Systems (TRAS) offer the opportunity to use EMS physicians more efficiently by reducing the on-scene time. By transmitting audio, vital signs and video data telematically, they bring the expertise of a remote elder EMS physician, a hospital or otherwise specialized institution to the emergency site. However, with increasing complexity of the systems the technical risks become harder to manage. A more formal approach to reduce these risks is needed to ensure patient safety. In this paper some exemplary properties that are more or less inherent in all TRAS are presented. Based on these features and existing work, a methodology to reduce technical risks in developing TRAS is introduced. Keywords Telemedicine · Emergency Medical Services · Risk Management · Telecare
threatening situation. Even when his presence is indicated e.g. for decision making, his manual abilities are needed in less then 15% of all missions [GHM03]. Provided he received all relevant information, diagnosis and treatment could be decided on from another location than the emergency site. Even in situations when time to definite treatment plays a crucial role like in heart attack or stroke [GdLA04]. TRAS deliver patient information instantly to any relevant institution. Shortening the time between such an emergency incident and definite treatment improves patient outcome. Besides, the EMS physician does not necessarily lose time shuttling between ambulance station, emergency site and hospital in cases where his physical presence is actually not needed on scene. In Aachen, Germany, researchers from the ZLW/IMA at RWTH Aachen University and the University Hospital Aachen along with the engineers from P3 communications and Philips Healthcare are developing a TRAS in the Med-on-@ix (medonaix.de) project. Their aim is to give an EMS physician at a remote site all the data needed to make timely informed decisions on diagnosis and treatment [PGH09]. Med-on-@ix is funded by the Federal Ministry of Economics and Technology and the industrial partners (www.simobit.de). TRAS are different in their approaches and their details, yet they can be described in general terms.
2 Characterization of Telematic Rescue Assistance Systems Different TRAS follow a similar structure. TRAS link an EMS physician with the emergency site. An emergency site is the spot where patient data are gathered. This can be the patient location at his home or elsewhere as well as the ambulance. For simplicity, the place from which the EMS physician or another specialized institution provides the EMS team with expertise will be referred to as the Competence Center. It can be a hospital, a fire department or relevant institution. Different TRAS have been developed. StrokeNet, for instance, transmits audio and video data in order to diagnose stroke patients directly from the ambulance, so that patients can be routed to stroke units directly when necessary (www.strokenet.de). The aim of Med-on-@ix is to transfer various vital signs via audio, photo, video on a portable solution directly from the patients’ location, where they – in Germany – are usually stabilized prior to transport. Ortivus has developed a telemonitoring solution for ambulances, which transmits vital signs and audio from a portable solution to a physician in a hospital (www.ortivus.se). All TRAS mentioned above are used in preclinical environments. Critically ill patients are stabilized for transport. At the facility of definite treatment, e.g. the emergency room, catheter laboratory or stroke unit, the patient is handed over. The less time this handover postpones definite treatment the better. Providing the admitting facility with data on the incoming patient early on shortens the critical interval to appropriate treatment by allowing time for preparation [Sea08]. Also, some time-consuming double examinations of the emergency
A Methodology to Reduce Technical Risk
119
patient can be avoided, provided the data are made available before the patient arrives. Data in TRAS are gathered electronically and also are automatically checked for plausibility. Electronic records of EMS missions are easier to data mine, facilitating an assessment of the quality of patient care. Also they may be exported to Hospital Information Systems (HIS). Because hospitals guard their HIS from outside access due to security reasons, neutral open data formats such as Health Level Seven (HL-7) may be used to provide HIS with TRAS data. TRAS must fulfill a number of requirements. At the emergency site vital parameters, audio and possibly video data must be acquired. An interface for entering patient data and documenting the mission must be provided. All patient data must be displayed to the supervising or physician responsible. Data from the emergency site and treatment directions from the EMS physician must be transmitted wirelessly, because there are no landlines to be relied on. The data must be archived for further evaluation as well as for legal reasons. The privacy of both the patient and the EMS personnel must be protected. Equipment used at the emergency site must be portable and durable. TRAS are technically located at the emergency site and the Competence Center. At the emergency site data have to be acquired, processed and communicated. A mobile computer that can be taken to the patient location is used to enter data and to control functions at the site. Medical devices supply a variety of information such electrocardiograms and pulsoximetry. An audio device is used for the communication with the Competence Center. Some TRAS allow the capturing of video data. At the Competence Center patient data are displayed to the physician responsible on one or more monitors. An audio device allows voice communication with the emergency site. Additional external information sources such as databases regarding treatment of poisoning may provide the EMS team with additional outside expertise. Support functions such as archiving and maintenance are needed.
3 A Methodology for Reducing Technical Risk in the Development of Telematic Rescue Assistance Systems Demand for TRAS is likely to grow as Patients’ lives depend on TRAS. Therefore technical risks to the proper functioning of TRAS have to be reduced. Currently each system that is newly developed or enhanced is designing its own approach to reducing technical risks. Several consortia are developing their own methodology to reduce technical risk, although their systems are similar. A common methodology could save resources spent on deciding on how to reduce risk, thus decreasing development costs leaving more resources for the actual identification, evaluation and mitigation of technical problems. Methodologies for reducing these risks have stood the test and been shown to be of value in other industrial sectors [Bie03]. TRAS are very complex systems, consisting of numerous geographically distributed components. Their details can be overwhelming, if not examined with a
120
Matthias Müller et al.
Fig. 1 Technical Views
focused perspective. Technical views on the system include hardware, software and IT security (Fig. 1). Both hardware and software are examined with their established methods. When assessing IT security risks, all relevant hardware and software is looked at. Non-engineering views such as legal considerations may also be needed, but are not the focus of this paper. A methodology to reduce technical risk of TRAS has to strike the balance between being too broad – fitting all systems, yet being short on the details – and being too narrow – specifying the details, but just fitting a few systems. Both hardware and software are examined during the concept phase, when requirements and the design of the system are defined, and during the implementation phase. Design is easier to amend during the concept phase. Therefore the system is examined in this phase. When the system is implemented, the individual components are tested for compliance to the requirements. If they comply, they are integrated and tested again. Some components like medical devices may have already been thoroughly examined by the manufacturer. They have to be included in the risk management of the system, albeit their subcomponents may not need to be examined again. In contrast to software, hardware wears out over time due to material stress. Therefore system hardware has to be replaced sooner (e.g. batteries) or later (e.g. displays). Hardware defects are usually easier to identify than software defects and limited to fewer components. Hardware risks are identified early on in the concept phase. Using the characteristics of TRAS appropriate methods are chosen. Instructions are given on how to combine them in the examination of a TRAS. Deductive methods like Fault Tree Analysis [Com06b] examine the possible causes of a defined event (e.g. a system breakdown). Inductive methods like Failure Modes and Effects Analysis [Com06a] identify the consequences of a component failing. Both inductive and deductive methods should in theory identify the greatest risks for the system. Since no method by itself is perfect they are combined to complement each other, one identifying risks that the other did not and vice versa. The widely accepted IEC 60300-3-1 already describes a number of methods [Com03], with their abilities. For instance, a Fault Tree Analysis can handle combinations of faults, whereas a Failure Modes and Effects Analysis cannot. During the implementation phase the components developed are tested to the requirements defined. They are integrated into subsystems and tested again. Software does not consist of anything physical, so it is not subject to material tiring. Often its components are interdependent. One wrong calculation in one module
A Methodology to Reduce Technical Risk
121
may cause another one to fail. The impact of a software error is hard to fathom. It may range from harmless to system breakdown. The methodology for reducing software risks starts out with a requirement analysis for orientation. For this the analyses of documents as well as interviews are used. Heide [Hei04] recommends conducting a Fault Tree Analysis to identify major risks. At the end of the concept phase design is reviewed exhaustively prior to implementation, which corrects design errors early on. During implementation code is reviewed. Once components are finished, they are tested using a combination of function-oriented tests (e.g. black-box tests), structure-oriented tests (e.g. white-box tests) and fault-oriented tests. Units are tested and integrated successively. IT-Security is highly relevant to TRAS and certain problems distinguish the field from others. EMS personnel at the site and the EMS physician in the Competence Center rely on accurate and timely data for treating an emergency patient. Patient data are highly confidential and have to be specially protected. Since portable devices may get stolen more easily, special protection is needed to secure the data on them. Failing to secure information properly can have serious legal consequences to the users of the system as well as to the developer. Common security risks of healthcare IT are given in ISO 27799 [fS08] and are adapted to fit EMS. Certain measures of IT security have to be built in when developing the system. Here the methodology will use the BSI-100ff [fI05] catalogues which give prescriptions for existing IT systems on how to achieve a standard level of security. Guidelines for implementing an information security process may be given.
4 Conclusion A model was created that fits a common TRAS. TRAS as a group share common properties. A methodology was presented which takes into account the characteristics of existing TRAS as well as previous work to build upon. As it has been shown that such a methodology is both necessary and feasible, the methods used should be further adapted to TRAS. Also further views may have to be developed. This paper has focused on hardware, software and IT security, but non-engineering views such as medical, legal and organizational considerations as well as acceptance by patients and personnel may be required. The management of technical risks of TRAS should become more standardized and more affordable. The methodology presented in this paper is a step in this direction. Acknowledgement The depicted research has been funded by the German Federal Ministry of Economics and Technology, P3 Communications GmbH, Philips Healthcare and the Fire Department of the city of Aachen.
122
Matthias Müller et al.
References [Bie03]
U. Biegert. Ganzheitliche modellbasierte Sicherheitsanalyse von Prozessautomatisierungssystemen. PhD thesis, Stuttgart University, Stuttgart, Germany, 2003. IAS. [Com03] International Electrotechnical Commission. IEC 60300-3-1: Dependability management - Part 3-1: Application guide - Analysis techniques for dependability - Guide on methodology. Technical report, 2003. [Com06a] International Electrotechnical Commission. IEC 60812:Analysis techniques for system reliability - Procedure for failure mode and effects analysis (FMEA). Technical report, 2006. [Com06b] International Electrotechnical Commission. IEC 61025: Fault tree analysis (FTA). Technical report, 2006. [fI05] Bundesamt für Informationssicherheit. BSI-Standard 100-1, Version 1.0. Technical report, 2005. [fS08] International Organization for Standardization. ISO 27799: Health informatics - Information security management in health using ISO/IEC 27002, January 2008. [GdLA04] C. Gibson, J.A. de Lemos, and E.M. Antman. Time is muscle in primary PCI: the strength of the evidence grows. Eur Heart J, pages 1001–1002, June 2004. [GHM03] A. Gries, M. Helm, and E. Martin. The future of preclinical emergency medicine in Germany. Der Anaesthesist, 52:718, 2003. [Hei04] A. Heide. Ursachenanalyse und Bewertung der Verantwortung bei Funktionsstörungen von softwaregesteuerten Komponenten im Maschinenbau. PhD thesis, ZLW/IMA, RWTH Aachen University, Aachen, Germany, 2004. [Mar08] W. Martin. Arbeitsmarkt für Ärztinnen und Ärzte: Der Ärztemangel nimmt weiter zu. Deutsches Ärzteblatt, 105:853–854, 2008. [PGH09] M. Protogerakis, A. Gramatke, and K. Henning. A System Architecture for a Telematic Support System in Emergency Medical Services. In 3rd International Conference on Bioinformatics and Biomedical Engineering, Beijing, 2009. [SBB04] R. Schmiede, H. Behrendt, and E. Betzler. Bedarfsplanung im Rettungsdienst: Standorte - Fahrzeuge - Personal - Kosten. Springer, Berlin, 2004. [Sea08] K.H. Scholz and et al. Optimizing systems of care for patients with acute myocardial infarction. STEMI networks, telemetry ECG, and standardized quality improvement with systematic data feedback. Herz, 33:102–109, March 2008. [Sea09] M. Skorning and et al. E-health in emergency medicine - the research project Med– on–@ix. Der Anaesthesist, pages 285–292, March 2009.
Defining a universal actor content-element model for exploring social and information networks considering the temporal dynamic Claudia Müller, Benedikt Meuthrath, Sabina Jeschke
Abstract The emergence of the Social Web offers new opportunities for scientists to explore open virtual communities. Various approaches have appeared in terms of statistical evaluation, descriptive studies and network analyses, which pursue an enhanced understanding of existing mechanisms developing from the interplay of technical and social infrastructures. Unfortunately, at the moment, all these approaches are separate and no integrated approach exists. This gap is filled by our proposal of a concept which is composed of a universal description model, temporal network definitions, and a measurement system. The approach addresses the necessary interpretation of Social Web communities as dynamic systems. In addition to the explicated models, a software tool is briefly introduced employing the specified models. Furthermore, a scenario is used where an extract from the Wikipedia database shows the practical application of the software. Keywords Social Web · Wikipedia · Social networking
1 Introduction The Social Web is composed of various applications, such as Flickr, twitter, del.icio.us, personal Weblogs and Wikipedia. The success or failure of these Web applications is heavily based on their social infrastructure (e.g. rules, policies, organization structure, etc.) [HSH+ 08]. Consequently, an understanding of the dynamics and evolution of social and informational structures, which emerge in such virtual spaces, is essential. Wikipedia, as a representative of Social Web applications, is utilized to introduce an actor content-element model. This model allows the description and analysis of C. Müller (B) Center of Information Technologies, University of Stuttgart, Allmandring 30A, 70569 Stuttgart, Germany e-mail: [email protected]
Social Web applications more precisely in terms of existing contextual and temporal interdependencies. However, there is a broad range of studies already dealing with the analysis of Wikipedia. These studies often answer questions in terms of usage patterns, the overall link structure, and relevant information in Wikipedia’s articles, and they can be roughly divided into explanatory data-based analyses, aggregated (i.e. graph-based) approaches, and visualization techniques. Selected examples are introduced in the following paragraph. Explanatory data-based analyses are carried out to investigate author contributions in terms of article quality [AdAPR08], to estimate the importance of privileged users for content creation [KCP+ 07], or to evaluate the information quality in Wikipedia [STSG05]. In aggregated approaches, Wikipedia is interpreted as Wikigraph, where the temporal development [BCD+ 06], the inherent characteristic of preferential attachment [CSC+ 06], and the complex nature [ZBSD06] are investigated. Moreover, hidden social patterns are revealed by applying social network analysis on so-called revert graphs [SCPK07] and tripartite relations consisting of categories, users and articles are defined [NT08]. Visualization techniques are carried out to substantiate specific aspects, for example the editing activity of Wikipedia is visualized by Chromograms [WVH07], changes on talk pages are illustrated by history flow diagrams [VWKvH07], and co-authorship networks show author-based relationships between articles [BA06]. Despite these various approaches, an approach that integrates these different perspectives is still missing. This gap is filled by the proposed actor content-element model which is composed of different models and a software employing these models. The contribution presented here is organized as follows: firstly, the most important concepts in Wikipedia are briefly explained. Secondly, the universal analysis model is specified. Using this model, two temporal network types are defined. Thirdly, a measurement model, considering different classes, is specified. Existing capabilities of applying the measurement model and the associated software is emphasized in a scenario. Here, different analysis approaches are presented and discussed. Finally, limitations are stated and future challenges are described.
2 Theoretical Foundation Volunteers around the world, who write different language versions, have established Wikipedia. The process of creating and sharing information in such a virtual information space can be explained by a simplified model: a person, let us call her Alice, writes an article by externalizing her knowledge as information. This information is stored as data in a database (i.e. a text file) which splits the data into different tables and uncouples them from the existing context. Another person, let us call him Bob, reads this article and internalizes the information contained. Bob might be able to extend the acquired information. Again, Bob’s knowledge is externalized in the article and then Alice can internalize this information. Between these two people, a
Defining a universal actor content-element model
125
timely shifted communication process takes place because the output of one person can be used as input by the other and vice versa. A similar process takes place in other information spaces such as Weblogs or Folksonomies. Only the characteristics of knowledge exchange and the name of the content objects differ slightly. All mentioned entities can be assigned to different dimensions which as a whole, constitute the social information space [Fuc04]. Actors (in the example Alice and Bob) belong to the individual dimension. The timely shifted communication between actors describes the interactional dimension. The article or content level is the integrating dimension in which the cooperation takes place. The main entities, actor and content elements, should be retrieved from available application data. The open source software MediaWiki is used by Wikipedia and all data is stored in a SQL database. The MediaWiki database scheme [Fou09] has been analyzed, the most important concepts identified, and modeled in Figure 1. A user can be an author of a page or just a reader. An author is a user who has created a version of an article. Each version belongs uniquely to one page. Pages are assigned to a specific namespace, which allows the separation of pages depending on their function and avoids page-naming conflicts. Wikipedia has 18 basic and two custom namespaces. Depending on the namespace, a page is interpreted differently. Selected page types are shown in Figure 1. These types are considered further: articles (ns:0 = dynamic content), files (ns:6 = static content), categories (ns:14 = structural elements) and templates (ns:10 = layout elements). There are different relationships between page types: links, inclusions and assignments. So-called Wikilinks ([[ar ticle_name]]) are established between articles. Furthermore, an article can utilize templates which are included by template links ([[T emplate:N ame]]), can tag categories by using category links ([[Categor y:N ame]]), and link to media files (e.g. PNG, PDF, etc.) by using media links ([[Media:N ame]]).
Fig. 1 UML class diagram showing the used concepts and their relations in Wikipedia
126
Claudia Müller et al.
2.1 Universal Actor Content-element Model In the subsequent section these MediaWiki-based entities are transferred to a universal actor content-element model1 that can be applied to different types of information spaces. In the universal actor content-element model, the main concepts consist of two types of vertices and three different types of relationships. Figure 2 shows the UML (Unified Modeling Language) definition of such a model. What an information space is made of is defined in the core. It consists of actors and content elements. Actors and content elements form bidirectional knowledge relationships: every content element in the information space has at least
WikiLink
Contextual Relationship
*
*
Content Element
*
*
Node
InfoSpace
Knowledge
*
*
Edge
* * 1..*
Actor
*
*
Social Relationship
Collaboration
Network Graph
Fig. 2 UML class diagram of a universal actor content-element model
1 A description of a content-actor model appears in [KPLB07]. Although their model includes network data, network analysis and visualization itself is not considered.
Defining a universal actor content-element model
127
one actor, while an actor can have multiple content elements. In addition, actors are connected with each other in social relationships (e.g. collaboration, acquaintance, communication, etc.), and content elements are linked to each other in contextual relationships (e.g. hyperlinks, common topics, wiki categories, etc.). Finally, an information space contains or aggregates other information spaces. Based on this informal description, a formal definition of an information space is derived: S = (A, P, Sd ), where A consists of a set of actors, P is a set of content elements, and Sd refers to a set of additional contained information spaces, which may be empty.
2.2 Defining Temporal Collaboration and Wiki-link Networks The universal model defines two main concepts, actors and content elements, which are related by specific relationships. In this section, the definition is applied to derive two network descriptions for a social network (collaboration network C = [ A × A]) and an information network (Wiki-link network W L = [P × P]). These networks are specified by using basic entities of a Wiki (cp. Figure 1). The time is one major variable in the network definition as opposed to conventional network descriptions. A temporal collaboration network is an undirected graph G C with a set V A of authors as vertices, and a set E C of collaborations as edges. Basically, an author Ai creates a version vti ∈ V E R at a certain time ti . This version vti belongs to a page P, which itself consists of versions P = {vt0 , ..., vti , ..., vtn } ⊆ V E R. A collaboration C = (Ai , A j ) exists if Ai is the author of vti , A j is the author of vt j and vti , vt j ∈ P. Based on the defined interval [tstar t , tend ] the version is incorporated into the network. The temporal collaboration network is a weighted graph, because each edge has a specific weight based on the number of mutual versions WV E R of two connected authors. Only the versions which rely on the same content elements and which are established and exist in the defined period of time [tstar t , tend ] are considered. This combined information leads to the following description: G C (V A , E C , WV E R , tstar t , tend ). A temporal Wiki-link network is a directed graph G W L consisting of a set of Wiki pages V P and a set of Wiki-links E L , where (P1 , P2 ) defines a connection from a page P1 ∈ V P to a page P2 ∈ V P . The first version vt0 of a page Pi is before the upper boundary of the analyzed time interval: t0 ≤ tend . Furthermore, links E L exist in a version vti with ti ≤ tend . Only a subset of the previously introduced page types is used for this network: V P = V A R ∪ VM E and E L = E A R L ∪ E M E L , where V A R are articles, VM E are media files, E A R L are article links and E M E L are media
Fig. 3 Kamada-Kawai-based visualization of a Wiki-link network; category: human computer interaction
128 Claudia Müller et al.
Defining a universal actor content-element model
129
links. The existence of broken links is considered in a single measure. All these considerations are summarized in the following description: G W L (V P , E L , tstar t , tend ). The network descriptions can be defined on a more abstract level. The collaboration network is then an instance of an interaction network, and the Wiki-link network is a context network. For example, in Flickr, content elements are photos or tags attached to photos. Actors share these content elements. The interaction network can be therefore described as a collaboration network in which actors are linked when they use the same tag. Based on these network definitions, the measurement model is presented in the next section. Finally, the connection of the universal model, the network descriptions and the measurement system provides a complete toolbox to investigate virtual information spaces.
2.3 Specifying Measures for Information Spaces At this stage, the universal model and the temporal network descriptions are defined. The last part of the approach is to specify a measurement system. Generally, a measure m is a function m : U → Q which assigns a measured value q ∈ Q to each entity u ∈ U , with q being the value of a quantifiable attribute of u. The measurement system is applied to time-dependent data. Measures are classified in two main categories: network specific and information space specific measures. These categories contain further classes. At the moment, the measurement system is only implemented for Wiki-based information spaces. The aim is to specify a solution which is independent of the investigated information space. An exploration of the previously defined networks is carried out by measures which belong to NodeMeasure, EdgeMeasure, and NetworkMeasure classes. On a microscopic level, the analysis, especially of vertex positions, is accomplished by measures such as Betweenness Centrality, Eigenvector Centrality, and Page Rank. An analysis of network structures is realized by measures such as Density, Diameter and Distance. The calculation of these measures is not explained here. Therefore further reading is recommended [dCRTB05]. The classes ContentElementMeasure, ActorMeasure, and InfoSpaceMeasure are domain specific measures. They are classified here into five groups (cp. Table 1). Activity related measures show dynamics in terms of conducted changes of content elements or by authors. Scope related measures evaluate the range of changes, i.e. the number of pages an author worked on, or the number of authors who worked on a page. The growth reflects the number of newly created pages. The size simply shows the extent of a page or contribution and the linking describes the connectedness in the information space. All defined networks and measures are realized in software which is introduced in the next section.
130
Claudia Müller et al. Table 1 Classification of Wiki-specific measures
Measure type
Actor (A)
Content element (P)
Activity Scope
Page Edit Count Page Changed Count
Version Count Author Count
Growth
Page Creation Count
Size
Contribution Amount
Linking
Added Internal/ External/ InterWiki Links
Information space (S)
Total Page Edit Count Total Page Changed Count Page Creator Total Page Creation Count Page Size Total Page Size/Contribution Amount Page Internal/ External/ Total Internal/ External/ InterWiki Links InterWiki Links
2.4 Applied Software All theoretical considerations are united in one software solution. SONVIS2 is a Java-based, open source software that is based on the Eclipse Rich Client Platform (RCP). RCP enables the creation of generic applications with a native graphical user interface. It offers a simple dynamic component-based model, which contains the functionality of the application, to develop plug-ins. Network visualization is implemented using the Prefuse Visualization Toolkit3 , and measurement on graphs is calculated by GNU R, an open source software environment for statistical computing and graphics4 . This architecture provides an Eclipse extension point for connector plug-ins to implement other data sources (e.g. other Wiki software, Weblogs, etc.) and extension points for other analyses. The SONVIS graphical user interface is based on completely configurable workspace layouts. Currently, two main perspectives are pre-defined: analysis perspective and manipulation perspective. A general overview of certain developments is given in the analysis perspective. After loading the selected database, basic Wiki measurements are automatically conducted and results are visualized in tables and specific diagrams. All measures are categorized according to the introduced measurement system (cp. Section 2.3). First statistical descriptive measurements (i.e. ratio of user/author) and temporal measurements (i.e. author growth, number of articles and rate of change) are graphically visualized. The manipulation perspective enables an enhanced visual manipulation of loaded networks. A filter function allows specific analyses of given categories, namespaces and tie periods. There are three different network layout algorithms implemented: Kamada-Kawai, Fruchterman-Reingold and Circle Layout. A visual filtering of the network is possible by adjusting the minimal node degree and the minimal edge
2
More information are available at http://www.sonivis.org. http://prefuse.org/ 4 http://www.r-project.org/ 3
Defining a universal actor content-element model
131
weight. The node size and color can be changed depending on selected measures. All calculated measures and networks are automatically transferred to the XMLbased GraphML format. In the next section, the existing potential of applying SONVIS to explore Wikibased information spaces is shown.
3 Scenario The English Wikipedia version is the largest with 2,708,022 content pages (15,771,620 pages altogether), 8,752,671 registered users and 146,069 active users (who had been active in the last 30 days) in January 2009. The complete Wikipedia content from April 2008 has been downloaded and extracted. On this local database copy, specific database procedures have been executed. These precomputed database tables significantly raise performance. For example, one of the pre-calculated tables contains all the links of every version. Therefore, the calculation time is reduced because the search for links in all versions is unnecessary. In the scenario, only a small part of the whole Wikipedia content is used to illustrate the practical application of the proposed model. All categories within the category human computer interaction (HCI) are extracted with a path length of 2. The used data set consists of 77 articles with 37,757 versions made by 16,904 authors (6,629 named authors, 10,275 anonymous authors), who have contributed 1,281,581 kBytes of raw text. Selected functions of SONVIS are introduced to investigate the page link structure and general usage patterns based on editing activities. The first step is to get a general impression of the content structure. For this, the Wiki-link network of the data set is visualized by selecting the maximal time period. The color-coding functionality of SONVIS is applied to the visualized network. The bigger the size of the node, the higher the Page Size value, while node color indicates the Version Count of a page. An interesting fact is that, a more precise image of assigned articles in the HCI category was expected, but categories such as virtual reality (with pages William Gibson and virtual reality) dominate the topical range. However, the network visualization reveals five conspicuous vertices in terms of their color and size: radio frequency identification (RFID), mouse (computing), virtual reality, ergonomics and William Gibson. Selected characteristics of these articles are shown in Table 2: Page
Table 2 Overview of five selected articles; category: human computer interaction Article
m p_sz
m ver
m au
m nau
RFID Mouse (computing) Virtual reality Ergonomics William Gibson
96,711 56,518 37,780 12,695 88,399
3,242 2,405 1,462 1446 1,216
1,739 1,260 745 671 380
582 668 329 262 227
132
Claudia Müller et al.
Size m ps z , Version Count m ver , Number of Authors m au and Number of Named Authors m nau . Based on these results, the collaboration structures of the RFID article are investigated more deeply. The size of the node shows the Contribution Amount, which is defined by the number of characters an author has contributed measured in kByte, whereas the color shows the editing activity (Page Edit Count) of the author. The first edit was on 15th January, 2003. The collaboration network is visualized in four different timeframes (cp. Figure 4). The aim is to identify authors who are steadily active during the investigated period of time. A point to note is that, all authors of the first timeframe, we call them pioneers, are only active in the first but not during the following timeframes. In the third phase, the “author” with the highest Contribution Amount is a bot that reverts vandalism. By considering the activity in terms of Page Edit Count, a user can be identified who has actively contributed to Wikipedia since 2005. The last period shows a similar contributor distribution. Two bots and one author with an administrator status have contributed most in terms of character count. The highest activity is from a Wikipedian who patrols recent changes. His Contribution Amount is therefore negative because of the high number of reverts he has carried out.
(a) 2003-01-15 to 2004-05-04
(b) 2004-05-04 to 2005-08-23
(c) 2008-08-23 to 2006-12-12
(d) 2006-12-12 to 2008-04-01
Fig. 4 Four timeframes of the collaboration network, article: RFID
Defining a universal actor content-element model
133
While presenting this very short scenario, more questions are asked than answered. However, the object was to show the existing potential of applying SONVIS to Wikipedia.
4 Conclusion and Further Research In this paper an approach is presented to explore virtual information spaces employing three models which are built on each other. Firstly, the general information sharing approach of Wikipedia was analyzed and the main entities, actors and content-elements, are identified. Starting from these concepts, a universal analysis model was composed which served as the basis for temporal network definitions. In order to understand the network evolution, a measurement system was defined. In a scenario, the existing potential of integrating different analysis and visualization methods was highlighted. Wikipedia was used as example to illustrate the viability of the proposed solution. However, the evaluation of the SONVIS software in terms of usability, extensibility, and interoperability is still missing. For this, it is planned to more exclusively integrate potential users to existing software development processes. The universal actor content-element model is a generally valid concept, because Social Web applications have the same inherent structure; they consist of content elements as well as actor information. The proposed model enables the investigation of previously separated Web applications, since an information space can contain other information spaces (cp. Section 2.1). For example, the Wiki analysis can be combined with an email analysis if there is a sufficient overlap in the actor group. At the present time, new connectors for Folksonomies (esp. Bibsonomy) and source code management systems (esp. SVN) are implemented in order to reveal existing similarities of such information spaces in terms of evolutionary community processes. In addition, it is intended to generalize and extend the measurement model. Text mining techniques are integrated to augment the traditional network analysis. Acknowledgements The development of SONVIS is a team effort. We thank the core members of the SONVIS Team: Anne Baumgraß, Sebastian Burkhart, Andreas Erber, Janette Lehmann, Robert Schmidl and Irene Sturm.
References [AdAPR08]
[BA06] [BCD+ 06]
B. Thomas Adler, Luca de Alfaro, Ian Pye, and Vishwanath Raman. Measuring Author Contributions to the Wikipedia. Technical report, School of Engineering, University of California, May 2008. Robert P. Biuk-Aghai. Visualizing Co-Authorship Networks in Online Wikipedia. Communications and Information Technologies, 2006. ISCIT ’06. International Symposium on, pages 737–742, 18 2006-Sept. 20 2006. Luciana Buriol, Carlos Castillo, Debora Donato, Stefano Leonardi, and Stefano Millozzi. Temporal Analysis of the Wikigraph. In Proceedings of the Web
134
Claudia Müller et al.
Intelligence Conference (WI 2006), pages 45–51, Los Alamitos, CA, USA, December 2006. IEEE Computer Society. A. Capocci, V.D.P. Servedio, F. Colaiori, L.S. Buriol, D. Donato, S. Leonardi, and [CSC+ 06] G. Caldarelli. Preferential attachment in the growth of social networks: the case of Wikipedia. Physical Review E, 74:036116–1–6, 2006. [dCRTB05] Luciano daF. Costa, Francisco A. Rodrigues, Gonzalo Travieso, and P. R. Villas Boas. Characterization of complex networks: A survey of measurements. Advances in Physics, 56(1):167–242, January 2005. [Fou09] Wikimedia Foundation. MediaWiki MySQL database layout. URL: http://www. mediawiki.org/wiki/Manual:Databaselayout, 2009. [Fuc04] Christian Fuchs. Knowledge Management in Self-Organizing Systems. Journal of Knowledge Management Practice, 5, 2004. [HSH+ 08] James Hendler, Nigel Shadbolt, Wendy Hall, Tim Berners-Lee, and Daniel Weitzner. Web science: an interdisciplinary approach to understanding the web. Commun. ACM, 51(7):60–69, 2008. A. Kittur, E. H. Chi, B. A. Pendleton, B. Suh, and T. Mytkowicz. Power of the few [KCP+ 07] vs. wisdom of the crowd: Wikipedia and the rise of the bourgeoisie. In 25th Annual ACM Conference on Human Factors in Computing Systems (CHI 2007), San Jose, CA., 2007. [KPLB07] Hyunmo Kang, Catherine Plaisant, Bongshin Lee, and Benjamin B. Bederson. NetLens: iterative exploration of content-actor network data. Information Visualization, 6:18–31, 2007. [NT08] Fawad Nazir and Hideaki Takeda. Extraction and analysis of tripartite relationships from Wikipedia. In IEEE International Symposium on Technology and Society, ISTAS 2008, pages 1–13, Sydney (Australia), June 2008. [SCPK07] Bongwon Suh, E.H. Chi, B.A. Pendleton, and A. Kittur. Us vs. Them: Understanding Social Dynamics in Wikipedia with Revert Graph Visualizations. Visual Analytics Science and Technology, 2007. VAST 2007. IEEE Symposium on, pages 163–170, 30 2007-Nov. 1 2007. [STSG05] B. Stvilia, M. B. Twidale, L. C. Smith, and L. Gasser. Assessing information quality of a community-based encyclopedia. In Proceedings of the International Conference on Information Quality - ICIQ 2005, pages 442–454, 2005. [VWKvH07] Fernanda B. Viegas, Martin Wattenberg, Jesse Kriss, and Frank van Ham. Talk Before You Type: Coordination in Wikipedia. In 40th Annual Hawaii International Conference on Systems Science (HICSS), pages 78–78, Jan. 2007. [WVH07] Martin Wattenberg, Fernanda Viégas, and Katherine Hollenbach. Visualizing Activity on Wikipedia with Chromograms. Human-Computer Interaction – INTERACT 2007, pages 272–287, 2007. [ZBSD06] V. Zlatic, M. Bovzivcevic, H. Stefanvcic, and M. Domazet. Wikipedias: Collaborative web-based encyclopedias as complex networks. Physical Review E, 74, 2006.
A Composite Calculation for Author Activity in Wikis: Accuracy Needed Claudia Müller-Birn, Janette Lehmann, Sabina Jeschke
Abstract Researchers of computer science and social science are increasingly interested in the Social Web and its applications. To improve existing infrastructures, to evaluate the success of available services, and to build new virtual communities and their applications, an understanding of dynamics and evolution of inherent social and informational structures is essential. One key question is how communities which exist in these applications are structured in terms of author contributions. Are there similar contribution patterns in different applications? For example, does the so called onion model revealed from open source software communities apply to Social Web applications as well? In this study, author contributions in the open content project Wikipedia are investigated. Previous studies to evaluate author contributions mainly concentrate on editing activities. Extending this approach, the added significant content and investigation of which author groups contribute the majority of content in terms of activity and significance are considered. Furthermore, the social information space is described by a dynamic collaboration network and the topic coverage of authors is analyzed. In contrast to existing approaches, the position of an author in a social network is incorporated. Finally, a new composite calculation to evaluate author contributions in Wikis is proposed. The action, the content contribution, and the connectedness of an author are integrated into one equation in order to evaluate author activity. Keywords Social Web · Wikipedia · social networking · collaboration networks
policies, organization structure, etc.) [HSH+ 08]. Such virtual communities lead to a new form of organizing information production which is decentralized, collaborative and non-proprietary. Resources, such as photographs, bookmarks, posts and tweets, are shared and new, collaboratively created outputs are distributed between loosely connected individuals [Ben07]. Understanding dynamics and evolution of virtual communities and the underlying social infrastructure is essential to improve existing infrastructures, to evaluate the success of available services or to build new virtual communities. The motivation is to transfer existing research results from the area of open source software development communities to open communities. A general model to describe author contributions in open communities is sought. In this contribution, a composite calculation to classify author contributions in Wikis is defined. This contribution is organized as follows. 1. Existing studies of Wikipedia research are introduced to present the different approaches of evaluating author activity. 2. The dynamic collaboration network is defined and three approaches to evaluate author activity in such a network is described. The edit count is utilized as an action calculation for author contributions, the content significance shows the real content contributions of an author, and the betweenness centrality is used to evaluate the leverage (i.e. topical influence) of author contributions. 3. The data set and the pre-processing of the data set are explained. All previously introduced calculations are applied. Firstly, different groups of contributors in terms of their importance to the content growth are investigated. Secondly, all presented calculations are separately applied. Thirdly, the action of an author, the added significant content, and the topical influence are combined to reveal the main and most important contributors. 4. A discussion about existing limitations in this research and an outlook on further research activities is given.
2 Related Work There is a broad range of studies that deal with Wikipedia, answering questions in terms of overall link structures and usage patterns. Existing studies can be roughly divided into explanatory studies, aggregated approaches (i.e. graph structures), and visualization techniques. In one study, the importance of privileged users for content creation at the beginning of Wikipedia’s growth is uncovered [KCP+ 07]. In the investigated period, the influence of the elite declines and the work of ordinary authors is increasingly important. In another study, existing methods evaluating editing patterns are extended by differentiating system roles (e.g. admin, sysop) of authors [OB07]. Author behavior is compared in different language editions. Furthermore, author edits are evaluated in terms of their influence on article quality [SH07]. Within an extensive overview about existing calculations, quantitative
A Composite Calculation for Author Activity in Wikis: Accuracy Needed
137
calculations are combined with qualitative measures [AdAPR08]. A contribution is only considered if other authors accept the added text of another. In aggregated studies, Wikipedia is interpreted as Wikigraph and its temporal development [BCD+ 06], characteristic of preferential attachment [CSC+ 06], and its complex nature [ZBSD06] are investigated. More than that, hidden social patterns are revealed by applying social network analysis to so-called revert graphs [SCPK07]. Visualization techniques are carried out to substantiate specific aspects: for example, the editing activity of Wikipedia is visualized by Chromograms [WVH07], changes on talk pages are illustrated by history flow diagrams [VWKvH07], and co-authorship networks show the relationship between articles [BA06]. The existing research shows different approaches to evaluate author contributions. An integrated approach which deals with a combined calculation from data mining, text mining, and network analysis is still missing. This research intends to close this gap by proposing a composite calculation to deal with an integrated approach.
3 Methodology In this section the dynamic collaboration network in Wikis is defined and three perspectives to evaluate author contributions in such networks are introduced. The open source visual analytics software SONIVIS [SON09] is applied. This software facilitates the exploration of social and information networks in Wiki-based information spaces. It is based on the Eclipse Rich Client Platform (RCP). Network visualization is implemented using the Prefuse Visualization Toolkit, and network measurements are calculated by GNU R, an open source software environment for statistical computing and graphics.
3.1 Dynamic Collaboration Network Wikis can be seen as dynamic social systems which can be described by two main entities, actors and content, linked by specific relationships. These entities are used to exemplarily derive one network description for a social network. As opposed to conventional network descriptions, time is considered as one major variable which addresses the dynamic character of a Wiki. A dynamic collaboration network is an undirected graph G C with a set V A of authors as vertices, and a set of collaborations as edges E C = {(ai , a j )}, where ai , a j ∈ V A with i, j ∈ N (i = j). Basically, an author ai creates a version vti ∈V E R at a certain time ti . This version vti belongs to a page P, which consists of P = {vt0 , ..., vti , ..., vtn } ⊆ V E R revisions. A collaboration C = (ai , a j ) exists if ai is author of vti , a j is author of vt j and vti , vt j ∈ P and the particular version is created in the defined interval [tstar t , tend ]. The dynamic collaboration network is a
138
Claudia Müller-Birn et al.
weighted graph with each edge having a specific value based on the number of mutual versions WV E R of two connected authors. The combination of this information leads to the following description: G C (V A , E C , WVER , tstar t , tend ). Basically, the dynamic perspective of network analysis allows for the analysis of network states, vertices and their edges, as well as changes in structure and configuration of the network [Car03]. There are different approaches to visualize dynamic networks: cumulative and sliding-window based analysis (cp. [GLZD03], [MMBd05]). They differ in the way of recording the network parameters. Cumulative analysis includes the aggregation of all changes and elements in one network over a relatively long time period. Changes in the network structure are visible, but the sequence and existing interdependencies in network processes are not identifiable. In the sliding-window based approach, a small time period of network life is analyzed. The network is completely detached and is based on single events. A sequence of these events can be evaluated in terms of time of occurrence and rhythm. In this study, a static visualization of the network based on the specified time periods is shown.
3.2 Action Analysis of Collaboration Networks The edit count is an often applied measure to assess author actions in Wikis. For example, edits of different authors are applied to evaluate the page based reputation of an individual [SH07]. The edit count is also used as a descriptive indicator for measuring Wiki dynamics [RTG08]. More formally, the Wikipedia content set C consists of content elements (pages) P = { p1 , ... pm }. Each page has different versions Pi = {vt0 , ...vtn }. Each version vti is created by an author ai at a certain time ti . The edit count can be described as follows [AdAPR08]: m EC (ai ) =
pm tn
vtn .
p=1 t0
The easier the calculation of the edit count, the more difficult the interpretation of this measure is. For example, a high edit count does not relate to a large text because it can be caused by “minor edits”. A person who profoundly extends the description of one article, might have the same edit count as a person who just improves the punctuation. Even more difficult is the fact that the quantitative edit count of an ordinary author might be equal to the edit count of vandals and spammers, whereas the qualitative result of the edit is completely different. However, the calculation to evaluate the activity level of an author is applied and the significance of how this measure can be improved is shown.
A Composite Calculation for Author Activity in Wikis: Accuracy Needed
139
3.3 Content Analysis of Collaboration Networks The measure significant content describes the relevant content a user has added. The significance of each term is calculated by weighting all terms depending on their local context. Only terms with a high local frequency (in the page) and a low distribution in the content set are taken into account. The content set C is a specific collaboration network, including all pages that are edited. Basically, the term frequency f i,n describes the amount a term tm i occurs in a page pn . Depending on the selected network adjustments, the function f is calculated for a specific period of time. The content set is defined by the actual growth of term development and all added and subsequently undeleted terms, all deleted and subsequently subtracted terms, and the union of both sets are considered. This definition refers to the dynamic nature of the defined network. Because pages exhibit different text lengths, the normalized term frequency is calculated. The normalized term frequency n f of a term tm i on a page pn is the term frequency f i,n divided by the total frequency of all terms contained in the defined content set C: n f i,n =
f i,n
tm i ∈C
f tm j,m
.
The normalized term frequency is thus the relative frequency of a term. The inverse term frequency id f of a term tm i is calculated by the total number of pages of the content set C, divided by all pages including the term tm i : id f i = log
|C| . |C : f tm i ∈ C|
Both measures are used to specify the importance of a specific term for a page: wi,n = n f i,n · id f i . Now, the relevance of the content a user added to a page can be seen. The calculation content significance is defined as: m C S (ai ) =
i max
f i,n,ai · wi,n ,
i=1
where fi,n,ai is the frequency an author ai has added a term tm i on a page pn and where wi,n is the importance of a specific term for this page. In the next section, a calculation to evaluate the position of an author in a collaboration network is introduced.
3.4 Leverage Analysis of Collaboration Networks The betweenness centrality is adopted to evaluate the leverage of author activities in dynamic collaboration networks. It is assumed that authors edit not only one
140
Claudia Müller-Birn et al.
article, but more likely various articles in a topic area. Authors who edited different articles are between other authors in a network. The more often a person has this position, the more central this author in terms of betweenness centrality is. Authors who work on different articles in a field have a better overview of existing topics or missing topics. They are able to compare the quality of different articles perhaps more objectively. Generally, a vertice that resides on many of the shortest paths between other vertices has a higher betweenness than those that do not [Fre77]. For an author in a collaboration network, the betweenness centrality is measured thus [WF97]: m C B (ai ) =
j
g jk (ai )/g jk
[(g − 1)(g − 2)/2]
,
where j = k = i, n i , n j , n k ∈ g, g is the number of edges and g jk is the shortest path (geodesic distance) between two authors a j and ak where ai is on this path. The aim is to identify authors who have the highest leverage in terms of influence. These individuals have better access to network information. The author contribution is based on the position of a person in a network. A value equal to zero means that this person contributed to only one article. The higher the value, the more different articles an author has edited. In this investigation, the amount of editing activities of an author on one article was not considered. The following sections contain the application of these calculations showing their advantages and disadvantages, and clarify the need of a composite calculation.
4 Data Collection and Extraction In this section the pre-processing of the chosen data set is explained. The English Wikipedia is the largest encyclopedic version of the whole project. A Wikipedia dump from April 2008 was downloaded and extracted. On this local database copy, specific database procedures were executed. These pre-computed tables significantly raised performance. For example, one of the pre-calculated tables contained all the links of every revision (i.e. version). The algorithm does not have to search for links in the revision’s text during the loading process of the network, but all information is available directly in one database table. In this scenario, only a small part of the whole Wikipedia was used because a better approximation of authors’ content contributions was expected because of a topical reduction. A category (i.e. topic range) in the Wikipedia corpus was selected. Human computer interaction (HCI) was chosen because of the familiarity with this topic. All categories within the category HCI, and all categorized articles were extracted with a path length of two. This data set now consists of 77 articles with a total of 37,757 revisions written by 16,904 authors (6,629 named, 10,275 anonymous).
A Composite Calculation for Author Activity in Wikis: Accuracy Needed
141
The next step is the pre-processing of the data set. Basically, the sum of all considered content elements (i.e. pages) in a Wiki represents the corpus. In a corpus, each content element consists of words. The set of all unique words in a corpus are called lexicon. Unique words are defined as terms. The pre-processing of content elements is carried out to prepare such a a lexicon. Firstly, all existing words of the investigated corpus (Wiki) are extracted. In this investigation, only articles in the main namespace of the Wikipedia corpus were considered. Secondly, all Wiki-specific (e.g. [category:]) and HTML-specific (e.g. ) syntax was deleted. The same procedure was carried out with stop words like “the” or “a”, because they do not add any meaningful content to pages. Approximately 1,584 specific and 34 general stop words were defined. Thirdly, the stemming process, which reduces words to their root by removing suffixes and prefixes, was considered. Similar words were aggregated, for example “networking”, “networked” and “networker” to the root word, “network”. In this analysis, the R package Rstem, which works with the Porter stemming algorithm realized by Snowball, was applied. Finally, all words which consist of less than three letters were deleted. The final corpus, the lexicon, became the working data set for this study. The period of analysis which proceeds from the first edit on 2001-07-23 to the last edit on 2008-03-14 was defined. The whole period was divided into 13 intervals, and each interval lasts 6 months. The first one was from 2001-07-01 to 2001-12-31, and the last one from 2007-07-01 to 2007-12-31. Consequently, the data set was reduced because in the first interval, the first 22 days and the last 3.5 months in 2008 were not included. In the first step, the number of pages (m PC ), the number of all authors (m AuC ) and the number of named authors (m i AuC ) from the HCI category were computed. Table 1 displays all categories ordered by the highest number of pages. It was decided to have a closer look at user contributions of the category virtual reality, because of the highest number of pages in this category. Table 2 contains all articles of the category virtual reality, as well as the corresponding page sizes (m P Si ze in Byte) and number of authors. This classification of pages refers to the categorization of the last time period. Of course, over time the categorization of pages changes, because the considered pages can be re-categorized, or a page or category can be newly created. In the next section, this data set was utilized to verify the hypothesis. Table 1 Characteristics of sub-categories in the category human computer interaction Category name
m PC
m AuC
m i AuC
Virtual reality User interface Human computer interaction (HCI) Computing input devices Graphical user interface History of HCI
18 17 9 7 7 7
4,362 2,936 1,467 2,188 1,969 1,978
1,977 1,394 726 1,015 962 957
142
Claudia Müller-Birn et al. Table 2 Characteristics of articles in the category virtual reality Article name
m EC
m P Si ze
m AuC
m i AuC
Virtual reality DirectX William Gibson Virtual community Anarchy Online OpenGL Flight simulator Cyberspace VRML User interface Artificial life Ray tracing Augmented reality Psychoacoustics X3D Bump mapping Gouraud shading Wire frame model XVROS
5 Results Using the defined data set, cumulative author activities based on action (edits made by users), content (added significant content by users), and leverage (betweenness centrality of an author in a network) were computed. The results were used to compare them with the results of the proposed composite calculation for author activity. In order to get an overall understanding of the distribution of author activity, three groups were specified. A defined approach [CH03], which proposed an onionlike social structure in open source software projects was adopted. Similar content contribution structures in the open content project Wikipedia were assumed. The onion model describes how contributors are positioned in such open communities. C ROWSTON AND H OWISON differentiate core developers, who are highly involved in the project, co-developers, who have specific but frequent contributions, and active users, who contribute only occasionally. A quite similar definition is given in the theory of “Participation Inequality” to classify authors in online communities. According to this concept, user participation often follows a nearly 90-9-1 rule, in which 90% of users are lurkers, 9% of users contribute occasionally, and 1% of users participate a lot and account for most of the contributions [Nie06]. Lurkers are here defined as persons who contribute minimal content. This group is characterized by high fluctuations and a low participation rate. In Wikis, this group consists mainly of edits made by anonymous users. In the following calculations, this group allocation was applied to define groups of low, middle and high contributions. It was proposed to compare the mainly applied measure edit count, to evaluate the wisdom of the crowds “idea”, with the
A Composite Calculation for Author Activity in Wikis: Accuracy Needed
143
defined content significance. This idea refers to the argument that the quality of Wikipedia articles is caused by the contributions of crowd members rather than by single experts (cp. [KCP+ 07]).
5.1 Author Activity Based on Edits Initially the cumulative edit count for each author in the whole period of time from 2001-07-01 to 2007-12-31 is calculated. Three groups are defined based on the results. The group of low contributions was defined by 0 ≤ m EC ≤ 3 and contained 4,126 authors (90.94 %). The group of middle contributions with 4 ≤ m EC ≤ 20 had 383 authors (8.44%) and the last group, the group of high contributions, had 28 authors (0.62%) with m EC > 20. Next, the edits in each of the defined time intervals were quantified and each author, depending on his period based edit count, was assigned to one of the groups. Figure 1 shows the cumulative edit count by groups. According to the results [KCP+ 07], the user group of low contributions exhibited the highest activity based on the edit count. The group of low contributors had the highest editing activity, if all edits were totalled. In comparison, the group of middle and low contributors only played a minor role. The differences in the chart compared to the results presented [KCP+ 07], were caused by the different group levels but the overall impression of the high impact of authors with fewer contributions was still the same. A closer look at the group of high contributors was taken and a ranking was created. It showed a list of authors with their highest (cumulative) value in all peri-
Fig. 1 Cumulative edit count per user group
144
Claudia Müller-Birn et al. Table 3 Edit count of highest ranked authors Author name
Active periods tactive tactive tactive tactive tactive tactive tactive tactive tactive tactive
=2 =2 =2 =2 = 11 =4 =1 =4 =1 =4
ods and the number of periods an author was active. Table 3 shows the ten highest values. The group comprises administrators and two rollbackers. For example, in period 13, the user Skomorokh (rollbacker) had the highest cumulative edit count over all periods, but the user Frecklefoot was the most continuously editing person in the network. In the next section, this distribution was compared to the significant content contributions, and again specific authors were looked at.
5.2 Author Activity Based on Content Significance Firstly, all terms and their frequencies in the category HCI were computed. Then the importance of each term in a specific article was calculated. Selected terms (highest ranked) in the category virtual reality were for example: xvros (Project “eXtensible Virtual Reality Operating System”), bump (Bump mapping, computer graphics technique), embm (Environment Mapped Bump Mapping—3D graphics technology), gouraud (Gouraud Method used in computer graphics), and ray (Ray tracing, graphics, is used for 3D image generation). After calculating the significant content added by each author (cp. Section 3.3), the group borders were specified based on the previously introduced procedure. The usage of the computed terms was examined in the whole period of time for all authors. The following group borders were defined: the group of low contributors 0 ≤ m C S (ai ) ≤ 45 exhibited 2,264 authors (90.38%), the group of middle contributors 46 ≤ m C S (ai ) ≤ 1, 000 had 216 members (8.62%) and the last group of high contributors (m C S (ai ) > 1, 000) contained 25 members (0.998%). Figure 2 shows the results of the group division in each period. The resulting chart is completely different compared to the diagram in Figure 1. Until the beginning of 2005, the added terms in each group were approximately the same. A different development existed from 2005. The high contributor group had an outstanding position compared to the other user groups. The contributors with middle activities were not very balanced in their activities, but overall, the group of low contributors showed the smallest topical engagement in the development of the category virtual reality. These results contradicted the previous findings. As opposed to various studies [OB07] and
A Composite Calculation for Author Activity in Wikis: Accuracy Needed
145
Fig. 2 Cumulative edit count per user group
[KCP+ 07], the importance of high contributors was revealed for the overall content development, even though the edits of this group were very low. These authors were examined further. Table 4 shows the most important authors in terms of their added significant content. Interestingly, the group of authors changed completely compared to the previous analysis, because of the changed perspective. Six of the ten authors in this list are administrators in Wikipedia. The bots (computer programs which do specific automated tasks) which have in this calculation the highest values were ignored. This was due to the fact that it was not considered whether the added content was new content or previously existing text. In future studies a potential to improve this measure exists.
Table 4 Added significant content of highest ranked authors Author name
Active periods tactive tactive tactive tactive tactive tactive tactive tactive tactive tactive
=1 =1 =2 =1 =1 =1 =1 =1 =1 =1
146
Claudia Müller-Birn et al.
5.3 Author Activity Based on Leverage In order to evaluate the topic coverage of authors, the dynamic collaboration networks in every period was visualized. One period was relatively long (half a year), but only an impression of the overall development was required. Based on the network visualization, the growing number of active people over three years and the growing number of individuals who worked on different articles could be seen. Even though every window had the same size, the number of authors increased above average. Exemplarily, two network slices of the dynamic collaboration network are presented in Figure 3. The node size depends on the significant content an author has added and the node color illustrates the edit count of an author. The different articles were designated to show the growing number of individuals who edited more than one article on the network. Whereas in period 5 (2003-07-01 to 2003-12-31), 83 authors changed 16 articles, in period 9 (2005-07-01 to 2005-12-31) 540 authors had already edited the same number of articles in the category. If the general activity and content development is observed, the same results will be obtained. The number of authors is growing, but the number of articles is almost the same. In earlier periods, the betweenness centrality of the collaboration network was lower than in following periods; that meant that more people worked on different articles on the network. In Table 5, the most central authors in terms of their betweenness centrality were observed. This list also shows different authors. Two of these authors were administrators in Wikipedia. As oppose to the previous two calculations, the betweenness centrality is not a cumulative value because a pure addition of the values changes the validity. The values shown here are the highest values in all periods. After the third calculation, it became more complicated to identify the most important authors. Each ranking showed important authors in terms of one specific aspect. The intersection between the different sets was very small. Consequently, a new measure is proposed to get a more consistent picture.
5.4 Integrated Calculation to Evaluate Author Activity In this section, the different interpretations of author contributions in one description are integrated. The three calculations deal with an author’s action, content and connectedness. Each of these calculations should be considered equivalently and high values of authors should not skew the result. Periods in which one value is equal to zero are regarded. For example, if an author has a very high edit count in one period but contributed no significant content, or worked on only one article, one of the values is zero. All values are incorporated in order to consider the continuity of author activities. All these considerations lead to the following equation:
m AC (ait ) =
m˜ C S (ait )0.5 ∗ m C B (ait )0.5 , m˜ EC (ait )0.5
A Composite Calculation for Author Activity in Wikis: Accuracy Needed Ray tracing
147
VRML
Virtual community X3D
Bump mapping
Direct X
Virtual reality
Cyberspace
Flight simulator
Psychoacoustics
William Gibson
Gourad shading Virtual reality
Anarchy online Open GL Augmented reality
(a) Period 5: Jul 03 – Dec 03 XVROS Augemented reality
Anarchy online
Artifical life
Virtual reality
Flight simulator
Psychoacoustics
Wire frame model
User interface design
Cyberspace William Gibson OpenGL Gourad shading VRML DirectX
Bump mapping
(b) Period 9: Jul 05 - Dec 05 Fig. 3 Dynamic collaboration network; node size m C S (ai ), node color m EC (ai ); pages are highlighted by dashed line elipses
148
Claudia Müller-Birn et al. Table 5 Betweenness centrality of highest ranked authors Author name
Active periods tactive tactive tactive tactive tactive tactive tactive tactive tactive tactive
=1 =1 =1 =1 =2 =5 = 11 =2 =2 =2
where m˜ EC (ait ) = 1 −
1 m EC (ait )α
and m˜ C S (ait ) = 1 −
1 1 + m C S (ait )α
with α = 0.25. The author contribution for all authors in each period was calculated. The different ranges of calculated values were adjusted in m˜ EC (ait ) and m˜ C S (ait ). The quadratic mean was applied to determine the cumulative author activity: tmax 1 m AC (ai ) = (m AC (ait ))2 . tmax t 0
There are different advantages of applying this measure. Authors who are doing maintenance work on Wikis have a high edit count and betweenness centrality, but low content significance. Authors who work mainly on a few articles and contributed a lot of content, have a comparatively low value for its betweenness centrality, but a high value for content significance. Table 6 contains the authors with the highest values of author activity based on the composite calculation. As expected, the ranking of the authors differs, but there are authors who are already known, such as Frecklefoot. This author has the highest betweenness centrality; however the edit count and the content significance is comparatively high. Therefore, this author can be seen as knowledge champion for the category virtual reality. Compared to all other authors shown, he has an outstanding position. However, the lower classified authors have very different activity patterns. For example, one author was active in a few periods and his activities were restricted to a few articles, but this author appears in the list because the overall activity and content contribution were relatively high.
A Composite Calculation for Author Activity in Wikis: Accuracy Needed
149
Table 6 Author activity of highest ranked authors Author name
Active periods tactive tactive tactive tactive tactive tactive tactive tactive tactive tactive
= 11 =1 =1 =5 =2 =2 =1 =1 =2 =2
6 Conclusion, Discussion and Further Research In this paper, author contributions in Wikipedia were analyzed considering three perspectives. First of all, the edit count was used to evaluate the activity of authors. Secondly, based on text mining techniques, authors contributions in terms of the added significant content were analyzed. Thirdly, the betweenness centrality of authors in a collaboration network were assessed to estimate the leverage of author activity. Each of the introduced measures had certain advantages and disadvantages in presenting author activity on Wikis. But each calculation revealed different rankings of author activity. It was realized that a pure ranking of the highest values distorted results. A new measure was therefore proposed that combined all three introduced calculations. This calculation balanced the different aspects of author contributions in a collaboration network. With the table based visualization of results used and the very small number of results shown, it was very difficult to achieve an overall impression of the author activity in the data set. Furthermore, the changing positions in the rankings were only visible in the subsets. Therefore, a more appropriate visualization would be very helpful. There is some potential for improvement regarding the proposed calculation. The considered periods are relatively long. A more precise calculation of author contributions is possible by using shorter periods. Our new calculation has some problems in identifying vandals and spammers, as well as peoples such as rollbackers. Here, another component, the so-called amount contribution, helps. The amount contribution is the number of characters in a Byte (UTF-16 → 1 character = 2 Byte). Another approach is to extend the composite calculation by considering how long edits last [AdAPR08], for instance, how long the added content stays in an article. This could be a fourth component of the equation. Furthermore, the calculation of the added significant content revealed the high influence of bots. A better result could be achieved by only regarding new and not previously existing content. In further studies, different content sets should be compared to validate the proposed calculation. In addition, the choice to select the data set based on one category would be revised. Results could be more comparable if an article selection is used.
150
Claudia Müller-Birn et al.
Another challenge is to extend all measures by regarding the time, because an incorporation of the number of active periods, in other words the continuity of author contributions, is missing. Altogether, it can be shown that only by considering more than one measure, and by integrating these into one composite calculation, can author contributions be evaluated more exactly. Acknowledgements We thank the core members of the SONIVIS:Team: Anne Baumgraß, Sebastian Burkhart, Andreas Erber, Benedikt Meuthrath, Robert Schmidl, and Irene Sturm.
References [AdAPR08]
[BA06] [BCD+ 06]
[Ben07] [Car03]
[CH03] [CSC+ 06]
[Fre77] [GLZD03] [HSH+ 08] [KCP+ 07]
[MMBd05] [Nie06]
B. Thomas Adler, Luca de Alfaro, Ian Pye, and Vishwanath Raman. Measuring Author Contributions to the Wikipedia. Technical report, School of Engineering, University of California, May 2008. Robert P. Biuk-Aghai. Visualizing Co-Authorship Networks in Online Wikipedia. Communications and Information Technologies, 2006. ISCIT ’06. International Symposium on, pages 737–742, 18 2006-Sept. 20 2006. Luciana Buriol, Carlos Castillo, Debora Donato, Stefano Leonardi, and Stefano Millozzi. Temporal Analysis of the Wikigraph. In Proceedings of the Web Intelligence Conference (WI 2006), pages 45–51, Los Alamitos, CA, USA, December 2006. IEEE Computer Society. Yochai Benkler. The Wealth of Networks: How Social Production Transforms Markets and Freedom. Yale University Press, 2007. Kathleen M. Carley. Dynamic Network Analysis. In Ronald Brelger, Kathleen Carley, and Philippa Pattison, editors, Dynamic Social Network Modeling and Analysis: Workshop Summary and Papers, pages 133–145. National Academy Press, Washington, D.C., 2003. Kevin Crowston and James Howison. The social structure of open source software development teams. In Proceedings of the International Conference on Information Systems, Seattle, WA (USA), February 2003. A. Capocci, V.D.P. Servedio, F. Colaiori, L.S. Buriol, D. Donato, S. Leonardi, and G. Caldarelli. Preferential attachment in the growth of social networks: the case of Wikipedia. Physical Review E, 74:036116–1–6, 2006. Linton C. Freeman. A set of measures of centrality based on betweeness. Sociometry, 40:35–40, 1977. Peter A. Gloor, Rob Laubacher, Yan Zhao, and Scott B.C. Dyne. Temporal Visualization and Analysis of Social Networks. Technical report, MIT, Center for Coordination Science, Cambridge, MA, 2003. James Hendler, Nigel Shadbolt, Wendy Hall, Tim Berners-Lee, and Daniel Weitzner. Web science: an interdisciplinary approach to understanding the web. Commun. ACM, 51(7):60–69, 2008. A. Kittur, E. H. Chi, B. A. Pendleton, B. Suh, and T. Mytkowicz. Power of the few vs. wisdom of the crowd: Wikipedia and the rise of the bourgeoisie. In 25th Annual ACM Conference on Human Factors in Computing Systems (CHI 2007), San Jose, CA., 2007. James Moody, Daniel McFarland, and Skye Bender-deMoll. Dynamic Network Visualization. American Journal of Sociology, 110(4):1206–1241, January 2005. Jakob Nielsen. Participation Inequality: Encouraging More Users to Contribute. URL: http://www.useit.com/alertbox/participation_inequality.html , 2006.
A Composite Calculation for Author Activity in Wikis: Accuracy Needed [OB07]
151
Felipe Ortega and Jesus M. Gonzalez Barahona. Quantitative analysis of the wikipedia community of users. In WikiSym ’07: Proceedings of the 2007 international symposium on Wikis, pages 75–86, New York, NY, USA, 2007. ACM. [RTG08] Camille Roth, Dario Taraborelli, and Nigel Gilbert. Measuring wiki viability: An empirical assessment of the social dynamics of a large sample of wikis. In Proceedings of the 4th International Symposium on Wikis (WikiSym 2008). Porto, September 2008. [SCPK07] Bongwon Suh, E.H. Chi, B.A. Pendleton, and A. Kittur. Us vs. Them: Understanding Social Dynamics in Wikipedia with Revert Graph Visualizations. Visual Analytics Science and Technology, 2007. VAST 2007. IEEE Symposium on, pages 163–170, 30 2007-Nov. 1 2007. [SH07] Klaus Stein and Claudia Hess. Does it matter who contributes: a study on featured articles in the german wikipedia. In HT ’07: Proceedings of the eighteenth conference on Hypertext and hypermedia, pages 171–174, New York, NY, USA, 2007. ACM. [SON09] SONIVIS. SONIVIS: Social networks in virtual information spaces. http://www. sonivis.org, 2009. [VWKvH07] Fernanda B. Viegas, Martin Wattenberg, Jesse Kriss, and Frank van Ham. Talk Before You Type: Coordination in Wikipedia. In 40th Annual Hawaii International Conference on Systems Science (HICSS), pages 78–78, Jan. 2007. [WF97] Stanley Wasserman and Katherine Faust. Social network analysis: methods and applications. Cambridge Univ. Press, Cambridge, 1997. [WVH07] Martin Wattenberg, Fernanda Viégas, and Katherine Hollenbach. Visualizing Activity on Wikipedia with Chromograms. Human-Computer Interaction – INTERACT 2007, pages 272–287, 2007. [ZBSD06] V. Zlatic, M. Bovzivcevic, H. Stefanvcic, and M. Domazet. Wikipedias: Collaborative web-based encyclopedias as complex networks. Physical Review E, 74, 2006.
Experiences from an International Student and Staff Exchange Program and Some Still Unsolved Mysteries Olivier Pfeiffer, Sabina Jeschke, Lars Knipping, Nina Reinecke, Erhard Zorn
Abstract This paper describes an ongoing exchange program between 20 partner universities; eleven from the European Union and candidate countries and nine universities form Jordan, Lebanon, and Syria, where a bilateral mobility flow between the European and the neighboring countries is implemented. While the idea of this program initially intended to focus on Information and Communication Technology (ICT) subjects only, it was later opened to students from all academic fields. Nevertheless, the better part of all participating students is from engineering disciplines. The described program encompasses undergraduate, graduate, and PhD students as well as postdocs and academic staff. By broadening their technical education, we think that all participants benefit from the reevaluation of their own cultures that occurs while functioning as part of another culture and communicating in a foreign tongue. Over 100 scholarships were awarded in the first year of the program and this number was exceeded for the second year of the program, which started in September 2009. The allocation process for this year is still underway and we expect to see a similar number of scholars in this third and final phase of the program. A part of the scholarships at the graduate and undergraduate level was granted to credit-seeking students. The rest of the graduate, undergraduate and all PhD scholarships were awarded to degree-seekers. One of the challenges that showed up was that, surprisingly for the Arab students and us, accrediting a degree obtained from some European university turns out to be much easier than getting a credit for just a single lecture attended at the very same university. The reasons for this are quite obvious: on the one hand, as a consequence of the Bologna Process, every European university today uses the European Credit Transfer and Accumulation System (ECTS) as a standard for comparing performance and achievement of students while on the other hand, all the partner countries are using the American academic system. Actually, one of them is an American university, i.e. the degrees awarded are officially registered by the Board of Education in New York State.
O. Pfeiffer (B) MuLF, TU Berlin, Straße des 17. Juni 136, 10623 Berlin, Germany e-mail: [email protected]
First experiences are reviewed, including the not exclusively academic obstacles that we had to overcome in teaching and learning within these different educational systems .We will also report on a survey, which we will conduct to learn about the students’ experiences and thus assure the quality of their mobilities. Keywords Internationalization · International Programmes · Higher Education
1 Overview of the Erasmus Mundus External Co-operation Window (EM ECW) The Erasmus Mundus External Co-operation Window (EM ECW) which has become part of the regular Erasmus Mundus program in the beginning of 2009 is a co-operation and mobility scheme within the area of higher education. The EM ECW was launched by the Europe Aid Co-operation Office in 2006 and has been implemented by the Education, Audiovisual and Culture Executive Agency (EACEA) of the European Union ever since. The aim of the EM ECW is to strengthen the ties between higher education institutions in the European Union and in third countries that are not part of the European Union. Specifically, the EACEA and EM ECW aim “to enable students to benefit linguistically, culturally and educationally from the experience of pursuing academic studies in another country, and to promote European Union (EU) values; to improve the transparency and recognition of studies and qualifications, in particular building on the ‘acquis’ and achievements gained of the Bologna process in this area; to enhance the skills and qualifications of foreign higher education staff so that they can contribute actively towards improvement of quality; to build the capacity of the administration and public and private sector by participation of their staff in higher education mobility activities (especially through doctorate and postdoctorate activities.” [Edub] To achieve this, student and academic staff exchanges are sponsored in order to promote the partnerships and institutional co-operation exchanges between the European and Third Country institutions. The European Commission sponsors these partnerships with grants that partially cover the costs of the organization of mobility of higher education students and academic staff and the costs of the implementation thereof. On the European side, all 27 member states as well as the candidate countries and European Economic Area (EEA) countries are eligible to become partners in this project. Each non-EU country eligible for this project is part of a certain geographical lot which usually consists of countries which lie in close proximity to each other. The partnerships sponsored through the EM ECW need to have at least five European higher education institutions of at least three countries whereas the required number of Third Country participants differs from lot to lot. In the end, a maximum of 20 partners can be involved in each partnership. However, an unlimited number of associates can be added to the project. These associates contribute to the implementation of the mobility scheme but do not receive any funding from the EU through the project.
Experiences from an International Student and Staff Exchange Program
155
The mobility itself is divided into different target groups as well as different individual mobility flows: there are three target groups, i.e. scholars from the partner universities (target group one), scholars from other universities in the Third Countries (target group two) and target group three that consists of scholars in particularly vulnerable situations such as refugees, asylum seekers, or individuals with disabilities. Scholarships vary in length between one month and 34 months and are given to undergraduate, master and doctorate students as well as post-doctorates and academic staff. The project grant consists of flat rates ( C 10.000 per partner for mobility organization costs) and unit costs to cover the individual mobilities. Scholars through EM ECW receive a monthly allowance between C 1.000 and C 2.500. The costs for travelling, health insurance and tuitions fees are also covered by the scholarship. Whereas the first call for proposals in 2006 provided funds with a total of C 36.4 million for projects in nine geographical lots, consisting of altogether 24 Third Countries, the overall budget of projects awarded through the last call in 2008 amounts to C 163.5 million. The call for proposals in 2008 covered 21 geographical lots of more than 50 Third Countries.
2 Implementation of the EM ECW at our University Technische Universität Berlin (TUB) has been a part of the partnership for the geographical lot of Jordan, Lebanon, and Syria since the start of the program in 2006. The calls were renewed in 2007 and 2008. Lunds Universitet is the coordinator of this project, other European partners are Masaryk University in the Czech Republic, University of Granada in Spain, Lille University of Science and Technology in France, University of Bologna and Catholic University of the Sacred Heart in Italy, Vilnius University in Lithuania, KTH (Royal Institute of Technology) in Sweden and University of Zagreb in Croatia. One of the original partners from the United Kingdom was later substituted by University of Leiden from the Netherlands. The European partners are joined by higher education institutions from the Middle East: five Jordanian universities (University of Jordan, Jordan University of Science and Technology, Princess Sumaya University for Technology, Tafila Technical University, and Hashemite University), two Syrian universities (University of Aleppo and University of Damascus) and two Lebanese universities (Lebanese University and American University of Beirut). Table 1 illustrates the number of students as well as home countries that TUB received in the first three phases of the project. The numbers for phase 3 are tentative. As this table illustrates, 18% of all mobilities have gone to TUB in the first three years of the project. When it comes to doctorate candidates, every fourth scholar has been sent to TUB. If one has in mind that TUB is just one of eleven European partners, the predominance of TUB becomes obvious immediately. This discrepancy is more apparent in the first two phases of the project and is being corrected in phase 3 in which only seven scholars are sent to TUB.
Scholars from TUB to Middle East Scholars from EU to Middle East
Call 2007 (Phase 2)
Call 2006 (Phase 1)
17 6 0 0 7
Scholars from EU to Middle East 0 0 0 0 1
Scholars from TUB to Middle East
Call 2008 (Phase 3)
9 7 0 1 3
Scholars from EU to Middle East
4 0 0 0 7 11
Scholars from TUB to Middle East
37 21 3 3 12 76
Scholars from EU to Middle East
Total (Phase 1, 2 & 3)
Table 2 Numbers of Middle Eastern scholars sent to TUB and the European partner universities for the first three phases of the EM ECW program
Experiences from an International Student and Staff Exchange Program 157
158
O. Pfeiffer et al. Table 3 Distribution of scholarship holders at TUB across the disciplines Computer Sciences
Undergraduate Graduate Doctorate Post-Doctorate Academic Staff Total
Engineering
Natural Sciences
Mathematics
Other
Total
3 4 6 1 1
8 4 2 1 4
1 1 2 2 0
0 1 0 1 0
0 1 2 0 0
12 11 12 5 5
15
19
6
2
3
45
Although this partnership came into being due to an earlier cooperation between these universities in the field of ICT, the EM ECW does not focus solely on ICT but was soon opened up to all academic fields. Nevertheless, the better part of all participating students is still from computer sciences and engineering disciplines as Table 3 shows. 42% of the scholars took classes or did research in one of the engineering departments at TUB. The funds for the implementation of the mobility (monthly scholarships, costs for health insurance and travel grants) are transferred to the bank account of TUB where it is the responsibility of the administrative staff to disburse these payments to the respective scholars. Since these administrative tasks were adding to the usual workload of the staff at TUB, an advisor for all matters related to the EM ECW program is now in office. Through this individual, TUB provides all services that are necessary to implement the mobilities, such as issuing visa invitation letters, enrolling the students at the university, arranging for health insurance coverage as well as travel to and from Berlin.
3 Highlights and Potential of the EM ECW The most positive aspects of the EM ECW program are the broadening of the participating scholars’ educations as well as the reevaluation of one’s own culture that occurs when one is living within a different cultural surrounding than the one used to. Students from this EM ECW have an Arabic background which usually also means that they have the Muslim faith. Moving to a western central European country serves as a culture shock for many of them. Not only do they have to learn another foreign language – not all Germans speak English – but they also have to get accustomed to a different way of life. Often, they find a path of their own between the Arabic culture they are used to and the European way of life: A female Jordanian student refused to live in a mixed dorm where female and male students lived. Therefore, she moved into her own apartment which was located off-campus. This move might have taught her a lesson in independence since she would now have to depend solely on her own because she was unable to rely on other students’ living close by.
Experiences from an International Student and Staff Exchange Program
159
Due to participation in the EM ECW program, TUB was able to strengthen its ties to the Middle East. Especially co-operations with Jordan universities have been intensified and TUB is looking forward to closely cooperate with two of the Jordanian EM ECW partner-universities on a Trans-European Mobility Scheme for University Studies (TEMPUS) project within the next year. The challenges of the project refer to the differences between the European and Middle Eastern education system. Whereas the majority of the classes at Middle Eastern universities are taught in Arabic, a substantial numbers of classes are also offered in English. Several students who came to TUB as EM ECW scholars were surprised to find out that about 95% of the programs offered here are taught in German. Currently, there are only six master programs that do not rely on knowledge of the German language. For all others, students need to prove that they know German on a certain level; they even need to pass state-wide examinations such as the German Language Test for Universities (Deutsche Sprachprüfung für den Hochschulzugang, DSH) before they can enter degree programs. Due to this unawareness, several of the scholars attending TUB have to take language classes to improve their German skills. According to Van Damme, “one of the most serious problems in policies and programs aimed at increasing international mobility surely is that of the recognition of study periods and credits obtained abroad. ... The lack of transparency and ‘readability’ of higher education regulations at national, but also at institutional and sometimes even faculty levels creates all kinds of problems, resulting in a widespread uncertainty among students about the recognition of a credit or the study period in the home university. ... Automatic transferability of credits among countries even with a rather similar educational system still is a dream.”[Dir01] This is exactly what we experienced at TUB: a part of the scholarships at the graduate and undergraduate level was granted to credit-seeking students who were essentially exchange students. The rest of the graduate, undergraduate and all PhD scholarships were awarded to degree-seeking students, i.e. those students who came to Berlin not to finish a degree they had started at their home universities, but to start a completely new program. One of the challenges that appeared was that, surprisingly for the Arab students and us, accrediting a degree obtained from some European university turned out to be much easier than getting a credit for a lecture attended at the very same university. The reasons for this are quite obvious: on the one hand, as a consequence of the Bologna Process every European university today uses the ECTS (European Credit Transfer and Accumulation System) as a standard for comparing performance and achievement of students [eur99, Con] while on the other hand, all the partner countries in the Middle East are using the American academic system. Actually, the American University of Beirut is an American university, i.e. the degrees awarded are officially registered by the Board of Education in New York State. Regel claims that “because of its large, flexible, and complex academic system, because English is the main language of communication, because many of the key journals and publishers are in the U.S., and because many scholars and policy-makers have studied in the United States, the American system is a powerful
160
O. Pfeiffer et al.
attraction” which, he later admits, is not readily exportable. [Omp92] Same applies for the ECTS. In this case, however, the ECTS and American credit system collide and the EM ECW scholars are the ones who suffer. While the Middle Eastern countries adopted the American system for the above mentioned reasons, it complicates the EM ECW exchanges to a certain extent. More flexibility of both home and host universities in accepting credits obtained at the other institution would be really helpful for the students and the future of the program.
4 Future Developments of the Program In late 2009 the EMECA (Education, Audiovisual & Culture Executive Agency) [edua] announced changes in the Erasmus program. From 2010 on EM ECW will be part of the Action II of Erasmus Mundus program. Within this change the geographic outreach of the external cooperation will be broadened. More partner countries have been added to the recent call for proposals, e.g. the United States, Canada, and the Gulf countries and TUB is looking forward to being an active partner in Action II.
5 Conclusion TUB has benefitted from its participation in the EM ECW program with Jordan, Lebanon, and Syria, and has been a dedicated partner from the start of the program. However, the financial aspect of the project should not be overlooked: Although the coordinating university receives C 10.000 for each participating university, less than a sixth part is actually transferred to each partner. This share is supposed to cover the administrative expenditure for the duration of this four year program. At the same time, additional costs such as language-classes are to be financed by means of this so called lump sum. In the end, each partner dedicates a lot of manpower for free which is undoubtedly alright to a certain extent but one needs to be aware of these circumstances before the start of the program and joining the partnership. Summarizing, the positive aspects of joining any EM ECW program outbalance the negative financial aspects. Any student who is able to receive a better education or broaden his/her personal horizon due to participating in an exchange and scholarship program is a plus. In the future, though, more attention should be paid to the financing part of the mobilities as well as the obstacles that hinder the transfer of credits between universities in different systems of higher education.
References [Con]
Confederation of EU Rectors Conferences and the Association of European Universities (CRE). The Bologna Declaration on the European space for Education: an explanation. Technical report. http://ec.europa.eu/education/policies/educ/bologna/bologna.pdf, Last accessed on March 19, 2010.
Experiences from an International Student and Staff Exchange Program [Dir01]
161
Dirk van Damme. Quality Issues in the Internationalisation of Higher Education. Higher Education, 41:415–441, 2001. [edua] The Education, Audiovisual and Culture Executive Agency of the European Union (EACEA). Technical report. http://eacea.ec.europa.eu/index.html, Last accessed on March 19, 2010. [Edub] Audiovisual & Culture Executive Agency Education. External Cooperation Window. http://eacea.ec.europa.eu/extcoop/call/index.htm, Last accessed on March 19, 2010. [eur99] European Ministers of Education. The Bologna Declaration of 19 June 1999. Technical report, 1999. http://ec.europa.eu/education/policies/educ/bologna/bologna.pdf, Last accessed on March 19, 2010. [Omp92] Regel Omporn. The Academic Credit System in Higher Education: Effectiveness and Relevance in Developing Countries. The World Bank’s PHREE Background Paper Series 59, pages 1–31, 1992.
A System Architecture for a Telematic Support System in Emergency Medical Services Michael Protogerakis, Arno Gramatke, Klaus Henning
Abstract The system presented in this paper is part of the research project Medon-@ix for the safe application of information technology in preclinical emergency health care. It aims at supporting emergency medical services (EMS) staff at the incident location from a remote Competence Centre (CompC). The increasing number of missions in the German EMS and the shortage of specialized emergency physicians leads to huge problems for the public health care system. Higher cost efficiency and treatment quality shall be achieved by the transfer of mission tactical data and medically relevant data such as vital signs, auscultation and video material from the emergency site to the CompC by the telematic support system. In this paper cases in which such a telematic support system can be used will be outlined. The crucial requirements and a possible hardware and software system architecture of a telematic support system for EMS will be summarised. Keywords Telematic Support · System Architecture · Middleware · Emergency Medical Services · Vital Parameter Transmission
1 Motivation The number of missions handled by the German emergency medical services (EMS) has increased by about 50 % in the last 19 years. In 2004 German EMS handled about 3.6 million incidents. The German EMS regulates the attendance of an emergency physician in the event of severe indications. The number of incidents in which an emergency physician was involved on site has increased from 33 % to approximately half of all operations in the same interval [BBS04]. These figures - as well as the number of false incidents - increased continuously over the last years. The
problem is intensified by a shortage of physicians in German public health care. The need for increased treatment quality and higher cost efficiency is a pan-European phenomenon. Figures presented by Gries/Helm/Martin [GHM03] on the percentages of different procedures in emergency cases suggest that in a maximum of only 15 % of all incidents, the manual abilities of a physician is needed. In contrast, paramedics could handle at least 85 % of all missions, if a physician could transfer his decision competence to the incident location via a telematic support system.
2 Use Cases for a Telematic Support System in EMS The Telematic Support System connects both the emergency site and the CompC and leads to a “virtual presence of the physician” on site. The system supports the staff • at the place of accident (portable) as well as • in an ambulance vehicle during transport of the patient to a hospital (mobile). Figure 1 shows the basic flow of information between the main participants in the system. Medical and mission tactical data is transmitted from the place of
Fig. 1 Scenario overview for Med-on-@ix
A System Architecture for a Telematic Support System in Emergency Medical Services
165
emergency or the ambulance vehicle to the CompC. The CompC can communicate with the staff at the place of emergency via an audio connection. An CompC will also help to optimise workflow by arranging communication with the health care facilities to which patients will subsequently be sent. In the first use case there is no physician on the location and a team of usually two paramedics treat an injured person at the emergency site. With the Telematic Support System the paramedics receive medical advice from the physician in the CompC. The scenario is applicable in cases where only the physicians decisions and no advanced invasive skills are needed on site. In the second scenario a emergency physician at the scene of emergency treats a patient. The CompC supervises the actions and the physician can consult the CompC in case of a seldom disease pattern. For the transport of a vitally stable patient the CompC can monitor the patient on his transport to the hospital. To develop the full benefit of a telematic assistance of EMS it is useful to design the system capable of a roaming of emergency cases between different CompCs. This allows a load balancing between the ressources and as a consequence reduces the number of necessary physicians in the CompCs.
3 Requirements The presented condensed functional and non-functional requirements concerning the system design were found in two expert workshops with physicians and paramedic staff.
3.1 Functional Requirements Table 1 shows the different necessary signals to be transmitted from emergency site as a base for the decision of the physician in the CompC with a classification into continuous live or intermittent transmission and the priority for transmission. No need could be identified by the experts for a more dynamic prioritisation or a control of these priorities by the CompC. All vital parameters shown in Table 1 must be measured by one single monitor/defibrillator device. The device must support the real-time export of all signals except the 12-lead ECG. An electronic stethoscope must allow the live wireless transmission of auscultation. The ambulance car must be equipped with at least one fixed remote control camera to allow the transmission of live video and high quality pictures. A portable camera must provide a video live stream and high resolution steady pictures from the place of emergency. Up to three wireless headsets must be comprehended into the voice communication on the place of emergency. The microphones of the on-site staff can be activated
166
Michael Protogerakis et al. Table 1 Signals to be transferred from site and their priorities Signal 12-Lead-ECG Voice Communication Rhythmn ECG Non-Invasive Blood Pressure Pulse Oximetry Kapnometry Defibrillator Invasive Blood Pressure Central Venous Pressure High Resolution Pictures ≈30 sec Video Sequences Video Stethoscope
by a central control device on site. The quality of the headsets must be of diagnostic quality for auscultation. As an important step towards better quality control the system must feature the documentation of medical and mission tactical data by a software on site and in the competence software [Ber06]. The documented data from the place of emergency must be displayed in real-time in the CompC. The documentation software must receive and display textual commands from the CompC, e.g. for the application of drugs. The documentation software must allow an automatic data import of patient data from the medical devices. Pictures taken with a camera device must be embedded in the documentation. Commercial mobile radio networks are well established and available at reasonable prices in most countries. If they are used redundantly they can be utilised even for safety critical applications to take advantage of the huge capabilities in data transfer compared to Professional Mobile Radio Standards such as Terrestrial Trunked Radio (TETRA). GSM and TETRA can be used for voice communication where they can still deploy their advantage of high availability. All transmitted data from the place of emergency must be archived for later evaluation and quality management. The latter access to (anonymized) archive data must be limited by a appropriate authentication system.
3.2 Non-functional Requirements To be easily adapted to the heavily varyiing local conditions in EMS the system architecture must be configurable in a modular way. This applies especially to medical
A System Architecture for a Telematic Support System in Emergency Medical Services
167
devices and to mobile communication technologies. The main issues concerning safety, security and reliabilty that have to be ensured in the system design are • the safety from interception of patient related data, • the prevention of unauthorized access to the system and • the data privacy of staff and patient related data. Thus all wireless communication must be secured by signing and encrypting the data. A mobile device such as a Tablet PC must be used as the central control unit and for the documentation software on site. The audio transmission of auscultation from the stethoscope must be of diagnostic quality. Therefore the stethoscope must allow the wireless transmission e.g. via Bluetooth with the A2DP standard. The video cameras must feature a resolution from at lest 640x480 pixels and a frame rate of at least 15 frames per second. A change of the frame rate must be possible during online operation. For energy and weight efficiency reasons a fixed camera in the ambulance vehicle must feature a hardware compression of the video stream. A portable camera must either be mounted on a small telescope tripod that is fixed on one of the other units or be a head mounted model. Bluetooth 2.0 must be used for the connection between the communication unit and the headsets. The headsets must feature the HSP/HFP as well as the A2DP profile to listen to the stethoscope signal. The system must provide a middleware solution to ensure its extensibility and adaptability. It must reestablish lost connections due to problems on the underlying communication channels. It must support the abstraction of hardware vendor specific interfaces, such as the real-time interface of the ECG/defibrillator unit, by the means of adapters between the vendor specific device driver and a middleware interface. In the case of insufficient bandwidth it must feature the prioritisation of data according to Table 1. The system must ensure the synchronicity of signals in configurable groups. The communication unit must offer a dedicated IP based packet tunnel and a seperate voice service to the middleware. It must support the use of Professional Mobile Radio services (PMR), such as TETRA and a Circuit Switched GSM mode for voice communication. Common mobile network technologies such as GPRS with EDGE, UMTS with HSPA must be supported through a generic interface for the IP based communication. The unit must offer one tunnel through the parallel channels of different providers. It must uplink a bandwidth information to the middleware.
4 Hardware Distribution and Network Architecture The distribution of hardware components and the physical network architecture is shown in Figure 2. The components on site are connected to the communication unit according to the requirements described above. The communication unit is a
168
Michael Protogerakis et al.
embedded computer which hosts the middleware, the networking logics and the hardware network interfaces. It connects to the ambulance vehicle via an 802.11 network or directly to Public Switched Telephone Network (PSTN) and the Internet through GSM/TETRA and GPRS/UMTS. The same applies to the communication unit in the ambulance. The connection between the On-Site Communication Unit and the Ambulance Vehicle Communication Unit is for redundancy only. Additionally peripherial devices such as a printer are connected to the Ambulance Vehicles Communication Unit by Ethernet. On the CompC’s side the servers are connected to the PSTN via ISDN and to the Internet. The Clients in the CompetenceC connect to the Servers through a local network.
Fig. 2 Hardware Distribution and Network Architecture Overview
A System Architecture for a Telematic Support System in Emergency Medical Services
169
5 Software Architecture 5.1 Architecture Layers On an abstract view, the architecture of the telematic system is divided into the Application Layer, the Session Layer and the Network Layer as shown in Figure 3. The Application and the Session Layer are distributed between the CompC site and On-site respectively the Ambulance Vehicle. Both sides are connected through one VPN Tunnel and Circuit Switched Audio connections in the Network Layer. The Application Layer encapsules the medical devices and other sensors as well as the clients and servers in the CompC. The Session Layer consists out of the Middleware which is responsibile for the session management and the conditioning of all in- and outgoing data. Multiple encrypted VPN Tunnels in the Network Layer are established over multiple concurrent physical links and bonded to one virtual tunnel interface. While techniques for vertical handovers like Mobile IP and the Session Initiation Protocol (SIP) in mobile networks have been well established over the last few years [GV06] the parallel use of multiple links to increase bandwidth is still a challenge [LB06]. The resulting networking properties of the bonded tunnel are determined by the fact that the order of arriving packets on the receiver side is non-deterministic due to the different behaviours of the underlying communication channels. That makes protocols like UDP without robustness against reordering of packets useless. The TCP protocol is much more robust against packet reordering. Bohacek et. al. described a new TCP protocol with high resistance against packet reordering [BHL+ 06]. The Real-Time Transport Protocol (RTP) extends UDP with a reordering robustness and is well suited for real-time streaming applications. ([IAC99], [SCFJ03]) RTP is one of the possible communication protcol used by ZeroC’s Internet Communication Engine (ICE) which was chosen to implement the Middleware. For the Messaging Part buffers are read from a Packer/Priority Calculator that polls data from the buffer according to the Priority Table if there is enough bandwidth and packs data to frames that are transferred through a TCP connection to the Middleware on the receiver side utilizing ZeroC’s Internet Communication Engine (ICE). [VVM+ 07]
Fig. 3 Overview of the different logic layers in the system
170
Michael Protogerakis et al.
5.2 Middleware Architecture Figure 4 shows the most important components of the Middleware on the sender side which is basically divided into a Streaming Part for the transfer of live-data, such as ECG and Video and a Messaging Part for Session Management, transfer of high resolution pictures and Control Purposes. Each signal path is feeded by a Device Driver that abstracts from vendor specific details and converts the incoming data. To achive the integration of services with real-time and synchronicity requirements the Streaming Part must balance different directives: • It must ensure the synchronicity of different signals in “synchronisation groups” such as the different ECG leads. • In the case of an insufficient total available bandwith it should prevent the “stuttering” of the signal output on the receiver side (continuity). The minimal length of continuous data is determined by the minimal length needed for diagnostic purposes. • The Middleware should not present too old data to the receiver side (timeliness). Only the most recent parts of the minimum useful length from the sender’s side must be transmitted. Therefore in order to achieve a trade-off between both continuity and timeliness, streams will be broken down into smallest segments that are still medically interpretable. To achive this, the Streaming Part design contains two buffers and a resampler or recoder in between. A controller calculates the parameters to achive the described balance for the buffer and resampling/-coding parameters.
Fig. 4 Middleware Architecture
A System Architecture for a Telematic Support System in Emergency Medical Services
171
There is one Packer for each synchronisation group of streaming signals. It fetches data from outgoing buffers of all signal paths in one synchronisation group and multiplexes them. Like for the Messaging Part the data is then packed and transmitted via ICE. The transport protocols used here are either TCP or RTP dependent on the data class of the synchronisation group.
6 Outlook and Conclusion Telematic assistance in the safety critical field of EMS can be achived with the already established Professional Mobile Radio in combination with commercial mobile network technologies. The parallel use of multiple physical links adds reliability and can be secured by using VPN mechanisms. A middleware architecture based on this technology is constricted in the choice of usable transport protocols. It must offer different channels for message oriented communication and streaming oriented communication with real-time and synchronicity requirements. For the future the architecture will be extended to fulfil the needs for the interoperability with other CompCs and for the interconnection with hospital information systems.
References [BBS04]
Holger Behrendt, Emil Betzler, and Reinhard Schmiedel. Bedarfplanung im Rettungsdienst - Standorte, Fahrzeuge, Personal, Kosten. Springer, 2004. [Ber06] S. Bergrath. Retrospektive Analyse der Datenqualität eines kommerziellen Datenbanksystems für den Notarztdienst und der Dokumentationscompliance des Notarztdienstes der Stadt Aachen. Phd thesis, 2006. [BHL+ 06] Stephan Bohacek, Joao P. Hespanha, Junsoo Lee, Chansook Lim, and Katia Obraczka. A new TCP for persistent packet reordering. IEEE/ACM Trans. Netw., 14(2):369–382, 2006. [GHM03] A. Gries, M. Helm, and E. Martin. Die Zukunft der präklinischen Notfallmedizin in Deutschland. Anaesthesist, 52:718–724, 2003. [GV06] R. Good and N. Ventura. A multilayered hybrid architecture to support vertical handover between IEEE802.11 and UMTS. In IWCMC ’06: Proceedings of the 2006 international conference on Wireless communications and mobile computing, pages 257–262, New York, NY, USA, 2006. ACM. [IAC99] Sami Iren, Paul D. Amer, and Phillip T. Conrad. The transport layer: tutorial and survey. ACM Comput. Surv., 31(4):360–404, 1999. [LB06] Ji Li and Jack Brassil. On the performance of traffic equalizers on heterogeneous communication links. In QShine ’06: Proceedings of the 3rd international conference on Quality of service in heterogeneous wired/wireless networks, page 33, New York, NY, USA, 2006. ACM. [SCFJ03] H. Schulzrinne, S. Casner, R. Frederick, and V. Jacobson. RTP: A Transport Protocol for Real-Time Applications, June, 2003. [VVM+ 07] F. J. Villanueva, D. Villa, F. Moya, J. Barba, F. Rincón, and J. C. López. Lightweight middleware for seamless HW-SW interoperability, with application to wireless sensor networks. In DATE ’07: Proceedings of the conference on Design, automation and test in Europe, pages 1042–1047, San Jose, CA, USA, 2007. EDA Consortium.
Designing Agile Processes in Information Management Uschi Rick, René Vossen, Anja Richert, Klaus Henning
Abstract Nowadays’ projects have to cope with increasingly dynamic and turbulent environmental conditions [HHL09]. Agile approaches are one possibility to successfully face this challenge. While combining agile with more traditional process models seems to be usual software development practice in industry [HDGZ06], it lacks of scientific reflection. In this paper, an approach for process design is presented that may be used in information management projects and that combines the advantages of agile software development methodologies and those of traditional information management methods. The agile information management provides process designers with a tool suite that consists of roles, values and principles and a set of various methods and that implements iterative and incremental processes in small steps. Early results of a case study confirm the appropriateness of the approach for challenging frequent changes (e.g. due to changing markets, user needs or vague requirements), interdisciplinary cooperation and communication between the involved roles. Keywords Agility · process design · agile information management
1 Introduction In organisations the trend may be witnessed that there is an increasing pressure to be agile with regard to information, knowledge, work and technology; hence the trend goes from planning on a long-term prospect to constant adjustments and realignment [Des07]. In this context, DeMarco [DeM01] argues that nowadays it is rather speed and mobility that matter, than the “right” development of a software product. Downsizing the processes is one method to account for this new mobility, i.e. slenderizing them to so called lightweight or agile processes [SB02]. As of the 1990s [Aue07, HGS09] several of such agile processes have been drafted for U. Rick (B) ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany e-mail: [email protected]
software development to cope with increasing dynamics in a business environment [Coc07, DeC04, BA05]. Today however, information and communication technologies are gaining increasingly more significance in business processes and support organisations to face the growing dynamics [GB03]. To date, there are only a few approaches which present or analyze eligible agile process models for specific use on information management. This paper presents an agile approach that combines classical information management and the advantages of agile software development. The focus is on the design of agile processes that can be used for information management projects and products.
2 State of the Art In recent papers and also in practice, “agile” and “agility” can frequently be found, yet there is no coherent definition. In addition, there are different interpretations of what agile information management or similar terms actually mean. Among these definitions, one finds approaches which primarily relate agility to information or the need for it itself (information centred), such as [BM07]. Other authors [Rou07, Yan06] describe agility as those properties of a system, which is utilized for information management (system centered). Further information management may be used in the environment of agile software development; [Dor04] and [Mel06] are advocates of these. Approaches which try to connect management and technology can be found in [Gal07, LP07] and [GS07] among others. The latter concepts however, are strategical rather than actually suggesting processes. The authors Knublauch [Knu02a, Knu02b] Baumeister [Bau04] and Auer [Aue07] at last, head into a similar direction as the one presented in this paper and apply the agile methods to knowledge management and more specifically to knowledge modeling and engineering. Their ideas are introduced in the following. According to Knublauch [Knu02a, Knu02b] heavyweight process models are often less useful to knowledge-based systems due to the high expenses for change they entail and the creative potential they fail to fully exploit. Knublauch hence aligns values, principles and methods to match the ones of Extreme Programming [BA05] for knowledge- and specifically ontology-based systems. In the case of Baumeister [Bau04, BPS04], capability for smaller teams and vague specifications provide the reason for agile processes in the realm of diagnostic knowledge systems. Following the concepts of Extreme Programming, the authors introduce an agile process model, which allows for early and continuous feedback, incremental planning and a flexible schedule. The methods used are adapted to match the actual case and complemented by an adaptation of refactoring procedures. Auer [Aue07] drafts and discusses an agile methodology for knowledge engineering called RapidOWL in order to render development and usage of knowledge bases even more efficient. This methodology is based upon Knublauch [Knu02a] and the Wiki-Concept [LC01] and is similar to Knoblauch’s approach. The author
Designing Agile Processes in Information Management
175
focuses on agile development of those knowledge bases, where the application is beforehand not known or not easy to determine, and on structured representations as opposed to unstructured textual documents of Wikis [Aue07]. As information management needs to face today’s dynamics and complexity of markets, there is a strong need for agile process models that are applicable to the whole domain of information management, not only to the information or knowledge modelling. Before presenting the approach for such a model, the following chapters describe the underlying concepts of agility and information management.
3 Agility According to the agile manifesto of agile software development [agi01], a consequent shift of values is proposed for information management projects in order to face today’s dynamics and complexity: • • • •
Individuals and interactions over processes and tools Working software (here: running processes) over comprehensive documentation Customer collaboration over contract negotiation Responding to change over following a plan
Fig. 1 Agility zones (according to [Til07])
176
U. Rick et al.
While the secondly mentioned aspects still are of value, the firstly mentioned aspects are more important. Agility is the ability to face growing complexity and dynamics [Til07] and can be seen as a gradual quality [LP07]. According to [Rie92], there are four zones depending on the actual dynamics and complexity (also known as “dynaxity”, see also [Hen09]): static, dynamic, turbulent and chaotic. Each zone requires a higher standard of tasks to be solved and of the involved people the ability to cope with turbulences and thus higher agility capabilities. Figure 1 illustrates the four zones of growing dynamics and complexity as well as the role of agility. Agile software development suggests implementing agility with common values, principles and practices. These are further developed by a team of software developers in order to manage fast deliveries of high quality software and its adoption to permanently moving business goals [DeM01]. One such approach is Extreme Programming according to [Bec99]. Before presenting this idea in the context of information management, the basic definition of information management that is used in the present paper is given in the following chapter.
4 Information Management Information management includes all management tasks within an organisation or another business entity that are concerned with a computer supported or computer supportable information and communication system; this system is developed according to the existing and possible technical support of the tasks to be solved and according to the needs of people that are assigned with these tasks [GB03]. Thus, an information and communication system consists of humans, organisational issues and technology also known as HOT approach [SH92, Mar91]. In case of complex or experience based information, it may be necessary to add non-computer supportable means, e.g. consulting services or workshops with experts, in addition to what is accessible by the computer supported or computer supportable information and communication system. Further, the focus of this paper is on more on processes within projects rather than on whole organisations. According to [Sch98] information management includes the management tasks planning, leading, coordinating and controlling of gathering, processing, transmitting, saving and providing information in order to support the business goals. As leading issues are of importance for all tasks to the authors’ opinions, they are implicitly involved in all tasks. In order to stress the relevance of decision making tasks, the task leading of the definition above is substituted by the decision making [GB03] in the present paper.
5 Process Design in Agile Information Management In this chapter, an approach for designing agile processes for the development of information management products is presented on the basis of the underlying concepts described in the previous chapters. The authors suggest agile processes to be
Designing Agile Processes in Information Management
177
consisting of the roles of people involved, common values, principles and methods used. These components relate to each other and may support each other [Til07]. The processes need to be individually designed according to the tasks, the actual agility zone and the team involved [Til07]. Further, there are different levels of abstraction that need to be taken into account (similar to the levels for tasks and activities in [GB03]): There are the processes of the whole project (macro level), the processes within teams, work packages or sub teams (meso level) and the processes of individuals and single tasks (micro level). The presented approach can be regarded as a toolbox for process designers and managers in the domain of information management. The different components are briefly introduced in the following.
5.1 Roles The allocation of roles to team staff is done according to professional competences and according to personality profiles. In this way, the teams may benefit the most from professional and social cooperation. In order to cope with complex tasks, it is of importance to ensure an interdisciplinary team [TRH06, TPH06] and enhance cooperation and communication [KHHM09]. In this way, different (disciplinary) perspectives may be integrated and may help to manage the challenges provided by complex tasks. Regarding the functional roles in information management projects there are usually the following roles involved: User, domain expert, technical staff, non technical staff (editorial staff etc.) and eventually operator and customer (if different from user). For software dominated projects it is appropriate to have the additional roles of a tester and an architect. Further, it might be useful to distinguish different roles on an organisational perspective: e.g. project manager, work package leader and team members. Finally, there are roles that can be derived from the environment of a project that also have to be taken into account. Examples are project partners or the organisational hierarchies beyond the project officer like a possible chief executive officer (CEO).
5.2 Values and Principles Managing agile processes requires a common system of values and principles that can be seen as a framework in order to ensure high quality teamwork, open minded and close cooperation with users and customers, excellent and rapid results that exactly meet the needs of users and customers. Besides the ideas that are stated in the agile manifesto, the values communication, simplicity, feedback, courage and humility [BA05] are to be implemented in the teams and processes. In addition to these values, several principles are presented in literature that link the abstract values with concrete methods. In the following some of the main principles are highlighted. Teams in an agile information management need to embrace change as it is the competitive edge of users and customers [Bec99]. These are to closely work with
178
U. Rick et al.
the team in order to ensure rapid and direct feedback [Bec99]. Team work is selforganised [agi01] and realised with flat hierarchies in order to allow spontaneous communication, creativity and a high motivation. Further, it is of great importance that the team regularly reflects their own work and processes [BA05].
5.3 Methods While the roles, values and principles mainly are taken from agile software development, the methods used here originate from both areas, software development and information management. They have been adapted to the new application field and new methods have been added. The methods can be clustered according to their strategic or operative dimension and assigned to the management phases planning, decision making, coordinating and control that have been presented above. The presented approach for designing agile processes proposes a toolbox that consists of about fifty methods that can be chosen by a process designer and adapted to the specific needs of a project. These needs can be derived from the tasks, the agility zones and the team involved as described above. Further, the process designer allocates the methods to the different processes (on macro, meso and micro level) and decides about their order of application. It is not possible to describe in-depth all methods in the present paper. Thus, some examples are pointed out. Creativity [PN03] and scenario techniques [PH06] can be used for strategic planning and decision making. Systems of aims and management ratios [GB03] allow a systematic control of the work done. Here it is important to keep the measurements as simple as possible [Bec99] (e.g. deadlines from week to week). Strategic workshops with the whole team seem appropriate to develop common visions and plans, to coordinate the tasks, to reflect on the work done, on team work and processes as well as to control the achieved aims (e.g. in task charts for the whole project). The conjoint work on common aims and plans attributes to the alignment of the different parallel processes and to the motivation of the team. Concerning the operative dimension of information management, weekly team meetings are obligatory for the whole team. In order to ensure efficiency, they are to be kept short. On-site users and customers [Bec99] allow for rapid feedback and spontaneous communication as they are more easily within reach. In this way, the adoption of work to specific needs may be ameliorated. Good team work is very important in an agile information management and supported by common code and document ownership and work in pairs (to ensure high quality and creative results) [Bec99]. All documents and results stick to a simple design and are developed incrementally and iteratively in small releases [Bec99]. This allows for frequent deliveries, rapid feedback and frequent testing (software, system, integration, usability, acceptance etc.). The methods cost accounting and human resource management [GB03] are examples for information management methods that are used for controlling within the operative dimension of the presented approach.
Designing Agile Processes in Information Management
179
6 Case Study The approach for designing agile processes has been developed and used within one of the information management projects that the authors are responsible for. The project is called KISSWIN, an online communication and information platform for young researchers (KISSWIN.de) that is funded by the German Ministry of Education and Research. KISSWIN aims at providing information, promotion and advice for young researchers concerning all aspects of a scientific career (career paths, funding opportunities, funding organizations etc.) in Germany. It provides the researchers with workshops, consulting services, job databases as well as databases providing information on funding opportunities and funding organizations, current news, announcements and events. KISSWIN is particularly suited for a case study of the approach due to the following reasons: The project has to cope amongst others with frequently changing markets, vagchanging markets, vague requirements and requires a very close collaboration with users, customer and domain experts. The various KISSWIN services can be seen as a large scale information and communication management system1 that needs to be implemented in small iterations by an interdisciplinary team. In this context, the means of traditional information management are limited and do not support adequately the need for iterative, flexible and cooperative work. Thus, the approach for designing agile processes within the development of information management products has been implemented within KISSWIN. The results still need to be evaluated in detail but a first insight will be provided here. The approach is well suited for fast and flexible reactions to the market and to user needs as the iterative and incremental development allows for early feedback and testing in small steps. The various principles and methods support the interdisciplinary cooperation and enhance the frequent and early communication between the team and users, customers and domain experts. The following challenges have become obvious while working with the presented approach. First, the flexible approach needs adaption to the specific needs of a project and its team. This requires experienced process designers and project managers. This is important to mention but equally matches to classical information management as well as to classical agile software development processes. Second, it might be hard for the project manager to synchronize the different processes due to their parallelism, phase-delayed implementation and dynamics. This again supports the claim for an experienced leadership person.
1 In this context, the organisation presented in the definition in section IV needs to be substituted by the term project.
180
U. Rick et al.
7 Conclusion The present paper provides an approach for designing agile processes that can be used in information management projects. The agile processes are consisting of the roles of people involved, common values, principles and methods used. The presented tool suite is very flexible and needs to be individually designed according to the tasks, the actual agility zone and the involved team on the one hand and according to different levels of abstraction (project, team and individual) on the other hand. First evaluation results have shown that the design of agile processes as it is presented in this paper is well suited to cope with frequent changes and vague requirements of information management products. Due to its iterative approach, it facilitates early feedback, testing and the communication between all stakeholders. However, it needs to be stated that using the approach requires experienced process designers and managers due to the flexibility of the approach and due to the complexity that is justified by parallelism, phase-delayed implementation and dynamics of processes. The case study and these early results still need to be analyzed and elaborated in more detail. Further, it seems interesting to gain more insights on the combination of agile and more traditional approaches as well as the practical implementation of such combined processes.
Agile Alliance. www.agilemanifesto.org, seen at 19th November 2009, 2001. S. Auer. “Towards Agile Knowledge Engineering,” Methodology, Concepts and Applications. Leipzig, 2007. K. Beck and C. Andres. Extreme Programming Explained. 2nd edition, 2005. J. Baumeister. Agile Development of Diagnostic Knowledge Systems. Berlin, 2004. Beck. Extreme Programming Explained. 1st edition, 1999. N. Bauer and P. Mandl. Agiles Informationsmanagement. Informationsbereitstellung im Unternehmen mit Web2.0. HMD – Praxis der Wirtschaftsinformatik, 255:88 – 96, 2007. J. Baumeister, F. Puppe, and D. Seipel. An Agile Process Model for Developing Diagnostic Knowledge Systems. KI Journal, Special Issue on “AI and Software Engineering”, 3:12 – 16, 2004. A. Cockburn. Agile Software Development. The Cooperative Game. Boston, 2007. D. DeCarlo. eXtreme Project Management. Using Leadership, Principles, and Tools to Deliver Value in the Face of Volatility. San Francisco, 2004. T. DeMarco. Vorwort. In K. v. Beck and M. Fowler, editors, Extreme Programming planen. München, 2001. K. C. Desouza. Preface. In K. C. Desouza, editor, Agile Information Systems. Conceptualization, Construction, and Management. Oxford, 2007. H. D. Doran. Agile Knowledge Management in Practice. In H. v. Holz and G. Melnik, editors, Lecture Notes in Computer Science, volume 3096, pages 137–143. Springer Berlin, 2004. R. D. Galliers. Strategizing for Agility: Confronting Information Systems Inflexibility in Dynamic Environments. In K. C. Desouza, editor, Agile Information Systems. Conceptualization, Construction, and Management, pages 1–15. Oxford, 2007.
Designing Agile Processes in Information Management [GB03] [GS07]
181
R. Gabriel and D. Beier. Informationsmanagement in Organisationen. Stuttgart, 2003. S. Gherardi and A. Silli. Agile Information Systems as a Double Dream. In K. C. Desouza, editor, Agile Information Systems. Conceptualization, Construction, and Management, pages 110–121. Oxford, 2007. [HDGZ06] C. Hansson, Y. Dittrich, B. Gustafsson, and S. Zarnak. How agile are industrial software development practices? The Journal of Systems and Software, 79:1295–1311, 2006. [Hen09] R. Henning, 2009. www.osto.de/fileadmin/Bilder/Publikationen/SYMAartikel.pdf, seen at 30th November 2009. [HGS09] K. Henning, A. Gramatke, and D. Schilberg. Informatik im Maschinenbau. SoftwareEntwurf, -Realisierung und – Management. Aachen, 2009. [HHL09] K. Henning, F. Hees, and I. Leisten. Unternehmenskybernetik 2020 – Betriebswirtschaftliche und technische Aspekte von Geschäftsprozessen. In K. Henning and C. Michulitz, editors, Unternehmenskybernetik 2020. Betriebswirtschaftliche und technische Aspekte von Geschäftsprozessen, pages 25–37. Duncker & Humblot, Berlin, 2009. [KHHM09] S. Kutscha, K. Henning, K. Henning, and C. Michulitz. Mission impossible – Erfolgsfaktoren im Projekt Toll Collect. In K. Henning and C. Michulitz, editors, Unternehmenskybernetik 2020. Betriebswirtschaftliche und technische Aspekte von Geschäftsprozessen, pages 67–78. Duncker & Humblot, Berlin, 2009. [Knu02a] H. Knublauch. An Agile Development Methodology for Knowledge-Based Systems Including a Java Framework for Knowledge Modeling and Appropriate Tool Support. http://vts.uni-ulm.de/doc.asp?id=2101, seen at 17th August 2009, 2002. [Knu02b] H. Knublauch. Extreme Programming of Knowledge-Based Systems. In Third International Conference on eXtreme Programming and Agile Processes in Software Engineering (XP2002), Alghero, Sardinia, Italy, 2002. [LC01] B. Leuf and W. Cunningham. The Wiki Way: Collaboration and Sharing on the Internet. 2001. [LP07] T.-W. Lui and G. Piccoli. Degrees of Agility: Implications for Information Systems Design and Firm Strategy. In K. C. Desouza, editor, Agile Information Systems. Conceptualization, Construction, and Management, pages 122–133. Oxford, 2007. [Mar91] S. Marks. Gemeinsame Gestaltung von Technik und Organisation in soziotechnischen kybernetischen Systemen. Düsseldorf, 1991. [Mel06] G. Melchard. Argumente für ein agiles Wissensmanagement. KM-Journal, 2, 2006. [PH06] L. Petry and K. Henning. Szenariotechnik für eine nachhaltige Mobilitäts- und Verkehrsgestaltung. In F. Wilms, editor, Szenariotechnik. Vom Umgang mit der Zukunft, pages 265–291. Haupt Verlag, Bern, Stuttgart, Wien, 2006. [PN03] P.-B. Paulus and B.-A. Nijstad. Group Creativity: Innovation Through Collaboration. London, 2003. [Rie92] H. Rieckmann. Dynaxibility – oder wie „systemisches“ Management in der Praxis funktionieren kann. In K. Henning and B. Harendt, editors, Methodik und Praxis der Komplexitätsbewältigung, pages 17–39. Duncker & Humblot, Berlin, 1992. [Rou07] W. B. Rouse. Agile Information Systems for Agile Decision Making. In K. C. Desouza, editor, Agile Information Systems. Conceptualization, Construction, and Management, pages 16–30. Oxford, 2007. [SB02] K. Schwaber and M. Beedle. Agile Software Development with Scrum. Upper Saddle River, NJ, 2002. [Sch98] J. Schwarze. Informationsmanagement. Planung, Steuerung, Koordination und Kontrolle der Informationsversorgung im Unternehmen. Herne, Berlin, 1998. [SH92] G. Strina and E. Hartmann. Komplexitätsdimensionierung bei der Gestaltung soziotechnischer Systeme. In K. Henning and B. Harendt, editors, Wirtschaftskybernetik und Systemanalyse, page 169 ff. Berlin, 1992. [Til07] T. Tiltmann. Agile Entwicklung von CSCW-Anwendungen für regionale Bildungsnetzwerke. Aachen, 2007.
182 [TPH06]
[TRH06]
[Yan06]
U. Rick et al. T. Tiltmann, L. Petry, and K. Henning. Der Ingenieur als agiler Projektmanager und der Soziologe als agiler Dienstleister im Softwarewareentwicklungsprozess – Erfahrungsbericht. In B.v. Oestereich, editor, Agiles Projektmanagement. Beiträge zur Konferenz, pages 181–195. dPunkt, Heidelberg, 2006. T. Tiltmann, U. Rick, and K. Henning. Concurrent Engineering and the Dynaxity Approach. How to Benefit from Multidisciplinarity. In P. v. Ghodus, R. Dieng-Kuntz, and G. Loureiro, editors, Leading the Web in Concurrent Engineering - Next Generation Concurrent Engineering. Frontiers in Artificial Intelligence and Applications, pages 488–495. IOS Press, Amsterdam, 2006. L. K. Yan. Testing Agile Information Management Systemt with Video Test Client. Case Study – DIMES,” 2006 CCRTS. The State of the Art and the State of the Practice. 2006.
Kulturveränderung oder kulturbasierte Veränderung? Eine strategische Entscheidung Robert Schmitt, Thilo Münstermann, Klaus Henning, Alexandra Ottong
Zusammenfassung Der Erfolg von Veränderungsprojekten muss für das Management und alle Beteiligten ersichtlich sein. Um die geeigneten Methoden für einzelne Projekte empfehlen zu können, sollten die Unternehmenssituation, der Typ des Veränderungsprojekts und die vorherrschende Unternehmenskultur berücksichtigt werden. Die individuellen Projekttypen beeinflussen das Gewicht einzelner Unternehmenskulturmerkmale, welche sowohl fördernd als auch hemmend sein können. Zusätzlich lassen sich erfolgreiche Veränderungsprojekte durch den Gebrauch der richtigen Methoden zur richtigen Zeit charakterisieren. Eine Toolbox hilft hier bei der Auswahl. Der theoretische Hintergrund fußt auf Kommunikations- und Motivationsaspekten sowie organisationalen Aspekten. Die Toolbox bietet flexible methodische Unterstützung und integriert die Unternehmenskultur. Ziel ist es, Veränderungsprojekte vor dem Hintergrund unterschiedlicher Unternehmenskulturen abzusichern. Schlüsselwörter Unternehmenskultur · Kulturmerkmale · Change-Management · Change-Projekte · Veränderungsprojekte Kulturbasierte Veränderung The success of a change-project must be obvious for management and all participants. To recommend the most appropriate methods for single projects, company specific conditions, the change-type and the prevailing corporate culture should be considered. The individual change-types influence the effect of specific culture characteristics, which may be supportive or inhibitory. Additionally, successful changeprojects are characterized by the use of the right methods at the right time. A toolbox supports in choosing. The theoretical background is based on communicational and motivational aspects as well as organizational phenomena. The toolbox provides flexible methodological support and integrates corporate culture. The aim is to safeguard the quality of change-projects on different corporate cultures.
R. Schmitt (B) Fraunhofer-Institut für Produktionstechnologie (IPT), Steinbachstr. 17, 52074 Aachen, Germany
1 Einleitung Warum gelingen manche Veränderungen scheinbar problemlos und andere haben mit enormen Widerständen zu kämpfen? Viele Antworten auf diese Frage konnten in den letzten Jahren gefunden werden. Transparente Kommunikation, konsequentes Vorleben oder Einbeziehung aller Gruppen von Beteiligten gelten als Erfolgsfaktoren für gutes Change Management. Zudem sind typische Phasen und Abläufe von Veränderung umfassend untersucht und ein breites Spektrum an Methoden vorhanden. Dennoch scheitern viele Change Manager, und das häufig obwohl sie in einem anderen Unternehmen bereits ähnliches erfolgreich umgesetzt haben. Gründe dafür können in der Unternehmenskultur – den impliziten Mustern, nach denen sich Gruppen und ganze Organisationen verhalten – liegen. Bestimmte Methoden passen zu einer Kultur besser als zu einer anderen und die generelle Reaktion einer Organisation auf eine anstehende Veränderung ist längst nicht immer gleich.
2 Stand der Forschung Kultur spielt eine entscheidende Rolle für das Funktionieren einer Organisation. Sie beeinflusst das menschliche Verhalten auf jeder Organisationsebene. Unternehmenskultur bezieht sich auf die fest verankerten Werte, Einstellungen, Glauben und Normen, die die Mitglieder einer Organisation teilen und beinhaltet den unterbewussten Teil des organisatorischen Lebens [SBG96]. Sie manifestiert sich in verschiedenen organisatorischen Faktoren, wie z. B. Organisationsstrukturen, Kontrollsystemen, Symbolen, Routinen sowie Ritualen [Joh92] und ist mit vielen Schlüsselprozessen wie Unternehmens- und Mitarbeiterführung aber auch dem Unternehmenserfolg verbunden [Sch85, CQ06]. So wird seit Jahren kaum bestritten, dass die Unternehmenskultur ein wichtiger Faktor für Change Projekte ist und eine große Rolle bezüglich der Veränderungsfähigkeit einer Organisation einnimmt [Sen00]. Diese wiederum wird beeinflusst durch die Einstellung der Mitarbeiter zu Konflikten, Kritik, dem Teilen von Informationen und Experimentieren in Prozessen und Produkten (vgl. Abb. 1). Kultur bedingt den Grad an Offenheit eines Managements für neue Ideen, an Willen sensitive Aspekte offen zu diskutieren sowie der Autonomie der Mitarbeiter und Unterstützung ihrer Aktionen. Darüber hinaus bestimmt sie das Ausmaß bis zu welchem die Organisationsstrukturen Veränderungen ermöglichen. Demzufolge kann eine Organisation, abhängig von der Kultur, viele verschiedene Veränderungsstrategien beschließen [KE02]. Sobald die Methoden in einem Veränderungsprozess jedoch nicht mit der existierenden Unternehmenskultur vereinbar sind, besteht eine große Gefahr des Scheiterns. Daher ist es ein wichtiges Element eines Change Prozesses die vorherrschende Unternehmenskultur zu verstehen [AC08, SD81]. Dennoch ist der Umgang mit diesem Wissen alles andere als einfach. Kultur ist häufig schwer zu beschreiben, nur langfristig zu verändern und in ihren Auswirkungen kaum einzuschätzen. Es stellt sich also die Frage wie Unternehmenskultur sinnvoll und operativ brauchbar in das Change Management
Kulturveränderung oder kulturbasierte Veränderung?
185
Abb. 1 Organisational culture and change [Sen00]
integriert werden kann. Herangehensweisen an Unternehmenskultur lassen sich grob in zwei Strömungen unterteilen. Die erste bildet Modelle zur Beschreibung und Erfassung von Unternehmenskultur, mit dem Ziel diese möglichst vollständig abzubilden. Daraus sind eine Vielzahl von Ebenen- und Dimensionsmodellen entstanden. Die andere Strömung bilden Ansätze zur Kulturveränderung, welche auf unterschiedliche Weise Unterstützung dabei bieten, von einem kulturellen Ist-Zustand in einen gewünschten Soll-Zustand zu gelangen. Da Unternehmenskultur sich aber in den Einstellungen und Gewohnheiten eines Unternehmens äußert, bedeutet Kulturveränderung nichts anderes, als dass die Einstellungen und Gewohnheiten zahlreicher Mitarbeiter und Führungskräfte verändert werden müssen. Und zwar nach Möglichkeit dauerhaft. Eine solche Kulturveränderung ist schwierig durchzusetzen und häufig ein langwieriger Prozess [Pul00], S. 199 ff., [Sch85], S. 5. Beiden beschriebenen Strömungen ist jedoch gemein, dass implizit oder explizit von einer „Veränderungskultur“ ausgegangen wird, also einem erstrebenswerten Kulturtyp, der jede Form von Veränderung positiv beeinflusst. Vor dem Hintergrund ihrer Wahrnehmung und Beurteilung der Realität und vor dem Hintergrund ihrer persönlichen Ziele, Werte und Interessen, verhalten die Menschen sich ihrer Ansicht nach völlig logisch. Dementsprechend erachten sie auch ihr Handeln im Unternehmensalltag – gestützt von der vorherrschenden Unternehmenskultur – als folgerichtig. Dauerhafter organisatorischer Wandel kann also nur dann erreicht werden, wenn er in der bestehenden Unternehmenskultur verankert wird [Kot96]. Nach Meinung verschiedener Autoren geschieht diese Verankerung der Veränderung in der Kultur durch Veränderung der existierenden Kultur [CQ06, Kot96, Sch85, SBG96]. Durch die feste Verbindung von einer Organisation und ihrer Kultur ist diese jedoch sehr schwer zu verändern. Auch kleine Veränderungen in der Unternehmenskultur bedürfen viel Zeit und je stärker die Kultur ist, umso
186
Robert Schmitt et al.
schwieriger wird es sie zu verändern. [Kot96, Sch87]. In vielen Quellen (z. B. bei [Kot96, Sch87, SD81, See00, Sen00] wurde diese Problematik erkannt und daher versucht eine effiziente Lösung hierfür zu finden. Nach Kotter [Kot96] ist es ein schwerwiegender Fehler mit dem Veränderungsprozess zu starten indem Normen und Werte geändert werden. Die Kultur kann nur dann verändert werden, wenn sich zunächst das Verhalten der Mitarbeiter ändert und diese neuen Verhaltensweisen einen sichtbaren Mehrwert hervorgebracht haben. Hinter dieser Sichtweise verbirgt sich auch Johnsons [Joh92] Begründung für erfolgreiche kulturelle Veränderungen: Das Erschaffen eines geeigneten Veränderungsklimas. Dann ist die Veränderung weit verbreitet akzeptiert und als notwendig für die Organisation angesehen. Schneider et al. [SBG96] sehen es ebenfalls als problematisch an, Unternehmenskultur direkt zu verändern. Sie sind der Meinung, dass kulturelle Veränderungen durch änderung der alltäglichen Gesetze, Praktiken und Routinen (z. B. Organisationsklima) geschehen, welche die Handlungen der Mitarbeiter führen. Da diese greifbarer sind, als Werte und Glauben der Mitarbeiter, sind sie auch leichter zu verändern. Wenn Wandel notwendig ist, muss die Organisation sich vorbereiten, um unvermeidliche Probleme handzuhaben und einige Zeit und Geld in den Prozess zu investieren [Sch87, SD81]. Außerdem erfordert eine erfolgreiche kulturelle Veränderung oft auch eine Veränderung von Schlüsselpersonen in der Organisation [Kot96, Sch87, SD81]. Veränderungen der Unternehmenskultur sollten also eher dann in Betracht gezogen werden, wenn die Kultur sehr schwach ist oder die Zukunft definitiv einen anderen Kulturtyp verlangt. Da vor allem bei kleinen und mittleren Unternehmen (KMU) die verankerte Kultur durch die engen und stark verwurzelten Beziehungen im Unternehmen und die Nähe der Mitarbeiter zueinander als eher stark anzusehen ist, lässt sich vermuten, dass eine Kulturveränderung hier auf massiven Widerstand stoßen wird. Auch wenn die Unternehmensführung entscheidet den Veränderungsprozess in der Kultur zu verankern, indem sie die Kultur verändert, sollte der Veränderungsprozess dennoch mit einer Analyse der existierenden Unternehmenskultur beginnen [Sch87, SD81]. Aufgrund der vielfältigen Schwierigkeiten, die eine Kulturveränderung mit sich bringt und der Tatsache, dass sie das zusätzliche Risiko unerwünschter Nebenwirkungen bietet, wird auch ein anderer Weg zur Vereinbarkeit von Kultur und Veränderung verfolgt. Scholz [Sch87] empfiehlt, die Unternehmens- oder Veränderungsstrategie gemäß der existierenden Kultur zu verändern, anstatt die Kultur selbst zu verändern. Dieser Ansatz zur Anpassung der Veränderung an die Kultur der Organisation scheint sehr vielversprechend, bedenkt man, dass es oft mehrere Wege gibt um die gesteckten Ziele zu erreichen. Maßgeblich für den Start eines Veränderungsprozesses wäre in dem Fall die Evaluierung des Grades zu welchem die existierende Kultur der Organisation die geplante Veränderung unterstützt oder hemmt [Sen00]. Die Identifizierung der Kulturklasse unterstützt die Implementierung der Veränderung zusätzlich, da diese hilft, mögliche Schwierigkeiten durch Unvereinbarkeit zwischen der Organisationskultur und der für eine geplante Veränderung
Kulturveränderung oder kulturbasierte Veränderung?
187
benötigte Kultur hervor zu heben [AC08, SD81]. Das Hauptaugenmerk sollte auf die Aspekte der Kultur gelegt werden, welche essenziell für strategischen Erfolg sind und nicht mit der geplanten organisatorischen Veränderung übereinstimmen [SD81]. Diese macht die neue Darlegung von Veränderungsstrategien einfacher und effektiver. Als Hebel zur Kulturveränderung lässt sich also folgern, dass, wenn die Mitarbeiter eines Unternehmens unter neuen Bedingungen bekommen, was ihnen wichtig ist, die meisten von ihnen relativ zügig auch ihr Verhalten ändern werden.
3 Ansatz zur kulturbasierten Veränderung Im Gegensatz zu den Versuchen und Ansätzen eine Kultur dahin gehend zu verändern, dass sie gewünschte Effekte erzielt oder Strategien unterstützt, wurde im Forschungsprojekt „Culture Based Change“ eine Vorgehensweise zur kulturbasierten Veränderung entwickelt. Wie können Veränderungen kulturgerecht, also angepasst an die bestehende und kurzfristig kaum veränderbare Unternehmenskultur, gestaltet werden? Ziel ist es also nicht, die Unternehmenskultur zu beeinflussen, sondern vielmehr sie angemessen zu berücksichtigen und im Sinne eines angepassten Methodeneinsatzes zu nutzen. Davon ausgehend, dass die meisten Unternehmen mit ihren Produkten oder Dienstleistungen erfolgreich am Markt sind, ist jede umfassende Kulturveränderung außerdem eine Gefahr für die Grundlage des bisherigen Erfolgs. Um den Erfolg von Veränderungsprojekten zu steigern wird daher vorgeschlagen, verstärkt das Change Management an die Kultur anzupassen, als umgekehrt die Kultur zu verändern. Es wird auf keinen Fall negiert, dass jede Veränderung auch die Unternehmenskultur beeinflusst, sondern lediglich dafür plädiert, den Fokus der Veränderung anders anzulegen. In solchen Fällen, in denen sich Rahmenbedingungen verändert haben und die Kultur explizit verändert werden soll, kann das hier beschriebene Vorgehen keine Hilfe bieten. Es wird aber davon ausgegangen, dass es eine Vielzahl von Veränderungsprojekten gibt, in denen nicht die Kultur direkt, sondern Strukturen und Prozesse verändert werden sollen. In diesem Sinne wäre die Veränderung der Kultur eine Folge des Change Prozesses und nicht die Ursache. Der Grund für die Bedeutsamkeit der Analyse der existierenden Unternehmenskultur ist das Versagensrisiko, falls die Praktiken während des Veränderungsprozesses nicht mit der existierenden Unternehmenskultur vereinbar sind. Die Veränderung wird höchstwahrscheinlich nicht gelingen, wenn kulturelle Normen durch Veränderungsstrategien verletzt werden [Sch85]. Daher ist die Analyse der Unternehmenskultur ein Herzstück von Veränderungsprozessen [AC08] und erfolgt innerhalb der ersten Phase im Veränderungsprozess. Praktisch noch bevor die tatsächliche Veränderung beginnt. Außerdem kann die Berücksichtigung der Rolle der Unternehmenskultur schon während der Planungsphase der Veränderung die Implementierung dieser unterstützen und eventuell sogar verkürzen. Häufig werden Veränderungsprozesse, gerade bei kleinen und mittleren Unternehmen aufgrund von begrenzten personellen und finanziellen Ressourcen mehr
188
Robert Schmitt et al.
Abb. 2 Häufige Umsetzungsbarrieren bei Veränderungsprojekten: Insbesondere weiche Faktoren dürfen bei Veränderungen nicht vernachlässigt werden (Befragt wurden Verantwortliche für Projektmanagementaktivitäten, Geschäftsführer sowie Mitarbeiter in leitenden Positionen, [bea07])
oder weniger unsystematisch vorangetrieben. Oft sind die Ziele des Vorhabens klar, jedoch nicht das Vorgehen, um diese zu erreichen. Sogar der umgekehrte Fall kann eintreten, indem Maßnahmen eingeleitet werden ohne dass ein Ziel festgelegt wurde [Pul00]. Vor allem bei kleinen und mittleren Unternehmen sind erzwungene Veränderungen der Unternehmenskultur schwer durchsetzbar. Neue Denkweisen zu erzeugen und somit die Unternehmenskultur zu verändern, wie dies bei Großunternehmen der Fall ist, wäre gegen die Funktionsweise der meisten KMUs. Die Nichtbeachtung derartiger Rahmenbedingungen liegt darin begründet, dass die so genannten weichen Faktoren nicht ohne weiteres ermittelbar oder auch nicht bekannt sind [Str06] (vgl. Abb. 2). Dass diese Faktoren im industriellen Alltag werden zu gering wahrgenommen, ist teilweise auf die Technik fokussierte Ausbildung der dort tätigen Ingenieure zurückzuführen, die sich vornehmlich mit harten Faktoren ¨ auseinandersetzen [B97]. Diese Konstellation führt zu dem Phänomen, dass häufig das mangelnde Engagement der von Change Prozessen betroffenen Mitarbeiter oder eine untransparente Kommunikation beklagt wird und gleichzeitig eine Ratlosigkeit vorherrscht, wie mit Widerständen in einem Change Vorhaben umzugehen ist. Es liegen zahlreiche Methoden zur Durchführung von Change Vorhaben vor, jedoch sind sie ohne die Berücksichtigung unternehmensspezifischer Kultur nicht Ziel führend einsetzbar. Das Commitment der Mitarbeiter zur Veränderung und die zielgerichtete Anwendung der Methoden durch die Mitarbeiter werden von den weichen Faktoren der Unternehmenskultur wesentlich mehr beeinflusst als durch die harten Faktoren wie Prozesse und Strategien. Zugleich existieren Typologien und Erfolgsfaktoren von Change Projekten, die es ermöglichen, bestimmte Zusammenhänge zu erkennen. Diese werden jedoch nicht in Form von konkreten Wirkgefügen dargestellt, als dass sie zur Steuerung von Change Vorhaben eingesetzt werden könnten. Das hier beschriebene Vorgehen setzt sich damit auseinander, insbesondere KMU Werkzeuge an die Hand zu geben, mit denen Sie diagnostizieren können, wie ihre Unternehmenskultur beschaffen ist und wie sich diese auf anstehende Change Vorhaben auswirken kann.
Kulturveränderung oder kulturbasierte Veränderung?
189
4 Vorgehensmodell zur kulturbasierten Veränderung für KMU Die vorgestellte Strategie der Kultur basierten Veränderung verfolgt die Arbeitsschritte „Identifikation der geplanten Veränderung“, „Analyse der Unternehmenskultur“ und „Ableiten von Methodenempfehlungen“ (vgl. Abb. 3), welche im folgenden kurz vorgestellt werden. Sie ermöglicht somit eine logische Abfolge von Arbeitsschritten und eine intuitive Vorgehensweise im Veränderungsprozess. Identifikation der geplanten Veränderung Eine Veränderung ist nicht gleich Veränderung. Der Begriff Veränderungsmanagement suggeriert ein einheitliches Fachgebiet, doch verbergen sich dahinter völlig unterschiedliche Problemstellungen, welche nur einen gemeinsamen Nenner haben: Es geht um eine Veränderung, von der eine Vielzahl von Mitarbeitern betroffen sein wird. Der erste Schritt in einem Veränderungsvorhaben ist daher, die Art der geplanten Veränderung zu identifizieren. Je genauer sich einschätzen lässt, welche Reaktionen das Vorhaben auslösen wird, umso besser lassen sich die erforderlichen Methoden auswählen. Unterschiedliche Veränderungen beinhalten in den meisten Fällen auch unterschiedliche Konfliktpotenziale. Neben den unterschiedlichen Auslösern für eine Veränderung, wie neue Technologien, Kunde und Markt, Shareholder sowie Normen und Gesetze, kann auch der Veränderungsaufwand zur Einteilung der Veränderungsvorhaben dienen. So erstrecken sich die unterschiedlichen Arten einer Veränderung über punktuelle Eingriffe zur schnellen Anpassung einzelner Teile im Unternehmen über drastische Senkungen der Kosten durch Stellenabbau und
Abb. 3 Strategie der Kultur basierten Veränderung
190
Robert Schmitt et al.
Outsourcing bis hin zur vollständigen Neuorientierung, dem sorgfältig geplanten Umbau des Unternehmens zur Sicherung der überlebensfähigkeit. Im Rahmen der hier vorgestellten strategischen Unterstützung zur Veränderung erfolgt die Einteilung der Veränderungsvorhaben nach der sogenannten Typologie der Veränderung. Dies erfolgt mit Hilfe von alltäglichen Begriffen, die leicht verständlich sind und in welche die einzelnen Unternehmen sich leicht einordnen können. Es werden folgende Change Typen betrachtet: • • • • • •
Personalabbau, Kostensenkungsprogramme, Produktivitätssteigerung, Outsourcing Reorganisation/Restrukturierung und Einführung neuer IT.
Wie erwartet zeigt eine im Rahmen des Forschungsprojektes „Culture Based Change“ durchgeführte Studie deutliche Zusammenhänge zwischen Unternehmenskulturmerkmalen und dem Erfolg der unterschiedlichen Change Typen. Darüber hinaus zeigen die einzelnen Change Typen zum Teil auch einen deutlichen Einfluss auf die Erfolgswirkung einzelner Kulturmerkmale. Einige Kulturmerkmale haben außerdem über mehrere Change Typen hinweg einen starken Zusammenhang mit dem Projekterfolg bewiesen. Es erweist sich somit als strategisch ratsam, die geplante Veränderung vorab einer der Typologien zuzuordnen. Analyse der Unternehmenskultur Der strategische Umgang mit kulturellen Fragen ist schwierig, da Unternehmenskultur ein großes und wenig klar abgegrenztes Feld organisationaler Phänomene bildet. Bestehende Kulturmodelle versuchen diese meist anhand von Dimensionen möglichst vollständig zu erfassen. Diese Herangehensweise ist aus einer forschenden, erkenntnisorientierten Sichtweise richtig und wichtig, bietet aber kaum pragmatische Ansatzpunkte, da die Dimensionen notwendigerweise sehr abstrakt sein müssen, um Kultur umfassend abbilden zu können. Mit Hilfe von vordefinierten Unternehmenskulturmerkmalen lässt sich Unternehmenskultur greifbar beschreiben. Als Merkmale einer Unternehmenskultur werden einzelne Teilaspekte verstanden, die für sich alleine einschätzbar sind und einen Zusammenhang mit der Gelingenswahrscheinlichkeit von Change Projekten aufweisen. Diese Merkmale erheben hingegen nicht den Anspruch, die Unternehmenskultur vollständig abzubilden. Beispiele für Kulturmerkmale können sein, ob Mitarbeiter eines Unternehmens sich duzen oder siezen, wie Führungskräfte auf Fehler reagieren oder auch architektonische Merkmale, wie etwa die Verfügbarkeit von Kaffee-Ecken. Als strategischer Umgang mit Unternehmenskulturmerkmalen wird vorgeschlagen, einerseits die individuell spezifischen relevanten Merkmale einer Organisation zu ermitteln und andererseits die für den identifizierten Veränderungstyp typischerweise fördernden und hemmenden Kulturmerkmale auf ihre Ausprägung in der Organisation zu überprüfen. Mit Hilfe eines im o. g. Forschungsprojekt entwickelten Verfahrens ist es möglich, anhand vergangener Change Projekte die für Veränderung erfolgskritischen
Kulturveränderung oder kulturbasierte Veränderung?
191
Kulturmerkmale einer Organisation individuell zu ermitteln. Möglich wird dies durch eine Modifikation der bewährten OSTO Systemdiagnose [Ise94, Hen05]. Diese Organisationsdiagnose hat die Besonderheit, dass sie Unternehmen anhand ihrer konkreten und sichtbaren Outputs „aufrollt“ und so nicht die postulierten Ziele und Strategien, sondern das tatsächliche Organisationsverhalten untersucht. Diesem Prinzip bedient sich auch die CuBa Change Diagnose. Um für Change-Projekte relevante Kulturmerkmale zu erhalten, werden nicht die Outputs des gesamten Unternehmens, sondern die eines konkreten Change-Projektes betrachtet [IFM09]. Es kann jedoch noch nicht unmittelbar auf die dahinterliegenden Kultur geschlossen werden. Outputs entstehen durch Verhalten der Organisationsmitglieder, so dass zunächst das zugrunde liegende Verhalten ermittelt wird. Verhalten wiederum darf keinesfalls mit Kultur gleichgesetzt werden. Kultur beschreibt Muster und Gemeinsamkeiten im Verhalten mehrerer Menschen und ihre überzeugungen. Einmalige Verhaltensweisen einzelner dürfen also nicht mit Kultur verwechselt werden. Die identifizierten Verhaltensweisen werden danach untersucht, ob sie typisch für Personengruppen oder die ganze Organisation sind bzw. ob das Verhalten durch gemeinsame Werte und überzeugungen der Beteiligten motiviert ist. Wird dies gefunden, zeigt sich eine klare Linie von Kulturmerkmalen über Verhalten hin zu den konkreten Auswirkungen des Projektes auf. So ist sichergestellt, dass genau jene Kulturmerkmale identifiziert werden, welche in dem betreffenden Unternehmen das Verhalten in Veränderungsprozessen in der Vergangenheit entscheidend beeinflusst haben. Weiterhin wurden die in der o. g. Studie gesammelten Erkenntnisse zur fördernden und hemmenden Einfluss von Kulturmerkmalen in einer Korrelationsmatrix zusammengetragen. Mit Hilfe eines in 2010 in der Schriftenreihe der FQS erscheinenden Anwenderleitfadens können KMU eigenständig die relevanten Kulturmerkmale untersuchen. Dazu bietet der Anwenderleitfaden verschiedene Varianten mit unterschiedlicher Intensität und Arbeitsaufwand, die auch durch ein Softwaretool unterstützt werden. Dies bietet insbesondere KMU den Vorteil, den Aufwand der Kulturanalyse an den Umfang und das Budget des Projektes anpassen zu können. Bestehende Ansätze hingegeben sind häufig mit umfangreichen Beratungsleistungen verbunden und daher für KMU bei den meisten Projekten zu kostspielig. Nach einer Bewertung der relevanten Kulturmerkmale lassen sich zum einen direkt die erfolgreichsten Methoden ermitteln. Zum anderen ist aber auch die Identifikation einer Kulturklasse möglich. Hier sind z. B. die Kulturdimensionen nach Hofstede [HH05] zu nennen. Diese beziehen sich auf Machtdistanz, Individualismus/Kollektivismus, Unsicherheitsvermeidung, Feminität/Maskulinität sowie einer Kurzzeit-/Langzeitorientierung. Bezüglich der Dimension Unsicherheitsvermeidung wird beispielsweise die Aussage getätigt, dass bei einer geringen Ausprägung die Veränderungsbereitschaft eher vorhanden ist als bei einer hohen Ausprägung. Das Zusammenspiel von Kulturklasse, Unternehmens- und ChangeTyp lässt also eine weitere Empfehlung bezüglich der Veränderungsstrategie zu. Diese bewegt sich zwar auf einer weitaus allgemeineren Ebene, als die direkte Empfehlung einzelner Methoden, kann der Veränderung an sich aber eine zusätzliche Grundlage bieten.
192
Robert Schmitt et al.
Ableiten von Methodenempfehlungen Den einzelnen Merkmalen der Unternehmenskultur wird in diesem Vorgehensmodell eine unterstützende bzw. hemmende Wirkung bei der Anwendung ausgewählter Methoden des Veränderungsmanagements attestiert. überspitzt bedeutet dies, dass je weiter ein Veränderungsprojekt fortschreitet, es umso wichtiger wird, von Seiten des Top Managements auch die Gefühle zu führen. Dies verlangt eine Empfehlung einzelner, das Veränderungsprojekt unterstützender Methoden. Die für dieses strategische Vorgehenden resultierenden Methodenempfehlungen basieren auf einer von Bisenius [Bis03] entwickelten Systematik zur Gestaltung strategischer Change Prozesse, welche sich an fünf Phasen orientiert, wobei die fünfte Phase die kontinuierliche Rückführung und Verbesserung einschließt. Veränderungen und damit verbundene Veränderungsvorhaben werden häufig und fälschlicherweise als einmalige Prozesse verstanden. Tatsächlich handelt es sich jedoch um kontinuierliche Prozesse, weswegen das Vorgehensmodell einen Regelkreis darstellt. Die fünf ausgeprägten Phasen unterscheiden sich in Entscheidung zur Veränderung, Vorbereitung, Gestaltung, Umsetzung und Absicherung der Veränderung. Konzept und Umsetzung eines Veränderungsvorhabens lassen sich nicht scharf trennen. Der Grund für diese Planungsunschärfe sind die auf weichen Faktoren beruhenden Unwägbarkeiten der Organisation. Getrennt werden die Phasen durch Quality Gates, an denen die Reife paralleler Schritte der betreffenden Phase bewertet werden. Checklisten für die überprüfung der jeweiligen Phasen wurden erstellt, um die Strategieentwicklung abzusichern. Voigt [Voi06] fügte dieser Systematik einen Methodenbaukasten in Form eines Leitfadens hinzu, so dass auch kleinen und mittelständischen Betrieben die Möglichkeit gegeben ist, Methoden des Veränderungsmanagements selbstständig anzuwenden. Die Erfolgsmessung spielt hier eine wesentliche Rolle, die Auswahl der Methoden ist jedoch nicht begründet. Weiche Faktoren werden in Form von Kommunikation und in groben Ansätzen von Motivation berücksichtigt, dennoch wird kein Bezug zur Unternehmenskultur und dem damit einher gehenden Verhalten und Wertegefüge der Mitarbeiter hergestellt. Die bis heute entwickelten Methoden bieten Anhaltspunkte, welche Maßnahmen in den jeweiligen Phasen von Change Projekten eingesetzt werden können, um zum Beispiel mit möglichen Hindernissen umzugehen. Die bisher postulierten Systematiken liefert aber keine Auswahlhilfe dafür, in welchen Unternehmen und bei welchen kulturellen Rahmenbedingungen gezielt Methoden einzusetzen sind. Die bisherigen Hinweise sind lediglich zeitlicher Natur und beziehen sich auf die einzelnen Phasen des Change Prozesses. Um diesen Missstand zu beheben wird dieses Vorgehensmodell nun um die Kulturanalyse erweitert (vgl. Abb. 3). Fokussiert wird dabei auf die Ausrichtung der Merkmale der Unternehmenskultur, die flexible methodische Unterstützung von Veränderungen sowie die Berücksichtigung der unternehmensspezifischen Situation. Zusammenfassung und Ausblick Unternehmen benötigen ein Instrument in Form einer Entscheidungsmatrix, um identifizieren zu können, welche Methoden zur Steuerung von Change Prozessen
Kulturveränderung oder kulturbasierte Veränderung?
193
vor dem Hintergrund ihrer speziellen Rahmenbedingungen einzusetzen sind. Ziel ist es, die Qualität der Change Prozesse vor dem Rahmen unterschiedlicher Unternehmenskulturen abzusichern. Welche Kulturmerkmale im Einzelnen sinnvoll sind, wurde im Rahmen der Strategieentwicklung durch eine Hypothesen geleitete Studie und anhand von Fallbeispielen untersucht. Die Besonderheiten dieses Modells liegen in der Kultur basierten Betrachtung von Veränderungsvorhaben, in der flexiblen methodischen Unterstützung sowie der Berücksichtigung der individuellen Risiken. Das Ergebnis ist ein Leitfaden, wann welche Methode des Veränderungsmanagements wie und zu welchem konkreten Zweck eingesetzt werden kann. Hinzu kommt die individuelle Berücksichtigung der Unternehmenskultur. Durch den Einsatz des Kultur basierten und Methoden gestützten Vorgehens, wird das Risiko bei Veränderungsprozessen minimiert. Es wurden spezifische Kriterien zur näheren Bestimmung der kulturellen Einflussfaktoren festgelegt. Für jeden einzelnen definierten Einflussfaktor werden aus allen prinzipiell in Frage kommenden die für die betrachtete Veränderung relevanten Methoden ausgewählt. Hinweis zum Forschungsprojekt Das IGF-Vorhaben (15593 N/1) der Forschungsvereinigung Forschungsgemeinschaft Qualität e.V. (FQS), August-Schanz-Str. 21A, 60433 Frankfurt am Main wurde über die AiF im Rahmen des Programms zur Förderung der industriellen Gemeinschaftsforschung und –entwicklung (IGF) vom Bundesministerium für Wirtschaft und Technologie aufgrund eines Beschlusses des Deutschen Bundestages gefördert. Die durchführenden Forschungsstellen sind das Fraunhofer-Institut für Produktionstechnologie (IPT) und das Institut für Unternehmenskybernetik an der RWTH Aachen (IFU).
Literaturverzeichnis [AC08]
Michael J Austin and Jennette Claassen. Impact of organizational change on organizational culture: implications for introducing evidence-based practice. Journal of EvidenceBased Social Work, 5(1-2):321–359, 2008. ¨ [B97] K. Büchter. Klein- und Mittelbetriebe: Träger des Strukturwandels? Relativierungen zu Prosperität und Qualifikationsbedarf. Zeitschrift für Arbeitsforschung, Arbeitsgestaltung und Arbeitspolitik, 6(4):412–428, 1997. [bea07] Status-Quo des Change Managements in Deutschland, 2007. http://www.bearingpoint.de. [Bis03] A. Bisenius. Systematik zur qualitätsgerechten Gestaltung und Absicherung strategischer Veränderungsprozesse. PhD thesis, RWTH Aachen University, 2003. [CQ06] Kim S. Cameron and R. Quinn. Diagnosing and Changing Organizational Culture. Jossey-Bass, San Francisco, CA, USA, (Revised edition) edition, 2006. [Hen05] R. Henning. OSTO Systemdiagnose, Leitfaden zur systemischen (und systematischen) Diagnose komplexer Systeme und eigener Verantwortungsbereiche aus der Hubschrauberperspektive. Technical report, OSTO Systemberatung GmbH, November 2005. [HH05] Geert Hofstede and Gert Jan Hofstede. Cultures and Organizations - Software of the Mind: Intercultural Cooperation and Its Importance for Survival. Mcgraw-Hill Professional, New York, (Revised and expanded 2nd edition). edition, 2005.
194
Robert Schmitt et al.
[IFM09] Ingrid Isenhardt, Paul Flachskampf, and Thilo Münstermann. Auf dem Weg zum kulturbasierten Change. managerSeminare Das Weiterbildungsmagazin, 135, 2009. [Ise94] Ingrid Isenhardt. Komplexitätsorientierte Gestaltungsprinzipien für Organisationen dargestellt an Fallstudien in Reorganisationsprozessen in einem Großkrankenhaus. Augustinus Verlag, 1994. [Joh92] G. Johnson. Managing strategic change - strategy, culture and action. Long Range Planning, 25:28–36, 1992. [KE02] Adrianna J. Kezar and Peter D. Eckel. The Effect of Institutional Culture on Change Strategies in Higher Education: Universal Principles or Culturally Responsive Concepts? The Journal of Higher Education, 73(4):435–460, 2002. [Kot96] J. Kotter. Leading Change. HBS Press, Boston, MA, 1996. [Pul00] K. Pullig. Innovative Unternehmenskulturen : zwölf Fallstudien zeitgemässer Sozialordnungen. Rosenberger Fachverlag, Leonberg, 2000. [SBG96] B. Schneider, A. Brief, and R. Guzzo. Creating a climate and culture for sustainable organizational change. Organizational Dynamics, 24:7–19, 1996. [Sch85] Edgar H. Schein. Organizational Culture and Leadership: A Dynamic View. Jossey Bass, San Francisco, CA, USA, 1 edition, 1985. [Sch87] C. Scholz. Corporate culture and strategy - The Problem of Strategic Fit. Long Range Planning, 20:78–87, 1987. [SD81] H. Schwartz and S. Davis. Matching corporate culture and business strategy. Organizational Dynamics, 10:30–48, 1981. [See00] R. Seel. Culture and complexity: New insights on organisational change. Organisations and People, 7:2–9, 2000. [Sen00] B. Senior. Organizational change and development. In N. Chmiel, editor, Introduction to Work and Organizational Psychology. A European Perspective, pages 347–383. Oxford: Blackwell Publishing, 2000. [Str06] G. Strina. Zur Messbarkeit nicht-quantitativer Größen im Rahmen unternehmenskybernetischer Prozesse. PhD thesis, RWTH Aachen University, 2006. [Voi06] T. Voigt. Systematik zur qualitätsgerechten Umsetzung organisatorischer Veränderungsprozesse. PhD thesis, RWTH Aachen University, 2006.
Network Management for Clusters of Excellence - A Balanced-Scorecard Approach as a Performance Measurement Tool Florian Welter, René Vossen, Anja Richert, Ingrid Isenhardt
Abstract Supplementary Cluster Activities constitute an important organisational part within the structure of the German Cluster of Excellence “Tailor-Made Fuels from Biomass” at RWTH Aachen University, because they focus on the entire clusters efficient networking process and successful strategic cluster development. As research teams from different scientific fields collaborate, the strategic management of interdisciplinary processes becomes necessary to enhance scientific cooperation. Therefore, amongst other measures of cluster development, a Balanced-ScorecardApproach is implemented to measure the performance of the entire Cluster of Excellence. With the annual implementation of the Balanced-Scorecard-Approach, crucial key performance indicators have been collected, compared and analysed for the strategic management of the Cluster of Excellence, to facilitate innovation activities through adequate measures. Keywords Scientific Networks · Management · Performance Measurement · Interdisciplinarity
1 Introduction Since 2006, huge financial investments in science have been made, due to the efforts of the German Federal Government, to promote excellent university institutions in Germany. The initiation and development of Clusters of Excellence were conducted in this context. Clusters of Excellence are scientific networks with heterogeneous partners of scientific institutions which follow a common vision. During the first
five years of the funding period, each Cluster of Excellence was financed with approximately 40 million Euros. The case study of this paper analyzes the Cluster of Excellence “Tailor-Made Fuels from Biomass” (TMFB) at Rheinisch-Westfälische Technische Hochschule (RWTH) Aachen University, which follows a common vision of developing sustainable tailor-made biofuels for clean combustion. In this context, scientists from engineering, chemistry, biology and sociology work together to explore solutions using a holistic approach. The fact that such a heterogeneous consortium of 20 university institutions, two non-university research institutions, and nineteen scientific and industrial advisors needs to be managed with adequate networking measures has already been taken into account before the establishment of the TMFB in November of 2007. Hence, the organisational structure of the TMFB includes the Supplementary Cluster Activities to support networking activities among its participants, located regionally, as well as globally, to monitor and enhance the strategic development of the entire network. Due to their overall function within the Cluster of Excellence TMFB, Supplementary Cluster Activities are aimed at the efficient networking of scientific processes. Therefore, a conceptual framework has been developed, considering closer cooperation among single cluster projects and a closer connection and exchange between all Integrative Research Fields (IRFs) and Core Interaction Fields (CIFs). The IRFs and CIFs are the scientific pillars in the overall organisational structure of the Cluster of Excellence, the Fuel Design Center (cf. Fig. 1). The biggest challenge in facilitating cooperation in the TMFB lies in meeting the demands of the different specialists, like engineers, chemists and biologists, located in different scientific pillars of the research cluster. Only if this challenge is overcome, will the entire cluster performance and scientific output be improved [Sau05]. Because of this, the enhancement of the satisfaction of all employees within the network constitutes an important aspect conducted by the Supplementary Cluster Activities, as well as the
Fig. 1 The Fuel Design Center and the structure of the Supplementary Cluster Activities [Uni]
Network Management for Clusters of Excellence
197
initiation of the learning processes and knowledge management within the cluster by a continuous reflection of all networking activities. The Supplementary Cluster Activities approach consists of three main tasks: Knowledge Management and Dissemination, Personal Development and Benchmarking (cf. Fig. 1). The focus of the paper is on the Knowledge Management and Dissemination task, which contains the implementation of a cluster-specific Balanced-Scorecard-Approach. By explaining this approach in detail, the necessity of measuring the network performance shall become obvious for successful strategic cluster development.
2 Overview of the Supplementary Cluster Activities One of the three main tasks belonging to the Supplementary Cluster Activities deals with Benchmarking. Instead of the other main tasks, all managed by the Center for Learning and Knowledge Management and the Department of Information Management in Mechanical Engineering at RWTH Aachen University, the Öko-Institut e.V. – Institute for Applied Ecology in Germany, will economically and ecologically benchmark the newly developed methods of the TMFB, in comparison to the existing processes or such underdevelopment [Uni]. Another main task - Personal Development - is divided into the sub-tasks Education and Lifelong Learning, Promotion of Young Researchers and Promotion of Equal Opportunities (cf. Fig. 1). An aim of the main task of Personal Development is to enhance the interpersonal exchange and networking among the staff of the TMFB. Due to the fact that scientists from different institutes and departments collaborate within the Cluster of Excellence, it is very important for a successful cluster development to arrange opportunities to discuss, or present, achieved milestones regularly. Because of that, strategy workshops for professors and leading researchers, or colloquia, for research assistants and PhDs are conducted once a year to foster scientific networking activities. For example, the intention of the colloquia for research assistants is the definition of the IRF- and CIF-specific objectives by all participants, as well as the placement of each objective into the entire scientific process of the cluster. By conducting this step, the potential of synergies within the cluster become clarified. A subsequent presentation of the content of each IRF and CIF for all participants enables everybody to inform themselves about details and ask further questions. Apart from the scientific exchange and the social aspects – the goal of the fostering of socialising and networking among the employees of the cluster can be reached by the conduction of regular colloquia. This is especially important during the initial phase of the Cluster of Excellence where socialising is crucial, because the scientists become acquainted with each other. Thus one can state that in this phase of cluster development the basis for the trust and cooperation in the forthcoming cooperationphases is generated [Ahr04, ACM06, Sau05]. In the sub-task Education and Lifelong Learning, the Supplementary Cluster Activities are setting up an advanced training programme, consisting of seminars and workshops for the staff of the TMFB. This programme enables employees within
198
Florian Welter et al.
the cluster to enhance their methodological, rhetorical and/or didactical skills. It supports this by the development of excellent scientific skills. Furthermore, not only are the contemporary employees of the TMFB in the focus of the Supplementary Cluster Activities, but also are potential scientists of the future. Due to this fact, a summer school for students interested in the scientific approach of the TMFB will be organised in the context of Promotion of Young Researchers by the staff of the Supplementary Cluster Activities in 2010. With the intention to recruit more young students for a scientific career in natural sciences and engineering sciences, this measure seems to be very promising and fits well into the future concept of RWTH Aachen University, aimed at the integration of more undergraduate students into research projects. The measure also underlines the growing importance of interdisciplinary research of natural sciences and engineering [Aac08]. In addition, a workshop has been organised in the context of the sub-task Promotion of Equal Opportunities to discuss the needs and possibilities of female researchers working within the Cluster of Excellence. With the input of the discussions of this workshop, a set of tailor-made measures will be developed for the promotion of female researchers. Furthermore, networking activities for the German national-project tasteMINT (funded by the German Federal Ministry of Education and Research) have been intensified by the Supplementary Cluster Activities, to reach more female pupils and inspire them into a career in engineering sciences or natural sciences. As a result, assessors for the MINT-subjects (Mathematics, Information Technology, Natural-Sciences and Technology) are educated within the TMFB, to introduce pupils into cluster-specific topics. Moreover, the presence of the representatives of the Supplementary Cluster Activities at the fair fam2008 (14th to 16th November 2008) and at the conference Going Diverse (29th to 30th November 2009) in Aachen, Germany, contributed to the further recruitment efforts of the TMFB among female researchers. To receive information about gender and diversity activities from other Clusters of Excellence at RWTH Aachen University, the contact to the gender representative of UMIC (Ultra High-Speed Mobile Information and Communication) has been strengthened, to share the best practices of both clusters in the future. The third of the three main tasks of the Supplementary Cluster Activities deals with the topic Knowledge Management and Dissemination. In reference to Dissemination, the Supplementary Cluster Activities initiated the conception and the design of a cluster-specific brochure for the TMFB in 2008. On the one hand, the brochure contributed to a general overview of the scientific approach of the entire cluster for every member of the TMFB. On the other hand, the brochure also contributed to a broader public perception in regard to further industrial or scientific partners interested in cooperating with the Cluster of Excellence. Concerning the sub-task Knowledge Management, an important step to a closer interdisciplinary connection of all scientific actors within the TMFB was realised by the implementation of the technical tool Knowledge Map in June 2009. The semantic web of the Knowledge Map enhances networking activities and simplifies the cluster-internal flow of information, because it connects project-specific information with associated information (e.g., personal contacts or published literature
Network Management for Clusters of Excellence
199
of the TMFB members). Hence, by using the web-based tool via the cluster-specific intranet, every participant of the TMFB has the possibility to fill in the Knowledge Map with personal content and information [BSHH06, SBH+06] Thereby, competences and responsibilities, as well as the cluster-specific scientific knowledge, can be shared and disseminated. Thus, the Knowledge-Map constitutes a tool in which the development of a “common language” [SGF03] of the entire network can be strengthened and cluster-internal synergies can be enhanced continually [HIO03] . The process of finding a common language in networks like the TMFB is a great challenge, because different scientific disciplines use different scientific terms and definitions. All these described measures are applied in the context of the Supplementary Cluster Activities to enhance the overall performance of the TMFB during the cluster’s runtime of five years. Moreover, product and process innovations were aspired as the scientific results of the entire interdisciplinary network with its overall vision to develop sustainable tailor-made biofuels for clean combustion. Thus, the scientific process to reach this vision is supported by the Supplementary Cluster Activities. During this process, the implementation of an adequate controlling tool is crucial to measure the performance (progress as well as stagnation) of different subprojects, as well as the level of the Cluster of Excellence. Hence, the implementation of a cluster-specific Balanced-Scorecard-Approach, which can monitor cluster performance to derive adequate new measures or adjust former measures, shall subsequently be explained.
3 Cluster Internal Performance Measurement Since the Cluster of Excellence TMFB constitutes a heterogeneous scientific network, publicly funded from 2007 to 2012, an adequate performance measurement system is needed to control the scientific output, personal development and the entire management strategy of the cluster. Because of that, the introduced Balanced-Scorecard-Approach in the TMFB enables the measurement of the key performance indicators of the entire Cluster of Excellence annually. The Balanced-Scorecard (BSC), developed in 1992 by Kaplan and Norton, constitutes a “performance measurement system” [KN92]. It was primarily created as a communication-, information- and learning-system within enterprises. Hence, the Balanced-Scorecard-Approach, implemented in the TMFB and specified to the needs of the cluster, tries to transfer the theory from classical corporate controlling to the controlling of highly complex, knowledge intensive networks, like the Cluster of Excellence. Due to the fact that the network of the TMFB is characterised by a completely other form of organisation and management than an enterprise, several modifications of the former BSC are necessary. A special character of networks affects the network-specific organisation of work, because cooperation within the networks is based upon the voluntary decision of each actor to take part in the network
200
Florian Welter et al.
[AC04, Sau05]. Because of this, no strong hierarchical structures can be found within the networks in contrast to the stronger hierarchical structures in conventional enterprises. Therefore, choosing the right way of decision-making depicts an enormous challenge for cluster management, because decision-making cannot just be done in a simple top-down manner, but needs to be conducted in a step-by-step manner through discussions with all actors in the network. Concerning the TMFB, the willingness to lead a dialogue between the layer of the cluster management, leading professors and leading researchers on the one side and the layer of PhDs and research assistants on the other side, is crucial to build up the trust for a successful development of interdisciplinary cooperation. To initiate a dialogue between these two layers within the Cluster of Excellence, the Balanced-Scorecard-Approach is a promising tool, as it allows the definition of a common cluster vision, common targets and milestones [FS02]. In how far the vision and targets are fulfilled by all participants of the cluster, can be examined by the implementation of a Balanced-Scorecard-based survey among all members of the TMFB. The questions and corresponding figures of the survey are defined by the Supplementary Cluster Activities in cooperation with the executive board of the cluster during the initial phase of the TMFB and can be compared and re-designed, if necessary, for the purpose of control after each survey once a year. But how exactly is the Balanced-Scorecard-Approach implemented within the TMFB? First of all, it is necessary to define a common vision for the entire cluster, which can be described as the development of ‘tailor-made fuels from biomass for clean combustion’. This vision has to be communicated among all of the layers of the actors in the Cluster of Excellence, meaning that everybody has to adapt his or her scientific work to the defined common vision. In addition, the vision can be divided into several sub-targets and milestones, helping us to make the vision more measurable for all participants of the TMFB and allowing a cluster-specific control. In contrast to the classical controlling approach, which stresses the importance of hard (quantitative) financial figures, a BSC tries to include soft (qualitative) figures, like the cluster-internal learning atmosphere or the development of cooperation among the researchers [FS05]. This is an important reason for the implementation of a Balanced-Scorecard-Approach for the purpose of control within scientific networks, which cannot just be reduced to their financial outcome [Ahn03]. According to the primary BSC of Kaplan and Norton, the modified BalancedScorecard-Approach for the TMFB possesses four perspectives necessary to create sub-targets. Nevertheless, these four perspectives specify the needs of the TMFB, because they are not identical to the former perspectives named by Norton and Kaplan. The Financial Perspective remains an integral part of a modified BalancedScorecard-Approach, but the focus within this perspective is set on questions concerning the working time of the researchers within the TMFB. Another of the four perspectives is the Customer Perspective/Output. This perspective is important, because sub-targets are defined in it, including the needs of the customer. For instance the industry can be defined as a customer who is interested in innovative scientific results of the TMFB. But also the generation of positive effects by the existence of the cluster of excellence on the research location of Aachen, e.g., the image of RWTH Aachen University can be defined as a sub-target, and
Network Management for Clusters of Excellence
201
therefore, it can be a component of the Balanced-Scorecard-Approach. The third of the four perspectives of the BSC in the TMFB is constituted by the Internal Perspective/Research Cooperation. One of the aims of this perspective is the achievement of a strong and productive cooperation among the staff within the cluster. In addition, facilitating direct and indirect communication between all actors in the TMFB is a crucial sub-target of the Internal Perspective/Research Cooperation, because a lack of scientific exchange within the cluster limits the development of interdisciplinary scientific results. Finally, the promotion of all employees in the network of the TMFB – individuals, as well as groups of scientists - is a sub-target of the Learning and Development Perspective. Reaching a level of highly motivated scientists in the Cluster of Excellence constitutes a common goal which has to be controlled regularly, so that adequate adjustments can be introduced by cluster management, if they become necessary. The implementation of the Balanced-Scorecard-Approach in the TMFB can be visualised by an iteration loop of different steps run through once a year (cf. Fig. 2). With the intention to measure the performance of the network (step 1), the accomplishment of a web-based survey is a crucial part of the Balanced-Scorecard-based evaluation (step 2). Due to the fact that the web-based survey is executed among the layers of executive board, leading professors and leading researchers, as well as among the layer of PhDs and research assistants, comparable figures can be collected. Hence, an adequate statistical approach for the comparison of the figures depicts the calculation of the average values throughout the scaled answers. The range of the scaled answers varies from 1 = ‘positive answer’ up to 5 = ‘negative answer’. For the statistical analysis of the data (step 3) these two layers are important and necessary to compare the corresponding answers of each group. In addition
Fig. 2 Implementation of the Balanced-Scorecard-Approach in the TMFB (Center for Learning and Knowledge Management and Department of Information Management in Mechanical Engineering (ZLW/IMA) RWTH Aachen University)
202
Florian Welter et al.
to the average values, the standard deviation of the respective answers has to be calculated to analyse the differences in the range of the given answers. Furthermore, qualitative remarks can be given by the participants, so that not only figures are collected through the survey, but also qualitative answers or suggestions. By doing so, on the one hand, differences among the answers between the two layers becomes obvious and, on the other hand, similarities and progress becomes clear [Jan03]. It seems obvious that a question, evaluated positively by both layers, is a worthwhile target, but the different evaluation of answers displays a need for action for the cluster management. Because of this, the reflection of all answers (step 4) is a significant task conducted by the representatives of the Supplementary Cluster Activities, together with the executive board of the TMFB. By reflecting on the development of the entire network, a process is initiated described by Kaplan and Norton as “double-loop-learning” [KN01a]. This means that management reconsiders decisions made in the past in the context of actual circumstances and the needs of the Cluster of Excellence. Furthermore, the identification of suitable measures, e.g., concerning the promotion of cooperation within the entire network or single groups of the network, is a very important step which follows after the reflection (step 5). With the implementation of the defined measures, the iteration loop is completed and a new one can follow to re-measure the performance of the network. The accomplishment of the iteration loops within the TMFB is planned every twelve months, so four loops can be accomplished during the five years of the project (first evaluation in 2009, last evaluation in 2012). An advantage of these four iterations is the possibility to compare the development of the answers nearly over the entire project period to promote the development of an ongoing strong interdisciplinary cooperation.
4 Results of the First Balanced Scorecard Approach Through the analysis of the answers of the first web-based survey, which constitutes a crucial step to measure the performance of the Cluster of Excellence, important insights concerning the status-quo and the further strategic development of the TMFB became obvious. Although participation in the first web-based survey was voluntary for all members of the cluster, a relatively high overall response rate of 66% was achieved between 2nd July 2009 and 29th July 2009. This response rate was positive and illustrated the successful implementation of the Balanced-Scorecard-Approach. Thus, 23 of 28 people (82%) took part among the representatives of the cluster management, leading professors and leading researchers, in comparison with 43 of 70 participants (61%) among the PhDs and research assistants.
4.1 Results of the Internal Perspective/Research Cooperation The analysis of the questions belonging to the Internal Perspective/Research Cooperation elucidated the ability to present the content and scientific line of
Network Management for Clusters of Excellence
203
argumentation of the entire Cluster of Excellence, as well as of the different subprojects of the cluster, was very high throughout all layers of participants. This illustrates that the vision and the sub-targets of the TMFB were communicated very well among all actors in the initial phase of the scientific network, an important achievement for the heterogeneous consortium of the TMFB. The internalisation of the common vision constitutes a crucial element of a cluster-internal Balanced-Scorecard-Approach, because just those visions and targets can be reflected each year, which are comprehended and accepted by all actors of the network [KN01b]. In spite of the internalised common vision of the TMFB, other results of the Internal Perspective/Research Cooperation illustrated that the flow of information within the Cluster of Excellence still needs to be enhanced. On the one hand, the established BSCW-Server used as a cluster-internal platform for communication and data was evaluated as being too slow and too static. Hence, a complement and more extensive use of the Knowledge Map by all actors of the network promise to augment a more dynamic flow of information within the TMFB. Additionally, the introduction of a newsletter has been considered by cluster management as a consequence of the first evaluation to provide actual information on all sub-projects of the entire network for all cluster members. Moreover, the necessity to foster face-to-face communication in the TMFB became evident in the questions concerning the quantity of the cluster- and sub-project meetings. For instance, the majority of PhDs and assistants would like to conduct a colloquium for research assistants more often than just once a year.
4.2 Results of the Learning and Development Perspective After the interpretation of the questions concerning the Learning and Development Perspective, one can underline the general atmosphere in the TMFB as perceived very positively by all participants of the first Balanced-Scorecard-based survey. This can be described as a crucial achievement for the initial phase of the heterogeneous network. Furthermore, the yearly International Workshop and the Thematic Workgroups of the cluster received an excellent evaluation, which accentuates the importance of these interdisciplinary activities, concerning cluster-internal learning processes, to optimise the entire scientific output. In contrast, a prevailing acceptance of reporting activities (e.g., the quarterly status report or the yearly report) could not be stated after the interpretation of the answers. Thus, some additional qualitative remarks of the participants illustrated that more time for scientific work is needed, instead of time invested in administrative activities like reporting. Another important aspect in the Learning and Development Perspective dealt with the introduction of an advanced training programme for the entire cluster. The corresponding remarks of this topic emphasised that some of the actors of the network demanded further training in scientific writing or presentations and communication skills, so that the organisation and development of adequate cluster-specific seminars can be intensified by the representatives of the Supplementary Cluster Activities within the TMFB.
204
Florian Welter et al.
4.3 Results of the Customer Perspective/Output In regard to the Customer Perspective/Output, a couple of insights became clear. On the one hand, the majority of the respondents underlined that within the entire Cluster of Excellence, more research output is achieved than in 50 single research projects. This means that the basic intention of the scientific cluster concept is accepted. This can be evaluated as positive, because the researchers obviously profit from synergies based on close cooperation and regular interpersonal exchange. In addition, the foundation of the TMFB promoted the image of the entire research location of Aachen, so that a long term goal of the entire German excellence initiative, shaping RWTH Aachen University as a globally known excellent research location, became closer. Furthermore, the participants stated that other projects were initiated beyond the borders of the cluster. This can also be evaluated as a very positive output. On the other hand, the contacts to industrial partners should be improved after the interpretation of the Balanced-Scorecard results. Although a formal exchange already exists between the scientists of the TMFB and the Industrial Advisory Board, which constitutes an important part of the organisational structure of the Cluster of Excellence, an intensification of contacts to industrial partners is possible, e.g., in the form of special industrial working groups for single sub-projects.
4.4 Results of the Financial Perspective In reference to the division of working time within the TMFB, the results of the Financial Perspective elucidated that relatively high differences exist between the answers of the two layers of participants, although one has to stress that those differences all turned up on a positive level of answers. For instance, the layer of cluster management, leading professors and leading researchers, emphasised that the research assistants received enough temporal free space to fulfil their tasks within the Cluster of Excellence. In comparison, the evaluation among the PhDs and assistants was worse, but overall, it was still on a quite positive level. This fact can be described as a typical dilemma of manager-employee-relationships characterised by arguments of the respective group on each side of the dilemma.
5 Concluding Remarks The results of the first Balanced-Scorecard-based evaluation in the TMFB support the executive board of the Cluster of Excellence in its management decisions, and thus, the results are helpful for the strategic development of the entire network. On the one hand, the executive board considers adequate measures by interpreting the results of the Balanced-Scorecard-Approach together with professors, leading researchers and representatives of the Supplementary Cluster Activities. On the other hand, decisions on some measures are discussed together with all research assistants and PhDs of the Cluster of Excellence to avoid too many top-down decisions,
Network Management for Clusters of Excellence
205
which can hamper voluntary scientific cooperation based on trust. Due to the fact that evolving trust is crucial for successful cooperation within the cluster, the opportunity for bottom-up proposals and suggestions from research assistants should be considered as very important concerning decision-making [FE04]. In addition a bottom-up-approach positively influences the development of the entire Cluster of Excellence through highly motivated employees. In regard to the further scientific development of the TMFB, one can stress as a concluding result that the scientific cooperation among researchers is already on a good level, but it still has to be enhanced to become excellent , not only on the level of IRFs, but also on the level of the entire cluster. Concerning this, the creation of a common figure canvassing the holistic scientific processes in the TMFB has been initiated by cluster management and the representatives of the Supplementary Cluster Activities, so that each single research project can be aligned in this figure by its representatives. With the aim to generate a more detailed overview of clusterinternal scientific processes for all members of the cluster, the figure can be assessed as an integral part of visualising the entire scientific cooperation.
6 Summary The Supplementary Cluster Activities of the German Cluster of Excellence “TailorMade Fuels from Biomass” (TMFB) at RWTH Aachen University are responsible for different measures that support stronger cooperation between the researchers working within the cluster, as well as the strategic management of the entire network. The main tasks including the corresponding measures are titled with Benchmarking, Personal Development and Knowledge Management and Dissemination. This paper explained the range of activities conducted by the representatives of the Supplementary Cluster Activities. The focus lied on the implementation of a cluster-specific Balanced-Scorecard-Approach. To what extent modifications of the classical controlling tool are necessary to match the needs of the TMFB have been elucidated. Hence, the Balanced-Scorecard-Approach of the TMFB contains four perspectives, although the qualitative figures play a more decisive role for the controlling of the scientific network than in the primary controlling approach for enterprises. Furthermore, the implementation of the Balanced-Scorecard-Approach in the Cluster of Excellence can be visualised by an iteration loop with five steps, starting with a web-based survey among the staff of the cluster conducted once a year. With the collected answers from the survey, figures in the form of average values can be calculated, which constitutes comparable values to measure the performance of the entire cluster. An advantage of this approach is the fact that adequate measures can be defined after the analysis of these figures to foster the success and scientific output of the TMFB. The web-based survey tries to evaluate all implemented measures to determine problems and opportunities. Within the Cluster of Excellence, this web-based survey was a great success. Not only did a high percentage of research staff participate in the survey, but basic problems within the cluster and
206
Florian Welter et al.
improvement suggestions became clear by the reflection of the results. One of these improvement suggestions is the generation of a common figure for the TMFB, in which every single project can be aligned as part of the entire scientific process. Moreover, the outcome of the survey illustrates that the majority of respondents believe that within the entire Cluster of Excellence, more research output is achieved than in 50 single projects. This illustrates the synergy effects caused by the structure of the cluster which focuses on interdisciplinary cooperation. This work was performed as part of the Cluster of Excellence “Tailor-Made Fuels from Biomass”, which is funded by the Excellence Initiative by the German federal and state governments to promote science and research at German universities.
References [Aac08]
RWTH Aachen University Rheinisch-Westfälische Technische Hochschule Aachen. RWTH Aachen im Exzellenzwettbewerb. September 2008. [AC04] A. Amin and P. Cohendet. Architectures of knowledge. Firms, capabilities, and communities. New York, 2004. [ACM06] B. Asheim, P. Cooke, and R. Martin. The rise of the cluster concept in regional analysis and policy: a critical assessment. In B. Asheim, P. Cooke, and R. Martin, editors, Clusters and regional development: Critical reflections and explorations, pages 1–29. New York, 2006. [Ahn03] H. Ahn. Effektivitäts- und Effizienzsicherung. Controlling-Konzept und BalancedScorecard. Frankfurt am Main, 2003. [Ahr04] D. Ahrens. Netzwerke: Von der Konfrontation zur Kooperation. In R. Oertel and F. Hees, editors, Das Netzwerk-Kompendium – Theorie und Praxis des Netzwerkmanagements, 1–8. 2004. [BSHH06] W. Backhaus, S. Sattari, F. Hees, and K. Henning. A Web-based Knowledge Map for integrating expert knowledge into higher education. In Conference Proceedings: 7th International Conference on Information Technology Based Higher Education & Training, Australia, Sydney, July 2006. [FE04] M. Fromhold-Eisebith and G. Eisebith. How to institutionalize innovative clusters? Comparing explicit top-down and implicit bottom-up approaches. Research Policy, special issue: Regionalization of Innovation Policy, 34(8):1250–1268, 2004. [FS02] H. R. Friedag and W. Schmidt. Balanced Scorecard – Mehr als ein Kennzahlensystem. Freiburg im Breisgau, 2002. [FS05] H. R. Friedag and W. Schmidt. Balanced Scorecard: Der aktuelle Stand nach 15 Jahren. Der Controlling Berater – Informationen, Instrumente, Praxisberichte, (7):433–458, 2005. www.rechnungswesen-office.de, downloaded 2009-05-22. [HIO03] K. Henning, I. Isenhardt, and R. Oertel. Wissen – Innovation – Netzwerke. Wege zur Zukunftsfähigkeit. 2003. [Jan03] C. Jansen. Scorecard für die Wissensmanagement-Performance in heterogenen Unternehmensnetzwerken. Fortschritt-Bericht VDI Reihe: 8: Meß- Steuerungs- und Regelungstechnik, 1024, 2003. [KN92] R. S. Kaplan and D. P. Norton. The Balanced Scorecard Measures That Drive Performance. Harvard Business Review, page 71–79, February 1992. [KN01a] R. S. Kaplan and D. P. Norton. Balanced Scorecard – Strategien erfolgreich umsetzen. Stuttgart, 2001. [KN01b] R. S. Kaplan and D. P. Norton. Die strategiefokussierte Organisation – Führen mit der Balanced Scorecard. Stuttgart, 2001.
Network Management for Clusters of Excellence [Sau05]
207
J. Sauer. Förderung von Innovationen in heterogenen Forschungsnetzwerken und Evaluation am Beispiel des BMBF-Leitprojektes SENEKA, volume 55 of Aachener Reihe Mensch und Technik. Mainz Verlag, 2005. [SBH+06] S. Sattari, W. Backhaus, K. Henning, E. Sjoer, N. Gronau, P. Pawlowsky, and M. Weber. The Web-based Knowledge Map - A Knowledge Management Tool for Preservation of Vital Expert Knowledge into Higher Education. In Mit Wissensmanagement besser im Wettbewerb! Tagungsband zur KnowTech 2006, pages 229–238, München, 2006. [SGF03] M. Schmette, E. Geiger, and M. Franssen. Phasenmodell für Netzwerke. In K. Henning, I. Isenhardt, and R. Oertel, editors, Wissen – Innovation – Netzwerke. Wege zur Zukunftsfähigkeit, page 65–71. 2003. [Uni] RWTH Aachen University, editor. Cluster of Excellence “Tailor-Made Fuels from Biomass”. Sample Proposal for the Establishment of Clusters of Excellence.
Part II
Next-generation teaching and learning concepts for universities and the economy
Application of Remote Technology to Electrical Power System Laboratories Saleh Al-Jufout, Sabina Jeschke, Abdullah Y. Al-Zoubi, Jarir Nsour, Olivier Pfeiffer
Abstract A prototype of a remote laboratory to conduct electrical power experiments over the internet has been developed to allow students to access the setup and perform measurement and analysis of typical electrical power experiments. The main goal of the paper is to provide a mean of resource sharing of expensive power equipment to students from other universities in Jordan and beyond in addition to students at Tafila Technical University, where the traditional electrical power systems laboratory is located. The design approach is based on modifying the existing traditional systems to facilitate remote access via the web. One typical experiment addressing synchronization procedure of two generators has been performed with a power system simulator. The data acquisition system and remote access of the power laboratory were designed based on the Lab-View programming language. Initial on students’ evaluation of the proposed online laboratory indicates that encouraging results may be obtained with remote experimentation when improved pedagogical aspects are integrated properly in the measurement procedures. Keywords Remote Experiments · Power Engineering · Engineering Education
generating units, different types of transformers, busbars, overhead transmission lines, cables and different types of static and dynamic loads in addition to the protection systems of the individual components and the overall system. The diversity of this laboratory depends mainly on the allocated budget. Thus, modern comprehensive power system laboratory may only be found at very few universities throughout the Middle East. Modern technology enables the remote access to equipment via the internet ([SRAA05], [AAZB08]). This proves to be particularly valuable in the case of electrical power engineering education to enable students to access such expensive laboratory instruments and equipment and to conduct experiments remotely. A consortium of universities may be formed to share laboratory resources by distributing online experiments over a given semester amongst the respective modules thereby allowing greater numbers of students to perform experiments with minimum cost.
1.1 Remote Laboratories A common point of criticism of the academic education at universities is that it is too remote from reality: theoretical knowledge is taught by lecturers and memorized by students, although it is often not thoroughly understood and cannot be used in real life applications. Hence closing the gap between theoretical and practical knowledge and checking the outcome is a major goal of online examinations. On the one hand using remote experiments this theoretical knowledge can be advanced to reality, while on the other hand the connection between abstract concepts and concrete challenges can be quantified. One of the intellectual challenges is to understand the role of a physical theory, the role of a physical model and the role of an experiment. Often, these terms are intermixed, and the classical curriculum offering separates lectures for theoretical and purely experimental laboratory classes do not make it easier for students to really comprehend the relations. Experiments play a central role in natural and engineering sciences and Modern eLearning technology may bridge the gap between theory and practice. The integration of new media into teaching and research has led to two principle kinds: virtual laboratories and remote experiments. Remote laboratories particularly enhance access to experimental setups for all students independent of limitations in time, budget, or access to classical laboratories and — maybe even more important — make the measured data electronically available for further analysis, enabling the students to directly compare the prognoses from the model with the results of their own measurements. Thus, web-based education [GNR05] has furthered the field of traditional electrical engineering education ([GMP+ 03], [DMM06]). It has also changed the traditional teaching style in higher education. When digital computers as well as various software and hardware are used in laboratories and classrooms, they can provide much more effective and efficient ways in teaching and make many mathematics related engineering problems easier to understand ([TPA06], [MRA+ 05]). Remote experiments are real-world experiments, remotely controlled from anywhere outside the laboratory, at almost any given time. Remote experiments consist
Application of Remote Technology to Electrical Power System Laboratories
213
of two vital parts, namely the experiment itself, which is supposed to be conducted remotely, and the remote control mechanism. Remote experiments are capable of enhancing the access to “‘real”’ experimental techniques which are often extremely complex or cannot be transported and are therefore restricted to a rather small community of students and researchers [JPRS07]. Remote experiments and simulations are actively used in various experimental sciences; related training courses have also been explored in chemistry [MLPB04]. An example for electrical engineering is given in [WH05]. However, the relation between experiment and simulation is rarely stressed. The capability of remote access through the Internet allows the student a direct comparison of theory and model on one hand and experiment and physical reality on the other, without having to switch back and forth between library or Internet and the laboratory. An interesting and related setup is the remote experiment and virtual lab for wind tunnels developed by Esche et al. [JXG+ 06]. For a multitude of remote laboratories, National Instruments LabVIEW® is used to control the hardware and collect the experimental data. LabVIEW also possesses a convenient web-interface enabling the remote-experimenter to perform any necessary adjustments. In order to view and control the experiment, a freely available web browser plug-in has to be downloaded and installed. Due to the modular programming structure of LabVIEW®, remote experiments can easily be combined or extended [col].
2 Electrical Power Laboratory The electrical power system laboratory is fourth-year senior electrical engineering laboratory course at Tafila Technical University as well as other similar programs in a number of universities in Jordan. The laboratory draws on and correlates with the knowledge obtained by students from their sophomore electrical engineering courses such as electric circuits and power systems. Usually, all of the laboratory experiments in this course were based on traditional equipment, devices, methods and techniques for measurements, data recording and result analysis. This typically makes experiments time-consuming and inefficient, and therefore, greatly limits the effectiveness in the students’ understanding of fundamental concepts and theories from the hands-on experimentation. A pilot project is currently underway at the Department of Electrical Engineering, of Tafila Technical University, to implement remote experiments using its comprehensive TERCO® power system laboratory. This laboratory is composed of different interconnected cubicles such as the power plant module with high voltage busbars and outgoing lines, transmission lines and distribution module, receiving substation module with high voltage side and load module. A number of experiments, under normal as well as fault conditions, are usually conducted manually and may be modified to accommodate remote setups. These experiments include settings of field control parameters, turbine control rectifier
214
Saleh Al-Jufout et al.
parameters, start and stop ramps and checking AC and DC supplies, alarm indications, acknowledge and cancelling procedures, status indications of isolators and breakers. Differential relays can also be tested by resistive faults or trim faults caused inside the protective zones. In addition, load distribution can be varied using auxiliary transformers to keep the currents within certain limits and generator performance under steady state and dynamic conditions can be studied for different types of loads. Finally, the dynamic characteristics of a controller is examined and the characteristics of over-current and under-impedance starting elements can be obtained. One of the experiments that can be performed with TERCO® power system simulator is synchronizing two generators (c.f. Fig. 1a). Each generating unit is equipped with speed and voltage controllers to adjust the generator terminal voltage and frequency as shown in Fig. 1b. This experiment is usually performed by three students to achieve the conditions of synchronization (equality of the voltage magnitudes and frequencies, same phase sequence and angles): the first student adjusts the first generating unit controllers; the second adjusts the other generating unit controllers; while the third student monitors the measuring devices and selects the
(a) Generating Units Panels and Measuring Devices for Synchronizing Two Generating Units.
(b) Voltage and Frequency Controllers of a Generating Unit.
(c) Measuring Devices for Synchronizing Two Generating Units
(d) Measuring Devices for Synchronized Two Generating Units.
Fig. 1 Building a Project on LabVIEW PDA Module
Application of Remote Technology to Electrical Power System Laboratories
215
suitable moment for connecting the two generators to a common busbars. The power system simulator is provided with dual voltmeter and frequency-meter that measure and compare the voltages and frequencies at both sides of the circuit-breaker through which the connection will be performed. Rotating three-light-bulbs are provided to check the phase sequence. These bulbs are connected between the phases of the two generators. If the phase sequence is the same, the bulbs will bright and dark in a rotating manner, but if the phase sequence is wrong; they will get bright and dark together. A synchroscope is provided to the check the voltage phase difference as shown in Fig. 1c. The light in the synchroscope rotates clockwise or counter clockwise depending on which generator has a higher frequency. When the light reaches the top green area, this means that the two generators are in phase (cf. Fig. 1d).
3 Remote Power Laboratory Remote laboratories deal with performing real laboratory experiments remotely via the Internet to help develop and implement new technologies enabling students to access real laboratory setups situated either in a central location on campus or distributed over several remote areas. The actual hardware is usually composed of basic data acquisition and control board with a digital I/O port, an analogue input module, and an analogue output module. The digital I/O lines are used to control the experimental setup under test. A video camera can continuously display the experiment activities on the remote user’s screen if real-time viewing is necessary. Once students have been provided with their username and password, they can log on into the URL address of the experiment remotely via their web browser. If a student is not a registered, the system will ask her/him to register for the lab before accessing and performing the experiment and keep this login data for future use. A student’s unique key-login name is checked against a database of students’ list. After logging in to the laboratory server the student can choose an experiment from the list of experiments to perform. When starting the experiment the student remotely takes over control of the hardware in the laboratory and is able to manipulate input values. The hardware, then, acts upon the input parameter and generates the same results as if the student was actually sitting at the control panels. The results are finally collected by the local host computer and sent back to the student’s computer. The student can rerun and submit different values to the experiment as many times as desired. Once satisfied, results are submitted for grading [col]. Remote technology thus facilitates the theoretical-practical comprehension of the subject and students feel more motivated when controlling the experiments according to their own devises and initiatives, they select their own schedule and program of their practices. It also allows the lecturer to contact the students in an autonomous way and include the control of an experiment in a theoretical class. It reduces the number of groups in practical lessons and allows the lecturer to set up a system in remote control.
216
Saleh Al-Jufout et al.
Fig. 2 Remote Power Laboratory Architecture
The application of interest here is to connect two generators in parallel. The main part of the paper focuses on the understanding benefits, procedures, conditions of synchronization and power flow for proper generators connections. The solution implemented in this work follows the architecture shown in Fig. 2. Students can access the power laboratory, from the classroom via local area network (LAN) or from anywhere beyond campus limits via the internet. A dedicated server hosts the virtual instruments (VIs), programmed in LabVIEW programming language to remotely control the experiments. The procedure of the application is generating two sine waves by determining the speed of rotation of the prime mover (rpm) and voltage parameter values for the two generators. For each parameter there are two controllers: coarse and fine, to determine accurate values. The value of rpm parameter must be 1500 since the generator is of four poles and the frequency is 50 Hz, and it is divided by 30 to determine the frequency of the sine wave signal. On the other hand, the value of voltage must be around 230 V and it is multiplied by 1.414 to determine the amplitude of sine wave signal. The rpm and the voltage values of the first generator are compared with those of the second one, and if there are equal and in phase the light is on, and then the user can switch on the circuit as shown in Fig. 3. In the front panel, the student can set the value of rpm parameter of the first generator using the coarse and the fine knobs, and then can watch the value on the indicator beside the knobs. The student can then set the voltage using other coarse and fine knobs, and watch the value on the other indicator, and set these the values for the second generator unit. The student’s goal is to achieve the equality by controlling the previous knobs; at the same time he must watch the scope to find the point when the two sine waves are in phase. When the conditions of synchronization are met, the green LED is on. In the study case the conditions are met when both the speed of the prime mover and voltage values for the two generators are equal with values corresponding to
Application of Remote Technology to Electrical Power System Laboratories
217
Fig. 3 Block Diagram for the Remote Lab LabVIEW Program
Fig. 4 Controlling and Measuring Devices on LabVIEW
1500 rpm and 230 V respectively. The last condition for synchronizing that is the two phases for the signals are equal too, as shown in Fig. 4. In the figure the student can observe both Generating Unit 1 and 2 and adjust all knobs until all the matching conditions lead to a completely synchronized system. This way, we have a system that is being run in real-time with the possibility to access the equipment directly providing the students with an environment that mimics the actual one. In cases where real measurements are required rather than just controlling equipment, the measured data can be made available through the LabVIEW program e.g. in the format of an excel-sheet, for further analysis, enabling
218
Saleh Al-Jufout et al.
the students to directly compare the prognoses from the model with the results of their own measurements. In this context further experiments are being carried out to see how students perceive this remote lab technology, particularly with a laboratory such as the power lab, where sharing of high power, heavy and expensive equipment is of paramount importance. This remote laboratory is to be integrated in the final examination of the lecture giving an opportunity to verify whether the theoretical knowledge from the lecture has been transferred to practical knowledge that can be applied to a real world scenario by the students.
4 Conclusions A prototype electrical remote laboratory consisting of a number of experiments, connected to a dedicated server, was developed at Tafila Technical University to enable students to access experiments via the internet. The remote lab was designed using LabVIEW and published to a university web-site. Student clients were able to access the remote laboratory using the internet. Further experiments are being conducted to study the perception of students of this remote lab technology, as related to the integration of new curricula into power engineering programs. The project’s long term objective is to become a hub for other universities in Jordan and the region to utilize its facilities through conducting a multitude or remote lab experiments. The project should in the end give a glimpse onto the future advancement of educational power engineering systems worldwide as we are approaching a new revolutionary age of advanced technology.
References [AAZB08]
Michael Auer, A. Y Al-Zoubi, Danilo Garbi Zutin, and Hatem Bakhiet. Design of Application-Specific Integrated Circuits for Implementation in a Network of Remote Labs. In Proceedings of the 2008 ASEE Annual Conference and Exposition, 22-25 June 2008, Pittsburgh, Pennsylvania, USA, June 2008. [col] A collection of remote experiments from the Institute of Solid State Physics (the “Remote Farm”). http://remote.physik.tu-berlin.de/farm/index.php?id=14&L=1. [DMM06] Zoe Doulgeri, Tilemachos Matiakis, and Student Member. A web telerobotic system to teach industrial robot path planning and control. IEEE Transactions on Education, 49(2):263–270, 2006. [FKU+ 92] A. Fitzgerald, Charles Kingsley, Stephen Umans, A. E. Fitzgerald, Charles Kingsley Jr., and Stephen Umans. Electric Machinery. McGraw-Hill Science/Engineering/Math, 1992. [GMP+ 03] E. Guimaraes, A. Maffeis, J. Pereira, B. Russo, E. Cardozo, M. Bergerman, and M.F. Magalhaes. REAL: A virtual laboratory for mobile robot experiments. Education, IEEE Transactions on, 46(1):37–42, 2003. [GNR05] D. Gillet, A. V Nguyen Ngoc, and Y. Rekik. Collaborative Web-based Experimentation in Flexible Engineering Education. IEEE Transactions on Education, 48(4): 696–704, 2005. [JPRS07] Sabina Jeschke, Olivier Pfeiffer, Thomas Richter, and Harald Scheel. On Remote and Virtual Experiments in eLearning in Statistical Mechanics and Thermodynamics. In Proceedings of the 2007 ASEE Annual Conference, Honolulu, HI, 2007.
Application of Remote Technology to Electrical Power System Laboratories [JXG+ 06]
219
Ruiqing Jia, Shanjun Xu, Songyun Gao, EL-Sayed Aziz, and Sven Esche. A Virtual Laboratory on Fluid Mechanics. In Proceedings of the 2006 ASEE Annual Conference: Conference CD, ASEE, Chicago, 2006. [MLPB04] R. Moros, F. Luft, H. Papp, and W. Bailey. VIPRATECH - Das online verfügbare Praktikum Technische Chemie. In Jantke Fähnrich, editor, Von e-Learning bis e-Payment 2004, Tagungsband LIT 2004, pages 332–328. Akad. Verlagsgesellschaft Aka GmbH, Berlin, 2004. [MRA+ 05] F.J.F. Martin, J.C.C. Rodriguez, J.C.A. Anton, J.C.V. Perez, C.B. Viejo, and M.G. Vega. An Electronic Instrumentation Design Project for Computer Engineering Students. Education, IEEE Transactions on, 48(3):472–481, 2005. [SRAA05] S.C. Sivakumar, W. Robertson, M. Artimy, and N. Aslam. A web-based remote interactive laboratory for Internetworking education. Education, IEEE Transactions on, 48(4):586–598, 2005. [TPA06] C.S. Tzafestas, N. Palaiologou, and M. Alifragis. Virtual and remote robotic laboratory: comparative experimental evaluation. Education, IEEE Transactions on, 49(3):360–369, August 2006. [WC99] B. M. Weedy and B. J. Cory. Electric Power Systems, 4th Edition. John Wiley, 1999. [WH05] H.D. Wuttke and K. Henke. In Jantke Fähnrich, editor, Von e-Learning bis e-Payment 2004, Tagungsband LIT 2005, pages 481–490. Akad. Verlagsgesellschaft Aka GmbH, Berlin, 2005.
Environments for Work and Learning 2020 Steps to the sustained preservation of an economic location Stefan Brall, Ursula Bach, Frank Hees
Abstract This paper highlights the challenges environments have to face in the field of work and learning up to the year 2020. Therefore, the paper shows the results of the questionnaire “The Future of Work and Learning” from 2008. The results demonstrate that the work and learning environments are affected by a huge number of dilemmas which must be reduced on a continuous basis. This ongoing process requires that the individual and the organization have the permanent ability to change. To preserve Germany as a business location it is necessary to step into dialog with international partners and review changes and innovative solutions. This paper shows the German project “International Monitoring” as an example. The overall objective of the project is to establish a continuous International Monitoring. Through observation, networking and dialogue a national and international opinion leadership in the field of innovative ability is aimed to be achieved in order to keep Germany and Europe globally competitive in the long run. Keywords Working Environments · Learning Environments · Future Studies · international Monitoring
(IMO - www.internationalmonitoring.com), as a starting point for the investigation, twelve conflict areas were compiled on the basis of the national research program “Working, Learning, Developing Skills. Innovative Ability in a Modern Work Environment” (Fig. 1). The research program addresses these dilemmas as an aspect of future research on work and innovation. This is why one purpose of the “International Monitoring” project is to deal with structuring the current areas of conflict determined in the program; another purpose is to identify and assess them in a national and international context. To this end, dilemmas are systematically identified, analyzed and structured, while new trends are investigated and processed for the target audience in the context of the program. Consequently, the monitoring activities are divided into four components: 1. Exploration: The international explorative study starts off the monitoring activities. By using open questions, the study establishes topics, problem areas and pressing research issues in the field. The answers mirror the diversity of the program and show the challenge spectrum for the coming years. Besides the current research program, the results of the explorative study also determine the starting points for further international analyses. The topic clusters derived from the answers mirror the range of the field and clarify the diversity of today’s research work and that of the future.
Fig. 1 Areas of conflict denominated in the current research program (Dilemmas)
Environments for Work and Learning 2020
223
2. Observation: By means of monitoring and exploration, the topical field of working, learning, competence development and innovative ability is rendered accessible. The analysis of all relevant sources, whether documents from the worldwide web or printed publications, can give some insight into trends and problem areas. The contents can then be presorted regarding various criteria for further analysis. On the one hand, automated monitoring facilitates finding relevant texts, and on the other hand it permits a first evaluation of the scanned data inventory. 3. Analysis: The core of the monitoring analyses looks at central focal points. First, qualitative analyses in the form of interviews or structured group questionnaires are utilized, and second, qualitative statements are verified in a quantitative manner. The central focal points are derived from the explorative study at the beginning of the project or from the automated analyses carried out in the course of the project. The results of the analyses, again, serve as foundations for further focal points ([VV09], [Bry09], [PI09]) and monitoring. 4. Combination: The result of automated monitoring and observation of focal points is the derivation of recommendations for further action. These are meant to be foundations for political decisions concerning topics of the project and thus need a strict evaluation carried out by an interdisciplinary and international group of evaluators. To this end, the results, but also the recommendations determined through monitoring and other areas of work, are verified in focus groups (e.g. by Delphi questionnaires) and thus placed on a broad basis for decision making. In the project, application of traditional scientific methods of monitoring is complemented by a dialog component. On a national level, conferences with researchers and practitioners are organized to discuss trends, best practice, research gaps and recommendations for action. But nonetheless the external view of Germany is a particularly essential component in the dialog. The International Panel of the project, for instance, discusses and assesses German practice and results of meta-trend analyses on a joint interdisciplinary level. Here, strengths, weaknesses, chances and risks to Germany as an economic venue are identified from an international perspective for the entire duration of the project, and with respect to that, appropriate action recommendations are made for Germany.
2 Research Method The questionnaire was applied as an explorative study, starting off an entire series of investigations accompanying the research program. With the help of the answers provided by experts, the conflict areas described in the research program are to be assessed, their actual significance determined on the basis of today’s research activities and, through denomination of future research requirements, additional demands for action are indicated. For this reason, and in order to exploit the research field
224
Stefan Brall et al.
in width, the main focus was placed on answers obtained through open questions. Hence, the main questions were: - Which subjects in the area of “Working, Learning, Developing Skills. Innovative Ability in a Modern Work Environment” currently influence your work? Which problem fields/challenges in the area of “Working, Learning, Developing Skills, Innovative Ability in a Modern Work Environment” are especially pressing today? - Which research gaps in the area of “Working, Learning, Developing Skills. Innovative Ability in a Modern Work Environment” do you consider to be relevant in the coming years? - How do you assess the relevance of these conflict areas with respect to the creation of work and learning processes during the next ten years? Do you see additional areas of conflict which will become important in these ten years? The survey was carried out online to guarantee worldwide accessibility. An invitation to participate was sent via email to a total of 1100 experts worldwide. The questionnaire was accessible from January to May, 2008, and available both in German and English. From altogether 214 returns, 140 questionnaires were filled out completely. The majority of participants came from Europe, 54 of them from Germany. In addition to the questionnaires, 18 in-depth interviews were carried out with the assistance of national experts. The majority of participants came from the field of Science.
3 Findings and Discussion 3.1 Introduction The program “Working, Learning, Developing Skills. Innovative Ability in a Modern Work Environment” addresses the dilemmas described in the introduction (chapter 1) as subjects of future work and innovation research. The questionnaire asked participants to assess the significance of the conflict areas both in their own country and on a global scale (chapter 3.2) and to determine other conflict areas of importance (chapter 3.3).
3.2 Assessment of the areas of conflict The relevance of the described areas of conflict was confirmed by the interviewees. Dissimilarities in assessing their relevance are mostly resulting from different perceptions of their relevance for Germany or from different points of view concerning global developments. Looking at Germany, the closely related areas of conflict “long-term strategies short-term profit expectations” and “social stability - economic flexibility” have
Environments for Work and Learning 2020
225
Fig. 2 Evaluation of the areas of conflict [BHS09]
been judged as particularly relevant. The conflict areas “competition - cooperation” and “globalization - regionalization” were regarded as less relevant, especially compared to their assessment when viewed in a global context (Fig. 2).
3.3 Naming other Areas of Conflict In addition, the questionnaire includes other areas of conflict which are presently deemed relevant from the perspective of the interviewees (Fig. 3). The most frequent choices dealt with the gap between the rich and the poor and closely related aspects of achieving employability. Another conflict area concerns the balance between standardization on the one hand and creativity and flexibility on the other. Also, participants remarked on the reliability of political conditions under which entrepreneurial action is to take place. The demographic change in its different facets remains a persistent phenomenon. The question of the integration of migrants is introduced from two different perspectives: On the one hand from the vantage point of demographic change and demands for highly qualified staff, and on the other hand from the perspective of countries suffering from Brain drain. The questions of specialization versus the demand for a wide competence basis are regarded as central topics against the background of innovation development and knowledge transfer. Apart from the already mentioned areas of conflict, several others were brought up; these however were suggested by less than five people each: - Individual responsibility for learning - Social responsibility for learning, Social responsibility - Individualism,
226
Stefan Brall et al.
Fig. 3 Further areas of conflict from the survey [BHS09]
- Rising work density - Staff reduction, - Employees - Management, and - Demand for personnel - Unemployment/Retirement (Fig.4)
3.4 A New Dilemma In the following section, the seven most frequently indicated areas of conflict are characterized more specifically based on the answers given by the interviewees. Wealth - Poverty The survey participants continue to see wealth and poverty as a conflict area which will increase in importance in the next ten years. Especially the increasing social gap between rich and poor - and thus being privileged or unprivileged - is recognized as a conflict area. With the increase of educational costs, the access to education for poor people is limited even further. As a consequence, this will lead to the impoverishment of larger social groups and the increase of precarious work arrangements such as short-shifting and contingent work, coerced or feigned self-employment and so forth. This, too, is part of the Wealth - Poverty dilemma. Employability - Lack of Skills and Educational Training According to the interviewees, rigid education policies combined with educational selection stand against the demand for competent and specialized employees. In analogy to the social gap mentioned above, interviewees see society splitting into educated and uneducated milieus, this often being determined by the educational level of the parents: if parents have a high level of education this usually results in
Environments for Work and Learning 2020
227
Fig. 4 Areas of conflict established after the survey [BHS09]
good educational opportunities for their children. Contrary to this, if children come from less educated families their access to education is a lot more limited. Standardization - Flexibilization While employees expect the greatest possible flexibility and businesses become more adaptable when co-operating with one another the desire for security and efficiency by means of suitable standardization continues. This dilemma appears in both micro- and macroeconomic connections. The question here is how to mate the desired creativity to business schedules and procedures, and how to reconcile innovation-conducive freedom with increasing control. This dilemma brings up another question, namely that of stiff organizational structures in general versus open or dynamic streams of information. Political Stability- Political Flexibility The problem of political stability, in contrast to the flexibility issue, does not only affect economic areas. We also have to look at the general disenchantment with politics stemming from a deficiency in reliability of politics. Daily changing political maxims, especially in education policy, are no rarity. Rather than developing long-term future perspectives, politics rather tend to ballyhoo a new dog-and-pony show after each published comparative study. What we need instead is solutions
228
Stefan Brall et al.
in the educational area which allow for a lasting learning effect and consequently foster employability. Demographic change This conflict area cannot be characterized by a mere comparison of two terms like “old versus young”. It rather describes an area that was often investigated and examined, but where few tangible actions were taken. The survey participants identified the main areas of conflict in the field of demographic change: the protection of employability for older employees, and social security where the so-called “generation contract” faces hard challenges. Also, ageing employees require new, or rather, different work arrangements concerning Work-Life-Balance. Additionally, with regard to the ageing population and large parts of society becoming “overaged”, the question of the “international competitive advantage in knowledge matters” comes up. Integration - Migration The dilemma in the field of integration and migration remains virulent. The fact that Germany needs migrants meanwhile is a proven and at least academically accepted fact. But the debate on migration should not be based exclusively on “utility values” and the international competition for highly skilled personnel. Germany as a destination country for immigration must also provide integration opportunities for lower educated immigrants. Opening and expanding the existing education systems would be one way to solve the integration problem. Specialist - Generalist The Specialist - Generalist conflict area is characterized by the necessity for high specialization and the concurrent need for extensive knowledge. In addition, the ability to integrate different fields of knowledge will become ever more important. The so-called future technologies - Micro-, Nano- and Biotechnology, as well as medical technology - require both interdisciplinary collaboration competence and extensive specialization. The opposition of theoretician and practitioner, as well as that of specialist and generalist, is dissolved due to the need to control both sides of the matter.
4 Conclusion The dilemmas described in this paper show that various efforts must be undertaken in order to sustainably secure the long-term protection of Germany as a business venue. In this process, work and learning will have to change radically. The consuetudinary separation of both areas will disappear over time, leading to lifelong learning in the course of work. Without continuous knowledge actualization and advancement, which also encompasses “unlearning” certain aspects [Hen09], individual employability, while growing in importance due to the demographic changes in Europe, cannot be maintained. Having an adaptable approach when exposed to dynamism and complexity is the key to cope with today’s tumultuous environments ([HHH09],
Environments for Work and Learning 2020
229
Fig. 5 Reduction of Dilemmas
[HHBH09]). The challenges described by present dilemmas require not only permanent individual transitions, but also continuous transitions on organizational levels as new areas of conflict will certainly emerge or increase in importance. (Fig. 5). It is impossible to meet current and future challenges on one’s own; they can only be overcome in networks and in dialog with other cultures and other perceptions. This is why international dialog is an essential component of the “International Monitoring” project. Only by active integration of a critical external view can Germany adopt a leading role in the creation of innovative work and learning processes and actively transfer these experiences back into the network.
References [BHS09]
S. Brall, F. Hees, and S. Sparschuh. Future of Work and Learning. Explorative Study within the BMBF Project “International Monitoring”. RWTH Aachen University, Aachen (in print), 2009. [Bry09] J. Bryson. Hybrid manufacturing systems & hybrid products. Trend Study within the BMBF Project “International Monitoring”. RWTH Aachen University, Aachen (in print), 2009. [Hen09] K. Henning. Innovation Champions. In Methods and Tools of Industrial Engineering and Ergonomics. Heidelberg, 2009. Schlick, C. (Hrsg.). [HHBH09] K. Henning, F. Hees, U Bach, and A. Hansen. Yes, we can! Warum Deutschland den Kopf nicht in den Sand stecken sollte. In Festschrift Gerhard Ernst ”Arbeits- und Dienstleistungsforschung als Innovationstreiber”. Stuttgart, 2009. [HHH09] A. Hansen, F. Hees, and K. Henning. Surfing Dynaxity! Entwicklungen und Trends in einer glo-balisierten Arbeitswelt aus Sicht des Internationalen Monitoring. In BMBFTagungsband zum zweiten Zukunftsforum ”Arbeiten, Lernen, Kompetenzen entwickeln“, Berlin (in print), 2009. [PI09] F. Piller and C. Ihl. Open Innovation with Customers. Foundations, Competences and Interna-tional Trends. Trend Study within the BMBF Project “International Monitoring”. RWTH Aachen University, Aachen (in print), 2009. [VV09] W. Veen and B. Vrakking. Homo Zappiens and his consequences for learning, working and social life. Trend Study within the BMBF Project “International Monitoring”. RWTH Aachen University, Aachen (in print), 2009.
Developing a PBL-based Rescue Robotics Course Frank Hees, Sabina Jeschke, Nicole Natho, Olivier Pfeiffer
Abstract Problem-based learning (PBL) denotes self-determined learning and learning through discovery, activity-based education, interdisciplinary education, and self-assessment. The participants in problem based learning courses learn to analyze a subject or a problem with minimal guidance by their teacher or rather their facilitator of learning. Students find and use the suitable sources of information by themselves, and finally, compare, select and convert the results. The essential highlight of the PBL approach is the examination of authentic (real life) and complex subjects. The origin of the PBL lies in application-based technical engineering subjects and later in medical education. Robotics education is perfectly suited for the application of PBL-scenarios as robotics combines a multitude of technological disciplines (ranging from computer sciences, software engineering, artificial intelligence, electrical engineering up to technology design) and its ubiquitous popularity with a variety soft skills (team skills, complex problem-solving strategies, etc.), required in the development process. The popularity of robots can be easily deduced from the large number of robotic heroes in literature and movies. Thus, robotics is ideally suited as a model project-oriented course of combining communication skills, development of strategies to solve complex interdisciplinary challenges, and different concepts of softand hardware engineering. Among the wide range of robotics applications, one field of particular importance is the field of “Rescue Robots”. Here, robots are developed that operate in catastrophe-scenarios, e.g. earthquakes or fires. Based on the data obtained from their various sensors (video cameras, infrared sensors, laser scanner and gas sensors), these robots have to manage their tasks autonomously in catastrophe-based scenarios. This comprises detection, rescue, and aid for victims should the situation arise. In order to fulfill these complex tasks, development of basic skills such as exact movements on unstable bedrock, field mapping, positioning and communication in weakly structured environments is necessary. Besides the construction F. Hees (B) ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany e-mail: [email protected]
of preferably all-terrain and robust robots, the improvement of innovative analysis procedures for complex sensor data is another focus of development. In addition, conception and realization of novel man-machine-interfaces come to the fore in order to support the operators of robots with their exhausting control tasks. Integrated in the “RoboCup”, the “Rescue-Robot League” clarifies the intensified orientation of the “RoboCup initiative” on real life applications. Another hint that rescue robotics represents a ideal playground for PBL scenarios in academic education. Beyond that, robotics is increasing the number of female students in the natural sciences and engineering. It has the potential of attracting girls and young females at their respective levels education by illustrating their own potential in a playful experimental setting. Independent design and construction of robots demonstrates the importance of creativity and social relevance, giving young women more confidence in their technical and scientific skills, facts affecting young women’s choice of degree. Keywords PBL · Robotics · Academic Education
1 Introduction The deployment of robots in disaster has received increasing attention in robotics research since the first suggestions in the aftermath of the Hanji-Awaji earthquake in Kobe, 1995, and the bombing of the Murrah federal building in Oklahoma City in the same year [Dav02]. Since then, robots have actually participated in 7 disaster, starting with the attack on the World Trade Center, New York City, USA in 2001 [Mur04]. While the success of the robots varied from scenario to scenario, due mostly to unforeseen challenges in locomotion and control of the specific systems, they proved that robots should and will play a growing role in future search and rescue (SAR) operations. As there are strong indications of global warming and the subsequent climate change increasing the chances of natural disasters [vA06, oCC07], the scenario of disaster response and relief has a strong social impact and relevance. Public awareness of the problem is high, providing excellent motivation for using such scenarios as the basis of problems in a PBL course. At the same time, the requirements placed on robotic systems deployed in such scenarios are both unique and technologically demanding. As a result, SAR robots are ideally suited to demonstrating the complexity of robot design, construction and control in a PBL course. The simple problem of navigating a robot within a collapsed structure alone leads to topics ranging from robust locomotive systems that will not get damaged on the extremely rough and uneven surfaces, cybernetics for precise control of the movement under such conditions, control as wireless remote control by a human operator may be blocked through steel components within the structure itself while fiber-optics cables might snatch and tear as the system moves through tight confines. In addition to the wide range
Developing a PBL-based Rescue Robotics Course
233
of robotic-related topics that may be introduced in the course of these scenarios, the unpredictable and ever-changing situations faced in rescue operations are ideally suited to train problem solving and decision finding skills.
1.1 Tasks in Search and Rescue The complexity and cost of a more flexible general purpose design result in SAR robots usually being designed to fulfill one or several specific task [MTN+08]: • Reconnaissance and Mapping: Disasters, both man-made (industrial accidents, attacks) and natural can drastically alter the layout of even well known areas or change the existing infrastructure and access options. Reconnaissance and mapping is meant to provide general information about the situation and a general geographical and topological layout of the stricken area, usually sacrificing minute detail in favor of quick overall coverage. • Search: In difference to reconnaissance, search is meant to provide detailed information about less accessible and often far smaller areas (collapsed buildings, landslides, flooded structures) for specific purposes, such as localizing persons, objects or specific situations (such as hazards). While speed is less of an issue, thoroughness and a minimal foot print to prevent additional damage may be essential. . • Structural Inspection: Most rescue operations require direct human intervention at some point. To ensure their safety close up sensor-data of the structural integrity of a damaged or collapsed structure is required before further access by rescue teams, both robots and humans, is possible. This task places high demands on sensor capabilities as well as maneuverability within the tight and confined spaces that can be expected. • Rubble and obstacle removal: Access to buried persons or hazardous situations often requires the removal of obstacles or the stabilization of badly damaged areas that resulted from the disaster. The manipulation of these objects requires more raw power than finesse combined with smallest possible size and foot print. • In situ medical assessment and intervention: Experience from the Oklahoma City bombing has shown the need of providing quick medical access to persons trapped within collapsed structures [BDM95], even before human rescuers can safely reach them. Robots can be used to reach these persons and provide diagnostic sensor data, a direct, usually verbal, channel of communication between victim and medical personal and transportation capacities for medical supplies or even life support. • Medically sensitive extrication and evacuation of casualties: Certain hazardous situations, such as the hot spots of radiological, chemical or biological disasters require the evacuation of victims without the possibility of more direct human intervention or beyond the capacity of humans operating in bulky and heavy protective gear.
234
Frank Hees et al.
• Mobile beacon or repeater: Robots can be used to form ad-hoc sensor and communication networks to enable or extend the range of all data traffic between the inside and outside of a disaster area. • On-Site surrogate for off-site specialists: Robots can be used to provide remote sensing, communication and manipulative capabilities to off-site specialists in support of on-site teams. The robot “acts as the body” for a remote controller, enabling the quick contribution of additional specialists, even in several different locations at once. Each of these tasks can be the basis of another task in the list, leading to a natural succession of problems ideally suited to a PBL course with subsequent tasks often requiring refinement in the skills and knowledge necessary. Reconnaissance and mapping might lead to the discovery of buildings destroyed in a landslide, requiring the search for survivors within these buildings. In consequence, the problem of best suited motive systems progresses, as reconnaissance can be performed from the air or surface, with relatively low demands on control and robustness of the drive system, while search will require entering into confined, rugged and hard to predict terrain, posing more challenging requirements. Search might result in the demand for structural inspection as a victim might be trapped in an unsound area of a structure and so on.
2 Related Work Under Socrates it was familiar that a priori ignoramus gains access to gradual solutions of complex problems by starting from questions. In nowadays’ knowledge and information society, this skill is of essential interest. To educate and therefore to improve this skill, a new style of teaching was developed in the seventies by Howard Barrows et al [BT80] called “Problem-based learning (PBL)”. He designed a problem-oriented curriculum for medical students on the basis of the ideas of David Boud [BF98] and John Dewey [Dew16], and this idea is used in several technical disciplines of today. The objective of PBL is the improvement of skills of acting by using problems that were designed to be real-world questions for a motivating workflow, and to impart social, technical and methodical skills. A further important characteristic is the basic attitude towards learning: learner and educator are equivalent persons regarding technical knowledge and behavioral role. According to Barrows [Bar96] a PBL curriculum has the following characteristics: • • • • • •
student-centered education, teachers are instructors, small student groups, problems form the organizing focus and stimulus for learning, problems are a vehicle for the development of problem-solving skills. new information is acquired through self-directed learning.
Developing a PBL-based Rescue Robotics Course
235
A problem-based curriculum is designed to develop the ability of solving problems. Operationalization of learning targets takes place via key skills, and not as in traditional teaching methods in form of knowledge snippets. According to Weber [Web04], the following key skills are important in a PBL curriculum: • • • • •
professional qualifications (expertise), methods and/or media qualifications (methods for the search of approaches), social skills (social behavior in learning environments and teams), personal qualifications (development of the personality), problem solving qualifications and decision-making and responsibility.
In regard to Barrows [Bar00] the essential educational aims are designed according to the explicit competencies of the students: 1. 2. 3. 4.
Acquisition of well thought-out knowledge regarding the problems. Improvement of problem solving strategies (reasoning) for real world questions Improvement of skills of self-directed learning and team work. Enhancing students’ motivation.
Real-world problems are designed by progressive asking. In addition, students should be solving problems as self-governing as possible. The instructor is a work flow and not a knowledge transfer moderator. Such a procedure for multifaceted problems such is difficult to realize. Therefore Meril [Mer02, Mer07] suggests integrating the elements of “instructional design” (cf. figure 1). In this way, complex problems are reconstructed by sub problems that are subsequently moderated according to PBL. The IITS at the University of Stuttgart offers a robotics courses for students of all fields, not limited to engineering, called “Robinson Mixed” [JKN+09]. The course design emphasizes supervising and mentoring students from non-technological fields and early semesters. Project and team work are an integral part of the concept. Completing the course enables the students not only to design, construct and program autonomous robots, but also qualifies them to teach basic robotics to high school level students. The students have the option to participate in instructor courses licensed by the Fraunhofer Institute as part of the Roberta [fAiSAP06] program. Aim of the course is improving the inter-disciplinarity of students in
Fig. 1 Instructional Design, modified from [vMBH04]
236
Frank Hees et al.
non-technological, particularly their interest in the technological fields. The course is offered as a regular, elective module credited with ECTS points. It has been held for the first time in the spring of 2009 and has received a positive resonance from the participating students.
3 Outline of the Course The students will be given the description of a disaster scenario such as a landslide, an area hit by a hurricane or earthquake, an industrial accident or a destroyed building. They will be asked to devise a plan of actions to safely evacuate the survivors of the disaster. This will lead to the identifying the task of reconnaissance and mapping as the starting point of the operation. The students are supposed to design and, if possible build, a robot or a team of robots to fulfill that task, with the scenario providing a natural transition and motivation to the next task and problem. Restricting ourselves to robotics, each task will usually have to address, amongst others, the following topics, under the specific restrictions and requirements of the task at hand: • Locomotion: which motive system is best suited for the environment to be expected and the task to be fulfilled? The decision will be influenced by considerations ranging from speed and size of the final system to ruggedness and range. Each task will typically require a different compromise between these parameters. Reconnaissance capability might be best provided by aerial robots (UAV) or unmanned surface vehicles (USV, basically unmanned boats) as large areas have to be covered and the robot’s payload is nothing but a sensor suite, while search might require small serpentine platforms to navigate narrow spaces and rubble removal might be best served by a large, powerful bulldozer-like tracked unit. • Manipulation: Several tasks like rubble removal, evacuation of casualties, or insitu medical intervention might require manipulative capabilities of the robot to place sensors or move objects. The control of such a manipulator will lead to the problem of direct and inverse kinematics. Design decisions have to be based on the requirements, as delicate operations or brute force tasks require very different systems. • Sensing: SAR operations are likely to involve the full spectrum of sensors available in robotics. Visual systems such as high resolution cameras will provide a general overview of the area or visual data from the inside of a building. Carbondioxide sensors can help in locating trapped persons, while a GPS can be used to help localize the robot (outside of screening structures) for high resolution mapping, with acceleration sensors or gyroscopes providing inertial navigation capabilities inside structures. Acoustical and seismic sensors are able to detect shifts in the rubble caused by buried victims or unstable structures. • Single-Robot Control: Many existing SAR robots are remote controlled, either through wireless communication or a fiber-optic cable, doubling as a safety tether to retrieve the robot if it gets stuck. The human operator can provide far more
Developing a PBL-based Rescue Robotics Course
237
complex and advanced decisions and reactions than most autonomous systems possess. However, direct control is not always possible, particularly in badly damaged collapsed structures where steel static components disrupt wireless data exchange while rugged surface and tight confines would tear a cable. Even under direct control of a human, cybernetic control systems can be used to fulfill sub-tasks such as moving in a straight line over very uneven surfaces with strongly varying traction [dWSB96]. Very simple behavior-based autonomous robots can fulfill very complex tasks without outside supervision if operating as a swarm [KZ93], making full autonomy possible under certain circumstances. How autonomous can/should the robot be? • Robot teams and team control: Cooperating teams of robots can provide capabilities beyond those of the single members. A UAV-USV team was used to inspect the damage of seawalls and bridges at Marco Island, Florida following the hurricane Wilma [MSPG06], with the aerial vehicle providing global positioning to the surface vehicle inspecting underwater damage. Teams of robots, especially heterogeneous teams, require control and coordination. Depending on the task and the capabilities of the robots, centralized, hierarchical or decentralized control architectures are best suited. The tasks to be solved have to be allocated and might require communication between the robots. • Communication and data exchange: Tasks like search or medical assessment might require direct data transfer between robot and human operators, or between robots. The students will have to decide what data has to be exchanged, if realtime capabilities are necessary (direct control commands or remote access to the camera on the robot for remote navigation) or if asynchronous communication is sufficient (still images taken of a large area in reconnaissance). The course can be easily extended beyond the scope of robotics, adding problems from logistics and crisis management.
4 Topics to be Covered in the Project In addition to the more generic aspects of robot design, each of the tasks defined above addresses a number of relevant topics within the field of robotics. The tasks defined in section 1.1 have a certain hierarchical order, adding new sub-fields of robotics and expanding in complexity those topics already touched upon in the previous tasks (cf. figure 2). As such, their order supports the natural work flow of the PBL-course as a whole. Reconnaissance requires far less complex sensor suits than structural inspection or medical assessment but can already be used to introduce the more basic topic of sensing and estimation. The control architecture for searching is less complex than that for human surrogate and remote control. This progression of tasks and the refinement of knowledge and skills to fulfill them keeps the learners’ motivation and interest up, while providing comprehensive coverage of all sub fields where desired. The following will provide two selected examples from the list of tasks defined in Chapter 1.1.
238
Frank Hees et al.
Fig. 2 Topics and Tasks
4.1 Reconnaissance and Mapping Requires: • Motive systems: (relatively compared to search) fast, but less reliable motive systems to cover large areas. These would typically include aerial robots and unmanned surface vehicles (boats). • Sensors: Sensor suits for this task range from mapping radar to visual and IRsystems. Sensing and estimation (creating an internal representation of the surroundings based on the available imperfect and often noisy sensor data) play a major role in this. This will introduce a number of methods and theory of statistic data analysis (e.g. Kalman-Filter [Sim06]) • Additional theory: Reconnaissance and mapping is closely related to the task of “Simultaneous Localization and Mapping” (SLAM) [DB06], [BD06], the ability of a robot to build up an internal representation of the world, the map, and localizing itself within this world, while exploring it.
4.2 On-Site Surrogates for Off-site Specialists Requires: • Motive systems: Since this group of tasks will usually involve direct, close deployment within the actual disaster area, the motive systems will be very robust in design, typically tracked, legged or serpentine systems. Uneven and slippery surfaces and lack of external localization require precise motion control while complex pathways caused by blocked passages require motion planning [dWSB96] in support of the human operator.
Developing a PBL-based Rescue Robotics Course
239
• Manipulation: Many remote controlled tasks require some sort of manipulative capability of the robot, typically a multi-jointed arm. Control of this arm requires knowledge of motion control [dWSB96] and an understanding of kinematics [SS96]. • Remote Control: Humans often require some sort of feedback from the robot to operate the manipulators in particular effectively and efficiently [MS93]. At the same time, a balance has to be struck between autonomy of the robot to reduce the work load of the operator and ensure safe operation independent of the control input, and the direct control of the operator over the robot’s behavior, leading to the theory of robotic control architectures [HBDH94].
5 Summary and Conclusions Rescue scenarios provide a well-suited background for a PBL project in robotics. The obvious social relevance enhances the motivation of students, while the technical challenges serve as natural entry points into any sub-fields of robotics. At the same time, the single tasks associated with rescue scenarios can be sufficiently focused to prevent overloading the students, providing a high degree of flexibility. The course described fits well into the existing framework of interdisciplinary robotics courses already taught at the IITS at the University of Stuttgart, Germany, in the “Robinson” programme.
H.S. Barrows. Problem-based learning in medicine and beyond: a brief overview. In Luann Wilkerson and Wim H. Gijselaers, editors, Bringing Problem-Based Learning to Higher Education: Theory and Practice: New Directions for Teaching and Learning. Jossey-Bass, San Francisco, 1996. H.S. Barrows. Problem-based learning applied to medical education. University School of Medicine, Springfield, Southern Illinois, 2000. Rev. 1994 Ed. T Bailey and H Durrant-Whyte. Simultaneous localization and mapping (SLAM): part II. IEEE Robotics & Automation Magazine, 13(3):117, 108, 2006. J.A. Barbera, C. DeAtley, and A.G. Macintyre. Medical aspects of urban search and rescue. Fire Engineering, 148:88–92, November 1995. David Boud, Grahame Feletti, and Feletti Boud. The Challenge of Problem Based Learning. Routledge, 2nd edition, 1998. Howard S. Barrows and Robyn M. Tamblyn. Problem-Based Learning: An Approach to Medical Education. Springer Publishing Company, 1980. A. Davids. Urban search and rescue robots: from tragedy to technology. Intelligent Systems, IEEE, 17(2):81–83, 2002. H. Durrant-Whyte and T. Bailey. Simultaneous localization and mapping: part I. IEEE Robotics & Automation Magazine, 13(2):99–110, 2006. John Dewey. Democracy And Education. Free Press, Original from The Macmillan Company, 1916. Carlos Canudas de Wit, Bruno Siciliano, and Georges Bastin. Theory of Robot Control. Springer, London, 1996.
240
Frank Hees et al.
[fAiSAP06] St. Augustin Fraunhofer-Institut für Autonome intelligente Systeme AIS and Gabriele Theidig Josef Börding Ulrike Petersen. Roberta - Anleitung zur Schulung von Kursleiterinnen und Kursleitern, volume 5. IRB Verlag, 2006. [GHL06] H.H. Gonzalez-Banos, D. Hsu, and J.C. Latombe. Motion Planning: Recent Developments. In Shuzhi Sam Ge and F.L. Lewis, editors, Autonomous Mobile Robots: Sensing, Control, Decision Making and Applications. CRC Press, Bota Racon, 2006. [HBDH94] G. Hirzinger, B. Brunner, J. Dietrich, and J. Heindl. ROTEX-the first remotely controlled robot in space. In Robotics and Automation, 1994. Proceedings., 1994 IEEE International Conference on, pages 2604–2611, San Diego, 1994. 3. [JKN+09] Sabina Jeschke, Lars Knipping, Nicole Natho, Ursula Vollmer, and Marc Wilke. The “Robinson” Programme: Interdisciplinary Education based on Robotics Curricula. In Xiangyun Du, Erik de Graaff, and Anette Kolmos, editors, Research on PBL Practice in Engineering Education, pages 185–198. Sense Publishers, P.O. Box 21858, 3001 AW Rotterdam, The Netherlands, May 2009. ISBN 978-90-8790-930-7 (paperback), ISBN 978-90-8790-931-4 (hardback), ISBN 978-90-8790-932-1 (e-book). [KZ93] Ronald Kube and Hong Zhang. Collective Robotics: From Social Insects to Robots. Adaptive Behavior, 2(2):189–218, 1993. [Mer02] M. David Merrill. A pebble-in-the-pond model for instructional design. Performance Improvement, 41(7):41–46, 2002. [Mer07] M. David Merrill. A task centered instructional strategy. Journal of Research on Technology in Education, 40(1):33–50, 2007. [MS93] M.J. Massimo and T.B. Sheridan. Sensory substitution for force feedback in teleoperation. Presence: Teleoperators and Virtual Environments, 2(4):344–352, 1993. [MSPG06] R. Murphy, S. Stover, K. Pratt, and C. Griffin. Cooperative Damage Inspection with Unmanned Surface Vehicle and Micro Unmanned Aerial Vehicle at Hurricane Wilma. In Intelligent Robots and Systems, 2006 IEEE/RSJ International Conference on, page 9, Beijing, 2006. IEEE Press. [MTN+08] R.R. Murphy, S. Tadokoro, D. Nardi, A. Jacoff, P. Fiorini, H. Choset, and A.M. Erkmen. Search and rescue robotics. In Springer handbook of robotics. Springer Berlin / Heidelberg, 2008. [Mur00] R.R. Murphy. Marsupial robots for urban search and rescue. IEEE Intell. Systems, 15(2):14–19, 2000. [Mur04] R.R. Murphy. Trial by fire [rescue robots]. Robotics & Automation Magazine, IEEE, 11(3):50–61, 2004. [oCC07] Intergovernmental Panel on Climate Change. 4th Assessment Report. Technical report, 2007. [Sim06] Dan Simon. Optimal State Estimation: Kalman, H Infinity, and Nonlinear Approaches. Wiley-Interscience, New York, 2006. [SS96] Lorenzo Sciavicco and Bruno Siciliano. Modelling and Control of Robot Manipulators. McGraw-Hill, New York, 1996. [vA06] Maarten K. van Aalst. The impacts of climate change on the risk of natural disasters. Disasters, 30(1):5–18, 2006. [vMBH04] J.J.G. van Merriënboer, Th. Bastiaens, and B. Hoogveld. Instructional design for integrated e-learning. In Wim Jochems, Rob Koper, and Jeroen Van Merrienboer, editors, Integrated E-Learning: Implications for Pedagogy, Technology and Organization, page 15. Kogan Page, London, UK, 2004. [Web04] Agnes Weber. Problem-Based Learning: Ein Handbuch für die Ausbildung auf der Sekundarstufe II und auf der Tertiärstufe. hep verlag, 2004.
Networking Resources for Research and Scientific Education in BW-eLabs Sabina Jeschke, Eckart Hauck, Michael Krüger, Wolfgang Osten, Olivier Pfeiffer, Thomas Richter
Abstract The major aim of the BW-eLabs architecture (networked virtual and remote labs in Baden-Württemberg) is the expansion of access to heterogeneous experimental resources (remote as well as virtual or hybrid) for cooperative execution of experiments in natural sciences and engineering as well as the reuse of raw data and experiments for research and education purposes. Thus, three major tasks take center stage for the BW-eLabs portal: 1st , the creation of efficient possibilities of external access to local experimental surroundings, 2nd , the guarantee of transparency and reproducibility of experiments, and 3rd , the promotion of cooperation and collaboration within the scientific community in experiment-driven high-technology fields. Nanotechnology and robotics serve as demonstrator disciplines because especially in these cost intensive areas access to experimental equipment is an important prerequisite for ensuring access to professional tools for all scientific communities involved. Corresponding raw data and related documents are examined along their life cycle and embedded into the entire process chain of experimental environments through sustainable indexing and field specific ontologies, traceable and reusable by means of semantic search. Existing infrastructure, e.g. digital libraries, decentralized tools, and repositories, are embedded into the BW-eLabs. As a framework, the 3D platform Wonderland (Sun) comes into place, taking the complexity of professional experimental set-ups into account. The BW-eLabs portal, together with its partner projects LiLa and NetLabs, is designed as an open network for scientific data and experimental set-ups under OpenSource/Open Access/Open Content policy. Keywords eLearning · academic education · virtual laboratories · remote experiments
1 Introduction The experiment is the essential research methodology component in natural sciences and engineering. Unfortunately, a contemporary experiment sometimes is subject to several restrictions such as financial and spatial limitations, and missing support [CJL+ 09]. The new media offer two fundamental models to tackle these challenges: virtual laboratories [JNR07] and remote experiments [JRST07]. A virtual laboratory is an elaborated software application to simulate experiments on the computer. These multimedia-based and interactive experiments are realistic, virtual simulations of an experimental set-up. The benefits, for example, are that many researchers or students can simultaneously work on the same experimental set-up or conduct abstract non-realistic feasible experiments. A remote experiment is a real experiment remotely controlled from outside the actual laboratory. Such an experiment consists of two main components: the experimental set-up and the technique and software needed to facilitate the remote access. An advantage is the simplification of the access to experimental set-ups, independent from space, time, and security considerations.
1.1 Scientific Networking of Virtual Laboratories and Remote Experiments Virtual laboratories and remote experiments are already used in many scientific research groups around the world. In addition, many existing experiments separated in laboratories could be easily equipped with remote access techniques. Regrettably, central access points (portals, repositories, etc.) do not exist yet to locate such useful experimental resources or information. This is an unsatisfactory situation for everyone involved, and therefore, a lot of experimental set-ups remain hidden. Moreover, collaboration between scientific research groups has to be furthered. To this end, an efficient networking between these groups would overcome challenges, e.g. to review experimental set-ups or enhance set-ups for better accuracy of results. Besides, seminal scientific studies do not emerge from isolated laboratories as individual achievements. It is vitally important to collaborate in a different kind of relation without organizational limits and geographical distances. At the same time, library and scientific document management systems, e.g. “eSciDoc”, do not contain links and information of those experiments that are explained in publications. Repeatability and transparency of results can hardly be ascertained in this way. Within this context the realization of scientific networking is a key factor in today’s information society.
1.2 Introduction to BW-eLabs The applicability of virtual laboratories and remote experiments is not limited to single disciplines. As described above, contemporary experiments are subject to
Networking Resources for Research and Scientific Education in BW-eLabs
243
several restrictions. A lot of these experiments are expensive and/or consist of very complex set-ups. Some of these experiments are very susceptible or need special environmental conditions (dust-free, temperature, pressure, etc.). A prominent example is the field of nanotechnologies having a huge influence on the current technology development as a key technology of the 21st century. Miniaturization of the observed objects and effects and the requirements for the accuracy of the measurements themselves and the equipment are closely connected to tremendous financial challenges (e.g. for clean rooms, electron microscopes, susceptible spectrometers). Hence, because only a small group of well-subsidized institutions obtain access to this equipment we chose nanotechnologies as a prototype for the BWeLabs [JPT06], [JNPT08]. The BW-eLabs project provides a repository of the “second generation” to face the technical challenges of experimental natural sciences and engineering in the field of nanotechnology. We use the service-oriented eResearch environment “eSciDoc” developed by FIZ Karlsruhe that can include dynamic and changeable data recorded along the scientific workflow (e.g. informal documents, experimental results, documentation) and experiments itself. The infrastructure keeps experimental results and the required workflow sequences available for subsequent use such as research and education. The benefits are the better networking mechanisms of geographically distributed research groups to share results through the internet. The former safeguards the reproducibility of the results by employing a post-connected “electronic laboratory journal” and therefore avoids repeated measurements by consequently processing existing results, the latter allows for the smooth exchange of laboratory facilities over distances, maximizing the usefulness of the corresponding equipment. To protect the results of research groups, all the material (raw data and equipment) is shielded by appropriate security policies and authentication via Sibboleth. Publications are an important means of information exchange in the scientific society. This can be realized by establishing the corresponding policies. The main target audience of the BW-eLabs project are scientists. Another target group are graduated students of the related topics. Additionally, the infrastructure can be useful in academic education scenarios to conduct complex professional experiments within a lecture or seminar.
1.3 Infrastructure of the BW-eLabs The infrastructure of the BW-eLabs Project consists of the following different components: 1. System Structure: The eResearch environment “eSciDoc” [esc10] used in the BW-eLabs comprises a set of services and solutions that facilitates user-inquires for static data to the corresponding repositories. It supports the development of suitable metadata concepts to describe experiments and raw data of the technical disciplines for documentation and repeatability purposes. In addition, different views on experiments are taken into account (experimenter’s perspective,
244
2.
3.
4. 5. 6.
7.
Sabina Jeschke et al.
experiment perspective, room perspective), thus providing a semantic infrastructure for indexing special data (semantic search) [RSS+ 09]. Digital Library-Static Content Provider: Libraries of the participating universities store and manage static data (journals, proceedings, etc.) including metadata such as a content manager of a digital classical repository. This is the first step to integrate scientific documents into the virtual laboratory spaces [SGK+ 04]. Experimental Content Provider: The corresponding research groups and institutions act as experimental content managers providing virtual laboratories and remote experiments. The infrastructure will be tested in a pilot phase tailored to the requirements of the partners. Access-Rights-Management: The infrastructure offers an adequate access-rightsmanagement similar to open access and open content. Integration of Digital Holography: By integrating the concepts of digital holography [OF06] real objects are visualized three-dimensionally in virtual spaces. Interfaces: To facilitate inserting data and experiments and to keep the entrance barriers as low as possible, interfaces for software used in the laboratory routine are created in the project. These software applications are e.g. LabVIEW for remote- and process-control of laboratory equipment, computer-algebra-systems like Mathematica or numerical packages, e.g. Matlab. Suitable interfaces are created on the side of the content providers for integration in already existing search engines and library portals to make the obtained data accessible and archive them. Communication tool “Virtual World”: The “virtual world” is a simulated threedimensional world providing the integration of available digital libraries and experiments, and a communication environment to promote collaboration between research groups. As a result, geographical distances between researchers are bridged and measurement data and experimental processes are archived. Integration of familiar tools and established infrastructure as well as intuitive system access and individualization procedures help to raise the acceptance and the usability of the portal.
2 Unique Features and State of Technology 2.1 Current State of Affairs in Baden-Württemberg and Germany in General The design, development and implementation of virtual knowledge spaces is being pursued on the national level throughout Germany, in particular at the University of Paderborn (sTeam [EBHE05] [EHGE06]), as well as CURE [Haa05] [HSH+ 04] at the Distance University of Hagen (Fernuniversität Hagen). However, despite years of extensive research and many visionary ideas, the field has yet to to come to fruition. This challenge is in no small measure the result of short comings in the existing base technologies and infrastructure required for the envisioned scenarios.
Networking Resources for Research and Scientific Education in BW-eLabs
245
Until now, most development fell into one of two categories: either isolated computer-based laboratory environments, each applicable to a few specific experiments, or portal technologies for the integration of diverse remote experiments and virtual labs. However, since the actual labs were designed for quite specific purposes, the lack of a coherent interface architecture prevented their content-oriented combination into more complex research scenarios.
2.2 Deployment of a 3D-Engine for Improved Usability The complexity of such extended or combined scenarios requires new concepts, posing further challenges not only for the IT-technologies involved but also in the field of usability. Intuitive access and operation of the system are vital to increase user-acceptance. To achieve this goal it is necessary to transfer real-world utilization concepts into the virtual world of the lab. In addition, this will reinforce the concept of the virtual lab or remote experiment as it trains the operation of real-life equipment. Vital input has come from research in the areas of virtual environments and gaming engines. Combined with developments in web service technologies, serviceoriented architectures, open 3D-Engines, social web and the digital extensions of libraries, this should provide the technological base for virtual knowledge spaces [SS06]. A suitable infrastructure is the open source three-dimensional platform “Wonderland” [won], developed by Sun, to be completed by suitable interfaces.
2.3 Document-Management Systems for Scientific Content Extending the eSciDoc infrastructure by incorporating these technological advances supports the publication, visualization and management of the finished objects as well as their continuous editing and use. The system is capable of handling textbased document at various stages of completion as well as primary (raw) research data from a wide scope of scientific disciplines. As a result, eSciDoc is well-suited to fulfill the knowledge and information management requirements of universities or other research or R&D related institutions, both public and private. The eSciDoc system goes beyond the capabilities of the widely deployed Dspace from MIT (Cambridge, USA, www.dspace.org), Eprints1 of the University of Southampton or the repository system OPUS2 , which is widely used in Germany and was developed at the University of Stuttgart (cf. A.1). All of these systems were custom-tailored to the requirements of an institutional publication repository, a limitation reflected in their user interfaces. 1 2
The eSciDoc infrastructure represents a generic infrastructure incorporating aspects of data quality, data management and long term availability. eSciDoc is based on the repository-software Fedora3 , and enriched by a series of newly developed services. The system provides features such as persistent referencing by assignment of appropriate identifiers, automatic extraction of technical metadata, and the administration of objects with heterogeneous metadata models. The system is designed to track and map the complete life cycle of a scientific document/data object, from its first inception to final publication and archiving. This is achieved by combining the concept of revision control with the mapping of formal and semantic relations between objects. By connecting data and documents/publications and by making them available to users, eSciDoc acts as an “enabling technology” for the comprehensibleness and the possibility to re-use research results required in scientific collaborations. Current solutions addressing these challenges only exist isolated. One example of an existing solution is the “Archer”4 system, promoted by the Australian Ministry of Innovation, Industry, Science and Research. It provides a software platform for managing and documenting the research data of Australian universities. Similarly to an existing set of applications developed for the humanities at the Max Planck Society, the eSciDoc infrastructure is suitable to be extended to a “scientific workbench”, supporting cooperation in virtual, distributed workgroups. The BW-eLabs will incorporate methods of data management in the natural sciences [SS05] [SBB+ 06] in co-operation with the scientific data processing servicegroup (Servicegruppe Wissenschaftliche Informationsverarbeitung) of the Freiburg Materials Research Center (FMF) at the University of Freiburg. Early implementations will address scenarios posed by the FMF, ranging from the characterization of the chemical analysis with high throughput technologies and the development of electronic laboratory journals to scientific information repositories for investigating individual technical questions. The FMF has developed methods for systematically managing analogue and electronic scientific information in the form of laboratory journal entries and primary data files in preliminary work, covering some basic aspects of scientific data managements [Sch07]. Based on the FMF-developed text-based full metadata format, these concepts permit the direct addition of experimental parameters and measured variables to the actual data, exactly as it is common in everyday work (e.g. in laboratory journals or measurement records), mapping the standard work-processes of the prevalent mode of operation. The BW-eLabs also address the complex problem of retrieving primary data: in difference to standard bibliographic indices, primary data catalogues have to support indexing by relevant experimental parameters. The BW-eLabs introduce the concept of using feature vectors consisting of experimental variables as a new classifier. The concept of feature vectors was first explored in the fields of pattern recognition
3 4
http://www.fedora-commons.org/ Australian ResearCH Enabling Environment, http://archer.edu.au/
Networking Resources for Research and Scientific Education in BW-eLabs
247
and machine learning. Complex objects are described as n-tupels of real numbers, facilitating semantic search algorithms based on custom-tailored distance measures within this n-dimensional vector space of primary data sets.
2.4 Remote Experiments with Real Objects Using Digital Holography Holograms possess a unique property: they contain the complete optical information (at the wavelength of the illuminating laser) of the recorded object. As a result, holograms can replace “real” objects in a number of experiments where purely optical properties about these objects are sufficient. In digital holography the wavefront resulting from the interference of light diffracted by the object and an undisturbed reference beam is recorded digitally, either directly from a CCD sensor or by digitizing an existing analogue hologram from film. This digital hologram can be reconstructed in one of two possible ways: by illuminating a suitable light modulator displaying the recorded hologram (for example the LCD-panel of a digital projector), or numerically in the computer. The former is basically identical to the classic optical reconstruction of an analogue hologram. In contrast, the latter method provides new possibilities not available in analogue holography. Numerical reconstruction does not require the exact replication of the original set-up to reconstruct an object, a severe limitation in sharing analogue holograms. A digital hologram can be stored on a computer or distributed over the internet, providing researchers in distributed collaboration access to the “real” object (or, at least, its optical properties). Current challenges in digital holography are the miniaturization of the recording devices, the robustness of the method and a simultaneous increase in measuring accuracy. Short coherence digital holography allows for layered scans of a threedimensional object (i.e. in depth, not just the surface of the object), enabling the reconstruction of layers of up to 20 micrometer thickness. This technique is used in microscopy and endoscopic medical procedures. It is capable of deducing the overall composition of an object by detecting areas of different elasticity using digital holography, in effect providing a surgeon with a sense of touch in endoscopic procedures. Comparative holography is a technique that compares nominally identical but physically different objects (sample-probe-comparison). By using digital holograms this method can be extended beyond the classic scope requiring an exact replication of the original set-up. As a result, it provides a very precise (fractions of the wavelength of the used laser), flexible testing method not requiring the simultaneous physical presence of the sample and the probe for comparison. The recorded hologram of the sample is reconstructed optically by illuminating a suitable light modulator. The reconstructed wavefront is used to illuminate the probe together with a reference wave. The resulting interference wavefront is recorded as a second digital hologram or simply viewed directly. Numerical reconstruction (or visual inspection of the probe) will show the interference phase between the wavefronts, the difference in the surface of the probe and the sample. This allows for real time
248
Sabina Jeschke et al.
Fig. 1 Schematic Set-up for Recording a Digital Hologram onto CCD.
inspection of objects revealing deformations of the surface on the scale of the laser wavelength, well suited for quality assurance.
3 Demonstrator Scenario Nanotechnology Nanotechnology deals with the synthesis, the properties, the characterization and the use of materials, on a scale between 1-100 nm. Moreover, it is considered to be the key technology of the 21st century. During the last 15 years, the nanotechnology developed more and more into a highly multi-disciplinary science. Its own Bachelor and master courses of studies have been established at universities and technical colleges. The high-grade multi-disciplinarity presupposes an active exchange and overlap between chemistry, physics, material sciences and even biology, if essential. The multiplicity of different scientific approaches and methods stimulates and requires collective experimenting, evaluating and publishing of completely diverse working groups necessary for the accomplishment of the research. This led to the installation and establishment of larger scientific centers and networks focused on nanotechnology in universities worldwide. Some examples are the German Center for Functional Nanostructures (CFN) at the University of Karlsruhe, together with Forschungszentrum Karlsruhe, the planned German Center for Nanotechnology at the University of Würzburg, the Swiss Nanoscience Institute (SNI), established at the University of Basel as a Swiss competence centre in the field of nanotechnology. The acquisition and shared use of expensive giant equipment in particular could be realized by the joint efforts of some of these centers. Due to the multi-disciplinarity and the necessity of networking, nanotechnological subjects are ideally suited for the application of virtual laboratories and remote experiments. In the context of BWeLabs, these experiments can be cross-linked in order to establish new standards in the field of knowledge management.
Networking Resources for Research and Scientific Education in BW-eLabs
249
The FMF at the University of Freiburg, Germany, analyzes synthesis, characterization and application of semiconducting and metallic nano-particles. Their insights are made accessible online/remotely as part of the BW-eLabs project made available for further transfer and application scenarios: The physical and chemical parameters relevant for the synthesis of nano-particles were evaluated exemplarily on the basis of the synthesis of high-grade luminescent CdSe nano-crystals [RBT+ 09] [WTWL07]. This synthesis is now being established in the context of the BW-eLabs project as a remote-enabled model synthesis. The parameters determined in the conventional reaction piston are transferred to microwave synthesis equipment, in which all decisive synthesis-parameters, e.g. temperature (profile) and concentration of reactants, can be standardized and are controlled and monitored remotely. The reaction process is amended by appropriate on-line analytics. By means of absorption- and emission-spectroscopy the synthesis process is monitored and controlled and can be compared to previously accomplished syntheses. Corresponding spectra of this on-line analytics are recorded and logged automatically. Alongside with this on-line analysis (small analysis) more complex analysis methods are connected in a later stage, e.g. connecting the transmission electron microscopy (large analysis) from another remote laboratory with the goal of a common structured datamanagement comprising a structured and clear representation of the results in form of an electronic laboratory journal. After the synthesis setup is concluded and the remote-ability of the synthesis process as well as appertaining analytics have been tested, further material syntheses are standardized and transferred to microwave basis. Gold nanoparticles are sought out to be the following model system and contacts to the department of inorganic chemistry of the University of Freiburg already exist on the basis of a common project for the synthesis of metalliferous nanoparticles. In addition, external partners can enter on this level and standardize their systems, accordingly. Thus, we develop a material synthesis library which is made accessible to other groups and contributes to the objectives of the BW-eLabs by securing the compiled research results of a working group beyond the period of the presence of a specific researcher. Moreover, this library serves as a point of contact for advanced research on the corresponding materials. Furthermore, virtual laboratories can be connected at this time, using the generated data for semi-empiric modeling [YRG+ 09] (structural characteristic statements of materials) and also provide input for the synthesis, structure and characteristics of new materials by using ab initio methods [dPTC08]. This forwards and accelerates tailored material development and permits the integration of nanomaterials with special characteristics into new applications. One example for the mentioned model systems CdSe nanocrystals and Au nanoparticles is the systematic introduction of impurity atoms (doping) of nanocrystals [NEE08] which opens new perspectives for the development of IR lasers or IR to emitting pigments to be used for bio marking. Then, nanohybrid materials, e.g. nanoscopic intermetallic phases as Au@Pt [YCZ+ 07] or Pt/Bi [DCL08] compounds can be developed as catalysts for reactions with a high industrial application potential.
250
Sabina Jeschke et al.
The responsible group at the FMF is working within the area of theoretical modeling of nanoclusters and nanocatalysts. Their results will be integrated at an appropriate time as virtual laboratory. Another applied component is the integration of nanomaterials in well-defined superimposed macro-structures. This is another area to integrate further remote experiments and virtual laboratories, especially in the field of nanostructuring of materials (remote experiment) and theoretical physics of complex systems (virtual laboratory). Synthesis as well as analysis and application of nanomaterials interact with the virtual laboratories of theoretical research groups. Figure 5 is outlining the linkage of a remote synthesis experiment with other remote laboratories, potential applications and virtual laboratories. An interactive platform linking diverse remote and virtual laboratories features sustainable long-term scientific advantages: • A more efficient and systematic development of new materials is promoted. • Compiled knowledge is indexed and secured on a long-term basis. This is of particular importance as once developed syntheses can by a successful standardization be made accessible to researches whose primary interests are, for example, physical measurements of that special material rather than its synthesis. • Constantly obtainable material quality that would be independent of the synthesizing chemist permits a continuously reliable research on this material. This item is of particular importance since syntheses are usually developed by graduate students and/or Postdocs who only stay at their respective research institute for a limited period and knowledge transfer does not always take the prominent role that it should.
Fig. 2 BW-eLabs Sample Architecture
Networking Resources for Research and Scientific Education in BW-eLabs
251
• Since only well-functioning syntheses are standardizable, some kind of quality control for the synthesis directives is automatically integrated. Thus, standardization of the synthesis will also be an important section of challenge within a synthetic PhD or master’s thesis in the future. • An interactive platform also facilitates the access to research methods that are not available locally, because of missing giant equipment, special devices, or particularly developed research equipment. Similarly as in internet chat rooms, extended ways of contact are possible on the scientific level which would not have come about in any other way (promotion of the development of national and international contacts). • An independently generated electronic laboratory journal has the advantage that all results together with the necessary parameters and all steps (synthesis ⇒ characterization ⇒ application) are retained uniformly and clearly arranged. Integratability into appropriate data bases prevents unnecessary parallel research and the amount of work for data evaluation is reduced for each researcher. In addition, research projects that have been classified as not successful by rule of a student’s thumb and therefore remained unevaluated are archived. At second glance, an expert might possibly come to a surprising discovery (increase the ratio of innovative random discoveries). • Established synthesis approaches are used also as exemplary sample experiments in teachings.
4 Demonstrator Scenario Robotics Robotics is one of the key technologies present in the 21st century. The essential research areas can be classified in: perception, actuator engineering, apperception and decision making. In terms of perception, the question of sensor technology and sensor combination arises. In the area of actuator engineering, manipulators and movement systems of different areas of application are decisive. With respect to apperception and decision making, the central points are data processing and filtering of knowledge representations, along with path finding and decisive derivation. Due to the accessibility of the different robotics applications and by means of evaluating any attempts already made in the decision making phase on an identical platform, new trials are tested faster and at a reasonable price. The environments where these systems can be examined in both a simulative as well as a realistic manner are manifold – two areas of application protrude in the context of the BWeLabs/NETLABS project:
Cooperative Scenarios: The availability for use of robotics systems in production environment continues to rise steadily. The necessary cooperation scenarios (between robots or between human and machine) lead to novel research questions that have yet to be answered.
252
Sabina Jeschke et al.
Fig. 3 Robocup-Team 2009, TU Dortmund
Further investigations however – in the case of realistic scenarios, e.g. with a collaborating installation – remain financeable only for a few research teams. Within the investigation of cooperative scenarios in robotics, the problem of costintenseness becomes obviously even more important due to the necessarily higher number of robots. Additionally, investigation for “cooperative behaviour” is only to some extend depending on the hardware – rather, concepts of awareness, decision making processes and task sharing play a central role. To investigate these research goals in a general and transferable manner, it is important to dispose of a large number of different robots in different application scenarios.
Fig. 4 Team of cleaning machines (SINAS), below: map of task distribution
Networking Resources for Research and Scientific Education in BW-eLabs
253
By means of suitable remote laboratories offered in BW-eLabs/NETLABS, production based research – also interdisciplinary – is carried out in multi-robotapplications and in human-machine cooperation scenarios. Any gained primary data is made available to other institutes. Autonomous vehicles: Since 1977, autonomous vehicles (a.k.a. driverless cars) have played a more and more important role in traffic technology. One opportunity to manage the increase of freight transportation and to optimize utilization of motorway capacities is the concept of truck platoons. With the aid of Advanced Driver Assistance Systems, trucks are electronically coupled keeping very short gaps (approx. 10 meters) to form truck platoons on motorways [FMH08]. This contributes to the optimization of traffic flow and reduction of fuel consumption advantaged by slipstream driving [MHS08]. Other concept and application scenarios of partly autonomous and fully autonomous cars are known. Within the DARPA Grand Challenge Cup, for example, different concepts of fully autonomous vehicles in substantial off-road courses are tested. BW-eLabs/NETLABS integrates movement-supported simulators in the form of virtual laboratories that allow for testing of passenger reaction, recognition of hazards, and even help to simulate path finding algorithms. Expensive, cost-intensive
Fig. 5 “Driver Organized Truck Platoons” [HP03]
254
Sabina Jeschke et al.
Fig. 6 “Spirit of Berlin”, FU Berlin, Team Rojas [Roj]
environments are linked and made accessible to other research teams. The simulators available within an institute are extended by means of functionalities, and thus allow for external control. In these areas of application, security plays a decisive role. This aspect must be taken into consideration when extending remote laboratories during the course of the BW-eLabs/NETLABS project. In addition, measures must be taken to examine how hazards for both humans and material can be prevented in remote-controlled laboratories. Aspects such as working and IT security must be taken into consideration. This is especially significant in human- machine-cooperation scenarios. For this case concepts are developed to guarantee the safety of the person interacting with a remote-controlled machine. In addition, robotics acts as a “connection discipline” within this application: actuators that control set-ups/experiments are necessary for remote-controlled laboratories. The aim to be achieved in the long run is a reconfiguration of laboratory environments. This will lead to a higher flexibility. Moreover, an infrastructure of intelligent “robots”, or better said “robot arms” has to be created to control experiments and laboratories. Additional concepts are developed, prototypically implemented and fitted accordingly to BW-eLabs/NETLABS.
5 Acceptance and Usability Usability and serviceability are decisive for the acceptance of the BW-eLabs. The usefulness has to prove itself exclusively by the increase in value it presents to researchers. The design of the user interface and interaction concepts are crucial for its operability. We will now address the latter in detail.
Networking Resources for Research and Scientific Education in BW-eLabs
255
Both the complexity of the planned research scenarios and the complexity of the planned overall system, exceeding the possibilities of classical repositories, which realize isolated access to static data or to special isolated laboratory environments, pose unique demands towards the interaction concepts and techniques to be used. Scientists’ creativity is promoted by a clear visualization of the data relevant for the experiments and is not restrained by difficulties with the interaction. Therefore complexity reduction is necessary in many application scenarios that, however, must not lead to a restriction of possibilities of use. Studies in psychology of perception of the last years showed that the use of closeto-reality metaphors facilitates a particularly intuitive use of software applications. A particularly complex environment as envisioned for the BW-eLabs must therefore most favorably be realized in three-dimensional representation. Such a communication environment in form of a virtual three-dimensional world on the one hand facilitates the conversion of interaction concepts that are already well-known from the real world and the integration and development of completely new approaches. On the other hand, this virtual world serves as a platform for the development and testing of new techniques of visualization of information. At the same time, mechanisms of information-based data analysis, i.e. processing of experimental data with automatic consideration of all meta-information like physical units, metric, etc. are employed. The preparatory work of the FMF in this area already shows that this approach leads to an effective complexity reduction, so that experimentally working scientists can test-drive, develop and implement complex data analysis algorithms. Effectiveness and efficiency of different interaction concepts and their acceptance by content providers and by the users is to be evaluated in usability tests with test subjects. With the help of the expertise won thereby, interaction concepts that encounter the broadest acceptance rate in the addressed target group are to be developed and implemented. It should in particular be examined which forms of social presence in the virtual world are essential for collaborative scenarios to be accepted by male and female scientists. These tests will be conducted in the usability laboratory at the Stuttgart Media University that is equipped with the appropriate devices for data acquisition and analysis.
6 Roadmap Right from the beginning, mechanisms for distributed authentication (shibboleth, and single sign on) are incorporated and usability studies for the 3D-interface are arranged. Meta data profiles and concepts for integrating digital content into intelligent web portals are developed. First concepts for the searchability of primary data are tested exemplary in the BW-eLabs. During the starting year of the project the 3D-infrastructure is established, embedding the already existing virtual laboratories and determining the intial connection to the eResearch environment. To this end the important components such as
256
Sabina Jeschke et al.
experiments, their life cycle, groups, and so on are defined, open standards are evaluated, and the first prototype of a remotely-controllable laboratory is implemented. Additionally, models of the implementation of digital holography into BW-eLabs are examined and evaluated. The second year is devoted to the extension and the development of the infrastructure, due to the newly collected requirements, and to the development of remote experiments as well as to the implementation of the usability studies. The first nanotechnological virtual laboratories and remote experiments are integrated and the entire process of data generation, filing, and access is reproduced and evaluated. In order to visualize real objects, initial methods of digital holography are included. A first prototype to semantic searchability of primary raw data is prepared and existing information resources are connected. For that purpose, a sophisticated policy-management for the integration of licensed data and open access concepts is prepared in co-operation with the university libraries and the partners. Evaluations of test users will take place. The third year is dedicated to consolidating services and infrastructure so that they can be sustainably integrated into production systems. By incorporating more institutions, additional scenarios are ascertained and if necessary further components will be supplied. Activities to widen the user-base that began in the starting year as well as community-building actions are strengthened (conferences, workshops, etc.). Technical support documents are provided to facilitate the integration of further experiments. Based on the evaluation, a plan is provided, not only for securing the results, but also containing a business model for forthcoming developments.
References [CJL+ 09]
[DCL08]
[dPTC08]
[EBHE05]
[EHGE06]
[esc10] [FMH08]
S. Cikic, S. Jeschke, N. Ludwig, U. Sinha, and C. Thomsen. VICTOR-SPACES: VIRTUAL AND REMOTE EXPERIMENTS IN COOPERATIVE KNOWLEDGE SPACES. In F. Davoli, N. Meyer, R. Pugliese, and S. Zappator, editors, Grid Enabled Remote Instrumentation, Signals and Communication Technology, pages 329–343. Springer, New York, 2009. Konferenz war 2007. L. Demarconnay, C. Coutanceau, and J.-M. Léger. Study of the oxygen electroreduction at nanostructured PtBi catalysts in alkaline medium. Electrochimica Acta, 53(8):3232–3241, March 2008. Marie Lopez del Puerto, Murilo L. Tiago, and James R. Chelikowsky. Ab initio methods for the optical properties of CdSe clusters. Physical Review B, 77(4):045404, January 2008. Bernd Eßmann, Peter Bleckmann, Thorsten Hampel, and Robert Elsässer. Distributed Persistence in CSCW Applications. Technical Report tr-ri-05-268, Heinz Nixdorf Institut, December 2005. Bernd Eßmann, Thorsten Hampel, Frank Götz, and Andreas Elsner. Embedding Collaborative Visualizations into Virtual Knowledge Spaces. In Proccedings of the 7th International Conference on the Design of Cooperative Systems (COOP 2006), pages 33–40, Carry-le-Rouet, France, 2006. eSciDoc Project webpage. http://www.escidoc.org/, last access 31.3.2010, 2010. A. Friedrichs, P. Meisen, and K. Henning. A Generic Software Architecture for a Driver Information System to Organize and Operate Truck Platoons.
Networking Resources for Research and Scientific Education in BW-eLabs
[Haa05] [HP03] [HSH+ 04]
[JNPT08]
[JNR07]
[JPT06]
[JRST07]
[MHS08]
[NEE08] [OF06] [RBT+ 09]
[Roj] [RSS+ 09] [SBB+ 06]
[Sch07]
257
In Vortragsveroeffentlichung, International Conference on Heavy Vehicles (HHVT2008), Paris, 19.-22.05.2008, Paris, 2008. A. Haake. CURE: Das CSCL-Portal der FernUniversität in Hagen - Benutzungshandbuch, 2005. K. Henning and E. Preuschoff. Einsatzszenarien für Fahrerassistenzsysteme im Strassengueterverkehr und deren Bewertung. Number 531 in 12. VDI Verlag, Düsseldorf, 2003. J.M. Haake, T. Schümmer, A. Haake, M. Bourimi, and B. Landgraf. Supporting flexible collaborative distance learning in the cure platform. Washington, DC, USA, 2004. IEEE Press. Sabina Jeschke, Nicole Natho, Olivier Pfeiffer, and Christian Thomsen. Networking Resources for Research and Scientific Education in Nanoscience and Nanotechnologies. pages 234 – 237, Melbourne, Australia, February 2008. IEEE Computer Press. Sabina Jeschke, Nicole Natho, and Thomas Richter. Integration Technology for Virtual Laboratories - Individualization and Flexibility through Computer Algebra Systems. In Proceedings of the 2nd International Conference on Engineering Education & Training - Preparing engineers to meet the challenges of the future (ICEET – 2), Kuwait, April 2007. S. Jeschke, O. Pfeiffer, and C. Thomsen. Vernetzung experimenteller Ressourcen in Forschung und Ausbildung für Nanotechnologien und Nanowissenschaft. In C. Hochberger and R. Liskowsky, editors, INFORMATIK 2006, Informatik für Menschen, Band 1, Beiträge der 36. Jahrestagung der Gesellschaft für Informatik e.V. (GI), Dresden, volume 1, pages 85–89, Bonn, October 2006. Gesellschaft für Informatik. Sabina Jeschke, Thomas Richter, Harald Scheel, and Christian Thomsen. On Remote and Virtual Experiments in eLearning in Statistical Mechanics and Thermodynamics. Journal of Software, 2(6):76–85, December 2007. ISSN1796-217X. P. Meisen, K. Henning, and T. Seidl. A Data-Mining Technique for the Planning and Organization of Truck Platoons. In Proceedings of the International Conference on Heavy Vehicles (HHVT2008), Paris, May 2008. David J. Norris, Alexander L. Efros, and Steven C. Erwin. Doped Nanocrystals. Science, 319(5871):1776–1779, March 2008. W. Osten and P. Ferraro. Digital Holography for the Inspection of Microsystems. In W. Osten, editor, Optical Inspection of Microsystems, pages 351–426. CRC Taylor & Francis, Boca Raton, 2006. Frank S. Riehle, Roland Bienert, Ralf Thomann, Gerald A. Urban, and Michael Krüger. Blue Luminescence and Superstructures from Magic Size Clusters of CdSe. Nano Letters, 9(2):514–518, February 2009. Team Raul Rojas. Spirit of Berlin, FU Berlin. http://robotics.mi.fu-berlin.de/pmwiki/, last access: 31.03.2010. Moritz Riede, Rico Schueppel, Kristian O. Sylvester-Hvid, Martin Kühne, Michael C. Röttger, Klaus Zimmermann, and Andreas W. Liehr. On the Communication of Scientific Results: The Full-Metadata Format. CoRR, 2009. G. Schneider, R. Backofen, H. -G Borrmann, W. Deck, A. Dedner, Luc De Raedt, K. Desch, M. Diesmann, M. Geier, A. Greiner, W. R Hess, J. Honerkamp, St Jankowksi, I. Krossing, A. W Liehr, A. Karwath, R. Kloefkorn, R. Pesche, T. Potjans, M. C Roettger, L. Schmiedt-Thieme, B. Voss, B. Wiebelt, P. Wienemann, and V-H. A Winterer. Bottom-up approach to grid-computing at a university: The BlackForest-Grid initiative. PIK - Praxis der Informationsverarbeitung und Kommunikation, 29(2):81–87, 2006. G. Schneider. Neue Medien als strategische Schrittmacher an der Universität Freiburg - wie Informations- und Kommunikationstechnologien Studium, Verwaltung, und Forschung verändern. Universitätsbibliothek Freiburg, 2007. Seiten 13-24.
258 [SGK+ 04]
Sabina Jeschke et al.
Werner Stephan, Bernhard Graf, Adalbert Kirchgässner, Sabine Laschat, and Wolfram Schüz. Das Zeitschriften-Paradoxon oder: Wer verfügt über wissenschaftliche Information?, volume 3 of Universitätsbibliothek Stuttgart, Bibliothek Bildung und Fortschritt. 2004. [SS05] Gerhard Schneider and Dirk Suchodoletz. Identity Management in der Praxis. In Jan v. Knop, editor, Lecture Notes in Informatics, number P-73, pages 255–264. Springer Verlag, 2005. [SS06] Frank Scholze and Werner Stephan. Retrieval on the Grid Results from the European Project GRACE (Grid Search and Categorization Engine). Zugang zum Fachwissen ODOK ’05, pages 118 – 127, 2006. Feldkirch: Neugebauer. [won] Wonderland (Sun) Project webpage. https://lg3d-wonderland.dev.java.net/, last access 6.8.2009. [WTWL07] Pin-Jiun Wu, Ku-Ding Tsuei, Kung-Hwa Wei, and Keng S. Liang. Energy shift of photoemission spectra for organics-passivated CdSe nanoparticles: The final-state effect. Solid State Communications, 141(1):6–11, January 2007. [YCZ+ 07] Lei Yang, Jinhua Chen, Xinxian Zhong, Kunzai Cui, Yan Xu, and Yafei Kuang. Au(at)Pt nanoparticles prepared by one-phase protocol and their electrocatalytic properties for methanol oxidation. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 295(1-3):21–26, March 2007. [YRG+ 09] Y. Yuan, F.-S. Riehle, H. Gu, R. Thomann, G.A. Urban, and M.J. Krüger. Nanosci. Nanotech, 2009.
Networked Virtual and Remote Laboratories for Research Collaboration in Natural Sciences and Engineering Sabina Jeschke, Arno Gramatke, Olivier Pfeiffer, Christian Thomsen, Thomas Richter
Abstract Based on the BW-eLabs platform, the goal of the N ET L ABS is the development of a software infrastructure, allowing for the interlinking and integration of experimental superstructures and simulations, as well as the software used for evaluation and archiving of data. In addition, a role- and rights-based model is developed, allowing access to experiments or measured data. The necessary components are integrated in the 3D Wonderland engine. Keywords remote experiment · virtual laboratory · metadata · digital holography · nanoscience · robotics
1 Introduction Experiments represent essential components of the research methodology in natural sciences and engineering. However, their execution is incumbent on numerous restrictions due to the financial equipment, spatial capacity, and support. The use of new media offers two essential concepts in order to reduce these challenges: • Virtual laboratories [RBJ09] are software environments providing a framework for execution of experiments at the computer. They can be used by arbitrarily groups simultaneously. Since experiments take place in a “virtual space”, setups are feasible that would never be possible in the “real life”. • Remote experiments [JRST07] are real experiments that are controlled by a location outside of a laboratory. They consist of two fundamental components: the real experimental setup and the technology of the remote access. Remote experiments permit the time and space independent use of natural sciences and engineering experimental setups.
Although numerous virtual laboratories and remote experiments are available, there are scarcely accessible repositories or portals of such experimental resources. Therefore, there are not only missing central access opportunities but also connections and networking of the laboratories among themselves. These connections could facilitate the transfer between similar experimental setups or experimental results. At the same time the link between virtual laboratories and remote experiments to other relevant scientific repositories such as digital libraries and scientific document management systems [esc] is missing. Thus, experimental setups and the executed experiments remain separated. The assignment of results to the setups as well as transparency and reproducibility of results is exceedingly difficult. Finally, collaborative methods are missed in isolated virtual laboratories and remote experiments. Nowadays, high performance research takes place only rarely in isolated laboratories and as individual performance, and scientific success is based increasingly on teamwork. This is not limited to local working groups – cutting edge research always take place above organizational boundaries and is geographically separated. Thus, the realization of scientific cooperation and cooperation in virtual knowledge spaces has a crucial relevance. The insufficient accessibility of experimental capacities is not limited to individual disciplines but it affects mainly cost-intensive experimental equipments. In a special way, this affects nanotechnology and robotics which have an extraordinarily influence on the current technological developments. Under the “extreme smallness” of the objects and effects, the requirements of high precision of measuring apparatus in nanotechnology is tremendous. The financial expenditure of preparation of necessary experimental equipment (electron microscopes, precise spectrometers, etc) is extremely. Generally, high professional equipment is only available for a small group of well-equipped institutions. Due to the complexity and the associated high costs of individual components that are used in the robotics (movement system, sensor system, etc.) only few laboratories can fall back to appropriate systems. Systems that integrated all these components such as human-machine cooperation (service robotic, assembly works in production) can be used today only by small group of researchers. The concept of virtual knowledge spaces became importance of academic education. In addition, virtual knowledge spaces are suited intensively as scientific co-operation tool (outriders: sTeam [ope], Cure [Haa05, HSH+ 04]. The fact is that many of early, visionary and revolutionary concepts not reach the desired level. Isolated laboratories for specific technical purposes were developed or begun with the development of portals for the integration of virtually and remote laboratories. However, the interface problems remained unresolved for the content wise networking of the individual components, which is crucial for the realization of complex research scenarios. The inter-connection of these technologies with the eSciDoc infrastructure - communication and a publication platform of the second generation for research institutions that focuses on the administration of scientific data - allows publications, visualization, and management of digital objects. The system goes thereby clearly over functionalities from publication repositories as for instance the common international repository DSpace [dsp], and Eprints [epr]
Networked Virtual and Remote Laboratories
261
or the German repository OPUS [opu]. The eSciDoc [esc] infrastructure with the repository software Fedora [fed] represents a generic infrastructure considered aspects of quality of data, the administration of data and the long-term availability. Characteristics are persistent quoting and referencing by assignment of appropriate identifier, automatic extraction of technical meta-data, and administration of objects with different meta-data models. By illustrating of formal and semantic relations under objects, the complete “life cycle” of scientific documents or data and their connection during the research process is provided. The transfer of real world concepts into virtual worlds plays a crucial role. Here, the research within the field of CyberWorlds [Wat09, pro] and Gaming Engines is an important motivation. Web service technologies and service-oriented architectures, open 3D-engines, developments of the social web and digital libraries techniques leverage the originally objectives. N ET L ABS is an open source 3d software platform for the external access to geographically distributed, heterogeneous virtual and remote-controllable experimental resources. Particularly, in the cost-intensive fields of nanotechnologies and robotics, we have limited access to professional experimental equipment. It is precious to get access for all technically involved communities to such experiments.The objective is the exploitation of experiments and their raw data as essential requirement for the realization of efficient, constant, and cooperative environments in natural sciences and engineering. The projects “eSciDoc” [esc] and BW-eLabs [JBH+ 09, JBH+ ] outline the technological fundament of this portal.
2 Portal Integration Based on the platform provided by the BW-eLabs [JBH+ 09, JBH+ ], the goal of the working programmer’s points of focus is the development of a software infrastructure, which allows the interlinking and integration of experimental superstructures and simulations, as well as the software used for evaluation and archiving of data. In addition, a role- and rights-based model is developed, which allows access to experiments or generated data. Therefore, the necessary components are integrated in the 3D Wonderland [pro] engine. Main focus point A is therefore subdivided in the following three tasks:
2.1 Creating a Subject Ontology to Distinct Experiments from Simulations: With the help of an ontology, the existing actuators and sensors of an experiment can be described- the input and output data of a simulation or an evaluation algorithm respectively. This way, the interconnection of experiments outside of laboratory limits becomes possible. In collaboration with main focus point E, the measurements and variables are converted into a so-called “Feature vector”, which guarantees the
262
Sabina Jeschke et al.
interchangeability and interpretation of data. What is also achieved through the semantic distinction of superstructures –by means of interlinking heterogeneous components- is the interchangeability of components for the purpose of either reproducing results or replacing single components in case of failure. In addition and parallel to this, information based mechanisms used for data analysis are introduced. With other words, the processing of experimental data in due consideration of all meta-information applied, such as physical unities, etc. In this case the preliminary work done for the Freiburg FMF is used, which already shows that such an approach can lead to an actual complexity reduction. This way, scientists who work with experimental data can apply and develop complex data analysis algorithms.
2.2 Developing a “Service Broker” as Part of a Roleand Rights-based Module: By means of the service broker, available resources can be booked; it also enables the access to raw and experimental data. For the authentication of the user, Shibboleth technology is used. Any access is automatically recorded and linked to experimental data in a suitable manner. This enables a complete reconstruction of the procedure attempted, including input and output parameters, participants and other conditions (possibly ambient temperature, light settings, etc.). This process serves both for the reproduction and verification of results, but also for protecting any intellectual properties of the experiment. Therefore, an “electronic lab book” is created, which facilitates the comprehension of experiments in a more efficient manner, without any additional expenditures.
2.3 Integrating the Main Focus Points in the 3D Wonderland Engine (Oracle, Formally Know as Sun): The purpose of this sub-project is to supply additional interfaces for virtual and remote experiments in the 3D Wonderland engine. This sub-task also aims at including in the project the previously described, ontology based experiment management system along with the service broker. In addition, suitable software interfaces for simulations, numerical algorithms and access controls are created in the Wonderland.
3 Nanotechnology Nanotechnology deals with the synthesis, qualities, characterization and application of materials which move within an area dimension of 1–100 nm. During the course of the last 15 years, nanotechnology started to become an interdisciplinary science, resulting in the establishment of subject specific Bachelor and Master Studies at
Networked Virtual and Remote Laboratories
263
Universities and Colleges. Strong interdisciplinary factors led to active exchange and overlapping of chemistry, physics, material sciences and in some situations even biology. The extensive number of scientific methods and rudiments necessary for the realization of research, stimulates and demands the joint collaboration of diverse study groups, should they choose to experiment, evaluate and publish any subject related works. This led to the construction of larger scientific centers in countless universities worldwide. In particular, the acquisition and common usage of larger, more complex devices can thus be realized in some of these centers. On account of the interdisciplinary and interlinking necessities, nanotechnology is an extremely fitting approach to be used within the BW-eLabs – in the form of interlinking virtual and remote laboratories to set up new standards in the knowledge management field [JPT06, JBH+ 09]. The Berlin Institute of Technology, and in particular the Thomsen study group [rem], disposes of extensive experiences in terms of nanostructure research at a top international level. Hereto, the research done on carbon nanotubes and related systems should be mentioned (nanoribbons, fullerene, graphene) but also the research on Si-nanowires and CdSe-nanorods. The experimental and theoretical preparations deliver the ideal conditions for the implementation of remote-nanotechnology labs. The available devices used nowadays in nanostructure research, are made accessible to the scientific community within the N ET L ABS scheme: • A Remote-Raman System with physical characterizations of carbon nanotubes is provided (e.g. the diameter regulations and grouping effects were examined). Raman spectroscopy permits a contact free analysis of physical properties and is by means of purely optical technology also indestructible. The provided RemoteRaman spectrometer has a stimulating wavelength of 532 nm and covers a spectral range from −1400 cm−1 to 3300 cm−1 . • A second device which can also be used universally and by many study groups is the scanning electron microscope. This device allows for image production, as well as other characterizations of nanostructures. Both experimental superstructures are integrated in the portal within the scheme of the N ET L ABS , something made possible with minor technical extensions. The scanning electron microscope has an accelerating voltage of 5 kV and a continuously adjustable spatial resolution of up to 30 nm; it is therefore ideal for the characterization of nanostructures. Both superstructures, which complement the “brief and extensive analysis “already commenced in the BW-eLabs, require suitable scheduling, a test dispatch and onsite support. Nevertheless, the main focus points during the term of the project are as follows: the integration of systems in the 3D Wonderland engine [pro], the development of more suitable interfaces between data delivered by the devices and data infrastructures originating from all focus points, the co-development of specialist nanotechnology ontologies, and the co-development of an intelligent access control system.
264
Sabina Jeschke et al.
4 Robotics Robotics is one of the key technologies present in the 21st century. The essential research areas focus on perception, actuator engineering, apperception and decision making. In terms of perception, the question of sensor technology and sensor combination arises. In the area of actuator engineering, manipulators and movement systems of different areas of application are decisive. And with respect to apperception and decision making, the central points are data processing and filtering of knowledge representations, along with path finding and decisive derivation. In the N ET L ABS project, the cost-intensive aspects required to follow through with the investigation of the previously mentioned areas are reduced. Due to the accessibility of the different robotics applications and by means of evaluating any attempts already made in the decision making phase on an identical platform, new trials are tested faster and at a reasonable price. The environments where these systems can be examined in both a simulative as well as a realistic manner are manifoldtwo areas of application protrude in the context of the N ET L ABS project: • The application ability of robotics systems in production surroundings continues to rise steadily; the necessary cooperation scenarios (between robots or between human and machine) lead to supplementary research questions which must be answered. Further investigations however – in the case of realistic scenarios, e.g. with a collaborating installation- remain financeable only to a few research teams. By means of suitable remote laboratories offered in N ET L ABS, production based research – also interdisciplinary–is carried out in multi-robot-applications and in human-machine cooperation scenarios; any gained primary data is made available to other institutes. Human-robot interaction becomes a significant application within the project. • Autonomous vehicles started to play a more significant role in traffic technology. N ET L ABS integrates movement-supported simulators in the form of virtual laboratories that allow testing of passenger reaction, recognition of danger situations, and even help simulate path finding algorithms. Expensive, cost-intensive environments are linked and made accessible to other research teams. The simulators available within an institute are extended by means of functionalities, and thus allow external control. In these areas of application, security plays a decisive role. This aspect must be taken into consideration when extending remote laboratories during the course of the N ET L ABS project. In addition, measures must be taken to examine how danger scenarios for both human and material can be prevented in remote controlled laboratories. Aspects such as working and IT security etc. must be taken into consideration. This is especially significant in human- machinecooperation scenarios. In this case, diverse concepts are developed to guarantee the safety of the person engaging with remote controlled machines. In addition, robotics acts as a “connection discipline” within this application: actuators which control superstructures/experiments are necessary for remote controlled laboratories. The aim to be achieved in the long run is a reconfiguration of
Networked Virtual and Remote Laboratories
265
laboratory environments; this will lead to a higher flexibility. Moreover, an infrastructure of intelligent “robots”, or better said “robot arms” must be created with the help of which experiments are controlled and laboratories varied. Within the scope of focus point C, additional concepts are developed, prototypically implemented and fitted accordingly to N ET L ABS.
5 Digital Holography One of the aims pursued within the N ET L ABS project is to make digital holography technology [OF06] available to a wider user community. Unlike conventional optical measuring technologies, holography uses a 2-dimensional recording system with the help of which it registers and reconstructs the 3-dimensional wave front in both amplitude and phase, with other words -in the case of a few specific wavelengths the complete optical information. Digital holograms add further advantages to classical characteristics of analogous holograms: • Reconstruction: Additional to the direct, optical reconstruction of the wave field and by means of spatial light modulators, or in an analog way to the classic case, the wave field can also be reconstructed on the computer. • Real-time property: Thanks to the quickly growing bandwidth and computing power, one can record a digital hologram, transfer it to any location worldwide and reconstruct it on a real-time operating system. Digital holography is therefore ideal for the optical transference of information in remote experiments; the miniaturization of devices and the ruggedness of the procedure coincident with an increase in measuring accuracy continue to be challenge areas. Alongside the general suitability for optical sensors and representation systems, digital holography opens unique possibilities in terms of measuring technology: • Short-coherent digital holography is especially suited for microscopic investigations: with the short-coherent method different levels of an object can be shown, and the layer thickness reconstructed up to 20 micrometers. Additionally, the technology can also be used in endoscopic procedures; by means of digital holography, the elasticity of different interiors can be determined and thus, the consistency of object as a whole can be deduced. • Comparative digital holography offers a new coherent-optical procedure. This allows the comparison (e.g. form or distortion) of two nominally identical but physically different objects (Pattern-Test-Comparison). By combining the principles of digital and comparative holography - comparative digital holography (VDH) - adaptable testing methods can be created. This way, there is no need to simultaneously apply the pattern and test functions when a comparison is made. In this manner, inspection results and quality statements based on a real-time operating system can also be obtained in the case of complex test objects.
266
Sabina Jeschke et al.
• Holography with multiple light sources, or light sources of different wavelengths permits a precise body surface measurement up to a fraction of the wavelength of the used laser. This enables a detailed analysis of surface contours, which plays an important role in both non-destructive testing methods as well as reverseengineering. The construction of an efficient infrastructure for the transference and storage of digital holograms is a valuable, sensor based enhancement for remote control experiments. This development is significant as it allows worldwide access to the complete optical information of an object; it also enables the direct measurement of properties.
6 Data Infrastructure Within the scope of the N ET L ABS project, the eSciDoc infrastructure [esc] – including the still to be developed extensions from the promoted BW-eLabs projectis extended and adjusted. As an “enabling technology”, eSciDoc supports the increasingly stipulated replicability and reusability of research results. It does so by interlinking data and documents/publications with their availability. The architecture of eSciDoc is based on two central layers: “eSciDoc Solutions” provides scientists with concrete working appliances in the form of applications, and focuses specifically on the discipline or on the field of work. For the purpose of creating a service-orientated architecture, the following layer, the “eSciDoc Infrastructure”, provides among other things a huge number of generic and comprehensive service suppliers, long-term storage, management, presentation and visualization, as well as the publication of digital objects. In this manner, eSciDoc covers the entire research process along with the arising resulting artifacts - from any experimental disciplines about the development of electronic lab books to scientific information repositories for the investigation of related questions. The eSciDoc infrastructure does not only take into consideration published results (e.g., in the form of academic articles), but also research and primary data along with various data information. In this case the complex dilemma of retrieving primary data is addressed: within the scope of the BW-eLabs project, primary data catalogues, along with classical bibliographical indices, indicate relevant physical dimensions. For this, a new classifier is used, one that can convert physical size into a feature vector: complex objects are characterized by an n-tuple displaying figures, so that by means of an appropriate distance determining function, still to be characterized objects can be compared and any resemblance can thus be detected. This approach allows for the first time a semantic search within primary data records. The semantic search system will be transferred and - where needed- fitted to the N ET L ABS setting; this still to be developed process will be carried out in the scheme of the BW-eLabs project. In addition, a search system will be developed to facilitate the discovery of labs and lab components; in terms of the usability of the
Networked Virtual and Remote Laboratories
267
portal, this process is particularly important for the further development of national and international groups (users, as well as laboratory providers). According to the requirements of the software components used, additional extensions - in terms of services and adapting interfaces - are carried out. It can be assumed that the interfaces and services developed for the virtual worlds presented in the BW-eLabs project can be adopted and reused. This approach particularly requires the robotics discipline extensions: data structures and metadata concepts for static and dynamic data of robotics are developed and as follows, integrated in the eSciDoc infrastructure [esc]. Already available decentralized services and resources of information – e.g. publications and data repositories relevant for respective communities – are integrated according to current license models present in the eSciDoc legal management system. Ultimately, extensive support is provided for the integration in the planned virtual research environments of the N ET L ABS project; this concerns in particular interfaces and the exchange formats of platforms.
7 Related Work A project similar to that of LiLa is driven by the MIT in the USA: Similar to N ETL ABS: the iLabs project [HdAL+ 08, ila] supported by Microsoft aims at making experiments remote-controlled, and thus having them accessible by web-services for other members over the internet. Very similar to iLabs, N ET L ABS is not fixed to a certain topic, but addresses all engineering and natural scientific fields. Similar to iLabs, we aim at a “single sign on” process to gain access to our resources. This process will be integrated into the Wonderland architecture. The Blekinge Institute of Technology in Sweden started in 2007 the VISIR project [GZH+ 07] Quite similar to N ET L ABS and iLabs, VISIR tries to increase laboratory utilization by sharing remote-controlled equipment across universities. Another similarity is that as many other projects [BI06, JRS+ 05] deploy LabVIEW [lab] by National Instruments to connect experiments to the internet.
8 Conclusion Even though N ET L ABS is an ambitious project for constructing an infrastructure for “virtual” experiments, we want to stress that it is not our aim to substitute the traditional experiment as its value lies beyond the gained scientific insight, namely in training the social communication skills with colleagues. Even though we try to mimic these structures as far as possible in the virtual world, their replication remains necessarily incomplete. Instead, we expect that N ET L ABS makes the best out of the financial problems universities are facing – and establishes a strong federation of universities that is able to support their scientists better than a single isolated institution could.
268
Sabina Jeschke et al.
Acknowledgments The authors like to thank the DFG (German Research Foundation) for funding N ET L ABS.
References [BI06]
H.A. Basher and S.A. Isa. On-campus and Online Virtual Laboratory Experiments with LabVIEW. In IEEE SoutheastCon, pages 325–330, 2006. [dsp] DSpace website. www.dspace.org, last accessed 23.9.2009. [epr] Eprints website. www.eprints.org, last accessed 23.9.2009. [esc] www.escidoc.org, last accessed 23.9.2009. [fed] Fedora. www. fedora-commons.org last accessed 23.9.2009. [GZH+ 07] I. Gustavsson, J. Zackrisson, L. Håkansson, L. Claesson, and T. Lagö. The VISIR project − an Open Source Software Initiative for Distributed Online Laboratories. In Proc. of Annual Int. Conf. on Remote Engineering and Virtual Instrumentation, 2007. [Haa05] A. Haake. CURE: Das CSCL-Portal der FernUniversität in Hagen − Benutzungshandbuch. http://teamwork.fernuni-hagen.de/CURE/doc/manual.pdf , last accessed 23.9.2009, 2005. Hagen, Germany: FernUniversität Gesamthochschule. [HdAL+ 08] V.J. Harward, J.A. del Alamo, S.R. Lerman, P.H. Bailey, J. Carpenter, K. DeLong, C. Felknor, J. Hardison, B. Harrison, I. Jabbour, P.D. Long, M. Tingting, L. Naamani, J. Northridge, M. Schulz, D. Talavera, C. Varadharajan, W. Shaomin, K. Yehia, R. Zbib, and D. Zych. The iLab Shared Architecture: A Web Services Infrastructure to Build Communities of Internet Accessible Laboratories. IEEE, 96(6):931–950, June 2008. [HSH+ 04] J.M. Haake, T. Schümmer, A. Haake, M. Bourimi, and B. Landgraf. Supporting flexible collaborative distance learning in the cure platform. Washington, DC, USA, 2004. IEEE Press. [ila] iLabs: Internet access to real labs - anywhere, anytime, available. http://icampus.mit. edu/iLabs/, last accessed 23.9.2009. S. Jeschke, B. Burr, J.-U. Hahn, L. Helmes, W. Kriha, M. Krüger, A.W. Liehr, [JBH+ ] W. Osten, O. Pfeiffer, Th. Richter, G. Schneider, W. Stephan, and K.-H. Weber. BWeLabs − Knowledge Management in Virtual and Remote Labs. [JBH+ 09] S. Jeschke, B. Burr, J.-U. Hahn, L. Helmes, W. Kriha, M. Krüger, A.W. Liehr, W. Osten, O. Pfeiffer, Th. Richter, G. Schneider, W. Stephan, and K.-H. Weber. Networking Resources for Research and Scientific Education. In 3rd IEEE International Workshop on e-Activity, Daegu, Korea, May 2009. [JPT06] S. Jeschke, O. Pfeiffer, and C. Thomsen. Vernetzung experimenteller Ressourcen in Forschung und Ausbildung für Nanotechnologien und Nanowissenschaft. In GI Jahrestagung (1) 2006, pages 85–89, 2006. [JRS+ 05] S. Jeschke, Th. Richter, H. Scheel, R. Seiler, and C. Thomsen. Das Experiment und die eLTR-Technologien: Magnetismus in Virtuellen Laboren und RemoteExperimenten. Bonner Köllen Verlag, 2005. LNI. [JRST07] S. Jeschke, Th. Richter, H. Scheel, and C. Thomsen. On Remote and Virtual Experiments in eLearning in Statistical Mechanics and Thermodynamics. In Innovations in E-Learning, Instruction, Technology, Assessment and Engineering Education, pages 329–334. Springer, Dordrecht, NL, 2007. [lab] LabVIEW by National Instruments. http://www.ni.com/labview , last accesses 23.9.2009. [OF06] W. Osten and P. Ferraro. Digital Holography for the Inspection of Microsystems. In W. Osten, editor, Optical Inspection of Microsystems, pages 351–426. CRC Taylor & Francis, Boca Raton, 2006.
Networked Virtual and Remote Laboratories [ope] [opu] [pro] [RBJ09]
[rem] [Wat09]
269
Open steam. www.open-steam.org, last accessed 23.9.2009. OPUS. www.ub.uni-stuttgart.de/wirueberuns/projekte/opus4/. Project Wonderland website. https://lg3d-wonderland.dev.java.net/, last accessed 23.9.2009. Th. Richter, D. Böringer, and S. Jeschke. LiLa: A European Project on Networked Experiments. In Proceedings of the 6th International Conference on Remote Engineering & Virtual Instrumentation, Bridgeport, CT, June 2009. Remote Farm. http://remote.physik.tu-berlin.de/farm/index.php?id=1&L=1, last accessed 23.9.2009. John K. Waters. Sun Makes Its MUVE, sidebar in “A ‘Second Life’ For Educators”. T.H.E. Journal, January 2009.
Natural Sciences in the Information Society First Experiences Grit Köppel, Sabina Jeschke, Nicole Natho, Lars Knipping, Grit Petschik, Christian Schröder, Erhard Zorn
Abstract The goal of the GALILEA project is to design and implement innovative programs and curricula, providing solutions to the changed job specifications for engineers and natural scientists and are capable of attracting more female students to these programs. In this article we outline the design, implementation and the first evaluation results of our pilot program, the bachelor course of “Natural Sciences in the Information Society” that started in the winter term of 2007/08. Keywords engineering education · natural sciences education · gender studies · e-education
Fig. 1 Left: Number of students in engineering and natural sciences programs from 1980-2004 in Germany (pre- and post unification), Right: Graduates in engineering and natural sciecnes per 1000 citizens in selected European countries as of 2003. Graph published in Spiegel Online [ONL], translation added by authors
of “standard” male students [Col]. In addition, not only women would benefit from such an alteration of curricula also supporting non-technological skills and expertise (e. g. soft skills, analytical competencies, or information literacy). The demands of the economy of the 21st century such as lifelong learning and the effects globalization has on today’s employees require engineers and natural scientists to be autonomous, and disciplined. Employees must have sophisticated communication skills such as speaking different languages and working in teams with different cultural backgrounds. In order to accept the challenges, universities have to reconsider their structural and educational concepts. Due to the Bologna process, European universities have the opportunity to reform their curricula. Within the Galilea project [EJN+ 07] our goal is to design and implement such innovative curricula, which answer to the changed job specifications of engineers and are capable of attracting more female students. Our first new Bachelor of Science program “Natural Sciences in the Information Society” started in the winter term of 2007/08. If offers a studium generale of natural sciences at Technische Universität Berlin (TU Berlin) and can be continued either by a corresponding Master of Science course “Natural Sciences in the Information Society” or Physics, Mathematics, Computer Science or Chemistry. This article reports on our first experiences, and evaluation results.
2 The Galilea Project Originally, Galilea was established at the Department of Mathematics and Natural Sciences; however, it is operating as a supporter for many courses at the entire university. The aim of the Galilea project is to design and implement new
Natural Sciences in the Information Society First Experiences
273
Fig. 2 Wishes of the industry regarding programs. Source [VDM], translation added by the authors
gender-sensitive courses within technical and scientific disciplines. Therefore new curricula have to be proposed, combined with modern educational styles overcoming the above described challenges. In this manner, practical aspects as freedom of scope, comprehensive projects, teamwork, and an internship play an important role. The program has been designed with educational preferences of women in mind. It allocates educational key qualifications and interdisciplinary skills as well as leadership and management qualities. Since language requirements have become important, especially in technical and scientific fields, we decided to offer at least some of the courses in English (Figure 2). One of the main deficiencies in academic education in Germany is the low attendance of young students in large universities. Galilea has an integrated mentoring program, especially for freshmen but also for older students.
2.1 Natural Sciences in the Information Society The “Natural Sciences in the Information Society” (NidI) program as the first Galilea course started in the winter term of 2007/08 with 15 students (9 female and 6 male) and 37 students (17 female and 20 male) in the winter term of 2008/09. It offers a wide repertory of natural sciences, engineering and non-technological courses at TU Berlin. The core attributes of all natural sciences are the close correlation between theory and experiments and the high standards in mathematical and computer sciences education. These connections are the guidelines of the two-tier program: 2.1.1 Bachelor program The NidI Bachelor program provides a great range of access to the basics of natural sciences. The curriculum of this three year program is built on a theoretical basis (59 % of the credit points) and supported by a compulsory elective (21 %) part, a freely chosen part (10 %), an internship (3 %) and a bachelor thesis (7 %). In fact,
274
Grit Köppel et al.
students have about 110 compulsory elective subjects to enrich their individual study plan. The technological broadness of the curriculum accommodates the multidisciplinary interests of women. Therefore, emphasis is put on integrating natural and life science aspects. During the concentrated study, the mandatory internship (of at least 12 weeks in duration) has to be concluded. It ought to give an insight into professional life. The theoretical basis of the curriculum is formed by mathematical and physical courses supplemented by several courses in computer science. Additionally, in the first year there are two mandatory modules in scientific information management chosen selected by at least one the following criteria: • • • •
Content-related course of study, Teamwork in co-ed teams, Teaching core skills, Project-oriented work.
Students are taught general scientific methods they will need for managing projects (i. e. experiments), regardless of the explicit fields they will specialize in. Therefore we provide two new project-oriented Bachelor program lectures “Scientific Information Management” and “New Media in Teaching and Research”. The students are taught basics of knowledge management, presentation techniques, multimedia education, and research. The first course is carried out by the staff of the university library, the second by the Center of Multimedia in Education and Research of TU Berlin.
2.1.2 Master program After graduating with a bachelor of “Natural Sciences in the Information Society”, students have the opportunity to join a master program that will result in the “Master in the Natural Sciences in Information Society”. Since the bachelor program offers a wide basis in the natural sciences it is also possible to join other master programs from the natural sciences (i. e. astrophysics or nanotechnology).
2.1.3 Mentoring program Students can participate in an internal mentoring program providing organizational, social, and technical aspects. This program aims at improving the educational and organizational atmosphere in large universities by a close mutual relationship between freshmen, older students and academic staff. Our goal is to increase the motivation and performance especially of freshmen and women. Beside individual meetings, we provide a wide offering of social events, and additional nontechnological and technological advanced training (i. e. exam stress, programming, or mathematical workshops). The mentoring program also allows us to react quickly if structural or technical problems of the program become apparent.
Natural Sciences in the Information Society First Experiences
275
2.2 Aim of the Program The industry and economy still have a high demand for interdisciplinarily educated graduates with great scientific knowledge (cf. fig. 2). Multi-disciplinary, application and research-oriented programs impart methods and fundamentals of computer science, mathematics and natural sciences. In the bachelor program students acquire the necessary knowledge and familiarize with general and specific methods for treatment and solution of problems in natural sciences. This enables them to transfer their knowledge to practice and create the basis for continuing their academic studies, e. g. in the master program. The following cross-technical competencies and social skills are mediated beyond the purely technical aspects: • responsible life-long learning, • problem-analysis and development of problem solving concepts, • social, scientific, gender-specific and ethic points of view in action and decision strategies, • multi-disciplinary communication and ability to work in a team, • presentation skills including the presentation of scientific results for different target audiences, • modern methods of scientific information management. After graduation the students are able to find jobs that require great scientific and methodological knowledge whereas the specific skills are acquired on the job. Some examples are: technology writers and other activities with scientific publishers, scientific librarians, advisory activities in politics/ministries/authorities, projectmanagement in scientific-technical areas, science management at universities and research institutes, activities in financial and insurance companies. Most of the structure of this multi-disciplinary program is tailored to suit women’s preferences, yet the large amount of experimental modules might be a problem. Due to their socialization women frequently underestimate their abilities and do not possess the same degree of experience as men [Gun03]. Experiments and theory are closely connected in natural sciences. Consequently it is neither possible nor desirable to design programs in these areas without experimental components. On the contrary: strict emphasis has to be put on the connection between theory and experiment and on offering additional possibilities for experimenting, e. g. virtual laboratories [JBH+ 09]. The bachelor and master programs “Natural Sciences in the Information Society” are the basis to modernize programs, e.g in physics where in general the ratio of female students is one of the lowest of all programs. The development of a program in astrophysics as the field of physics that is traditionally very attractive for female students could be an example. This is reflected in the numbers of students graduation in physics at TU Berlin: Only about 10 % of all physics students are women, yet approximately 30 % of all female graduates and scientific assistants choose astrophysics as their specialization.
276
Grit Köppel et al.
For a detailed summary of the “Bachelor of Natural Sciences in the Information Society” programs please see [JNPZ09].
3 Evaluation of the New Courses Overview So far students’ feedback indicates that we are on the right track. We learned that the high relevance of practical courses and projects in these modules, the extensive access to laboratories and independent experimenting, the possibility of a wide choice and the internship are of special interest for our female students. The mentoring program of the course also allows evaluating the success of the program itself. This feedback from students is crucial, especially in the first year, in order to make adjustments to the program. This hopefully leads to more support from graduates in the future, e. g. by becoming mentors or providing internships for students. We evaluated both courses and found out that the concept was generally well accepted. Male students appreciated the courses focusing on gender aspects even more than female ones. The courses were offered a second time, but an evaluation has not been conducted so far To find out if and how the students’ point of view changes during the term, each course was evaluated twice. There was a weekly poll for “Scientific Information Management” and a final questionnaire evaluation, whereas “New Media in Teaching and Research” has been assessed monthly and ended with a final questionnaire session. The results will be presented in brief.
3.1 Integrated Lecture: Scientific Information Management 22 students took part in the weekly evaluation over a total of nine weeks. Every time five male and five female randomly-sampled students were interviewed. The male students followed the lessons a bit more concentrated and took part more actively than female students did. They also expressed to have learned a lot for use in the future. In the final evaluation, the students evaluated the course with average grades. They expressed that they had learned a lot and were able to transfer it into practice. The students were able to handle the level and scope of this course very well. Different kind of media used in the lessons motivated the students. They preferred to work in mixed teams. The students even appreciated the mandatory attendance of this course because they understood that it was adequate due to the teaching methods applied [JNPZ09].
3.2 Integrated Lecture New Media in Teaching and Research Each of the 20 students (12 male and 8 female) and the three lecturers were interviewed to compare their impressions of the students’ participation during the
Natural Sciences in the Information Society First Experiences
277
lessons (monthly evaluation). While the assessment of their own contributions to the lessons was balanced around a good average, the lecturers were more pleased than the students thought they would be. However, this changed in the second half of the term. For the final evaluation, only students were interviewed. They considered this course to be useful for their future and that they would use the learnt methods (e. g. presentation skills) in their everyday life. They really liked this course and graded it with an average of 1.75 on a scale from 1 (best) to 4 (worst). The level and scope of this course were adjusted to the students’ previous knowledge. There was no mandatory course attendance, yet 45 % of the students visited almost every lesson, 70 % attended more than half of all lessons. Four of them would have even preferred a compulsory attendance. The students enjoyed the different kind of examinations (“prüfungsäquivalente Studienleistungen”): There were four different types of exams: an oral exam, several assignments, delivering a scientific paper and a presentation. The students appreciated the oral exam the most (average: 1.42 on a scale from 1 to 4) and the assignments the least (average: 1.95). Referring to these types they claimed to be able to present the topic and their knowledge more easily. Altogether, we found that the concept was well accepted both by male and female students. Besides the acquired knowledge students developed useful soft skills and applied them immediately in other courses. Although there are some aspects that still need to be improved, the concept seems to be successful. During the present term similar evaluations are carried out to validate these results and to find out what has improved and what else needs to be done.
3.3 Mentoring The evaluation of the mentoring program was included all participants – mentors and mentees. All mentors were evaluated through means of a short questionnaire, and all mentees completed four guided interviews with some additional questions included in a short questionnaire. Two female mentees and two male mentees filled out the questionnaires. Respectively, one of each mentees was actively participating in the mentoring program. At the time of the questioning, all mentees were at the end of the second term. Generally, the acceptance of the mentoring program by active mentees was positive, whereas the temporal effort strongly varied (2 to 9 hours per term). But only one non-active mentee positively reviewed the program. According to the asked mentees the aim of the program is to assist students through a contact person during their studies and especially at the beginning. However, the interviewees seem to have no specific conception of the program. Therefore, the mentoring program was used mostly at the beginning of the studies, in order to clarify organizational questions. If no concrete questions arose, the mentees did not make use of the program. It was criticized that there were difficulties to make appointments and that clear guidelines were missing. Both mentors and mentees should have had a clear conception of the meaning of mentoring. However, if there was a personal relationship between
278
Grit Köppel et al.
the mentor and the mentee, the conceptions and expectations of the mentees were satisfied. This relationship was characterized as amicable. The general aspects of the mentoring program were of interest to the mentees. This part of the program was also used in parts if the student did not have a mentor. An obstacle of active participation in the mentoring program is represented by the first appointment with the mentor. If this took place, then generally a good relationship between the two formed. However, different challenges appeared due to the status group of the mentor (student, scientific staff, professor). Thus, appointments with the status group “professor” hardly took place. In contrast, friendly relationships arose out of the status group “students” in which the mentoring becomes less important.
4 Summary and Future Work Especially the German industry needs more qualified engineers and natural scientists at this point in time. The number of students currently enrolled in the corresponding programs is too small to fill this gap. While more female students could stand in, the majority of these programs are rather unattractive to women. In particular, many students in these fields are not well prepared for their future professional life. Thus, important soft skills or communications skills are needed. Numerous evaluations [Col], [Sch99], [Sch78] prove that the quota of women in the programs mentioned is considerably increased by special adaption to their needs (e. g. multidisciplinary). However, this can only be achieved if the curricula are readjusted. Moreover this can also motivate numerous male students. The awareness in Germany to readjust courses of studies is developing very slowly, yet constantly growing. In addition, there are many efforts to encourage female students. Thus, female students will be supported from kindergarten onward all the way to studies at the university level through scholarship programs. However, the criticism grows that many male students are now facing disadvantages due to the changing gender stereotype settings and activities [e.V]. Therefore a lot of male school graduates would not continue to university any longer and would also not come into consideration for a qualified profession. At TU Berlin the mentioned challenges are well-known, and different solutions are being sought out. TU Berlin strives to be a forerunner for new challenges in research, economics, and society, in order to achieve a new image of engineering and natural sciences. One of these approaches is the Galilea project financed by the European Social Fund (ESF). Galilea is designing programs in order to increase the quota of women in engineering and natural sciences by integrating specific female requirements and new educational paradigms. Through the Bologna declaration [eur99], [con] the incentives for completely new programs are provided since the two-tier bachelor and master system is a completely new structure in the academic education in Germany. Therefore, the Galilea project designed the new program “Natural Sciences in the Information Society” aiming at a quota of about
Natural Sciences in the Information Society First Experiences
279
50 % female students. This new approach is integrated not only into the concept of the entire program but also through the individual lectures “Scientific Information Management” and “New Media in Teaching and Research”. These two lectures for the Bachelor program were evaluated. Generally, the lectures were positively assessed. During the first run of these lectures students identified some technical procedures such as high expenditure of work or too many examined topics to be disadvantageous. The multidisciplinary approach was assessed as good, and the students had the feeling that the lectures impart soft skills that can be useful in further studies or future jobs. In addition to the changed curriculum a mentoring program is integrated into the Galilea program, accompanying students throughout their studies. An evaluation of the mentoring program took place as well. Here, the biggest issue found was that the meaning of the notion “mentoring” is not universally understood, in spite of efforts of explanations in training courses. Students are already aware of the existence of a contact person, but this offer is accepted mainly in the first term. Nevertheless, the program was evaluated positively. A good approach was to enlist sophomore students as mentors for their freshmen fellow students. The developing relationships are considered amicable and lasting. In further evaluations, we would like to examine specific problems within the lectures and the mentoring program to improve our courses.
References [Col]
S. Collmer. Wie Gender in die Technik kommt – Computerkompetenz für Frauen. Talk, available online. [con] Confederation of EU Rectors Conferences and the Association of European Universities (CRE). The Bologna Declaration on the European space for Education: an explanation. Technical report. [EJN+ 07] Maria Elsner, Sabina Jeschke, Nicole Natho, Olivier Pfeiffer, and Christian Schröder. Attractive Universities: New Curricula in Natural Sciences and Engineering. In Meeting the Growing Demand For Engineers and Their Educators 2010 - 2020 International Summit (IEEE), pages 1 – 7, Munich, Germany, November 2007. IEEE Computer Press. [eur99] European Ministers of Education. The Bologna Declaration of 19 June 1999. Technical report, 1999. [e.V] Kompetenzzentrum Technik-Diversity-Chancengleichheit e.V. Neue Wege für Jungs, Ein geschlechtsbezogener Blick auf die Situation von Jungen im Übergang Schule Beruf. Technical report. [Gun03] Cathy Gunn. Dominant or different? Gender issues in computer supported learning. Journal of Asynchronous Learning Networks, 7(1):14 –30, 2003. [JBH+ 09] Sabina Jeschke, Barbara Burr, Jens-Uwe Hahn, Leni Helmes, Walter Kriha, Michael Krüger, Andreas W Liehr, Wolfgang Osten, Olivier Pfeiffer, Thomas Richter, Gerhard Schneider, Werner Stephan, and Karl-Heinz Weber. Networking Resources for Research and Scientific Education in BW-eLabs. In Software Engineering, Artificial Intelligences, Networking and Parallel/Distributed Computing, 2009. SNPD ’09. 10th ACIS International Conference on, pages 47–52, Daegu, South Korea, May 2009. IEEE.
280 [JNPZ09]
[oE] [ONL]
[Sch78]
[Sch99]
[vde] [VDM]
Grit Köppel et al. Sabina Jeschke, Nicole Natho, Olivier Pfeiffer, and Erhard Zorn. New Media in Education and Research - a Sophomore Lecture at TU Berlin. In Proceedings of the 5th IEEGCC Conference and Exibition, March 17-19 2009, Kuwait-City, Kuwait, Al Kuwait, Kuwait, March 2009. IEEE Computer Press. VDI The German Association of Engineers. Ingenieurmonitor August 2009. Technical report. SPIEGEL ONLINE. Wirtschaft bangt um den Technik Nachwuchs. http:// www.spiegel.de/unispiegel/jobundberuf/0,1518,500843,00.html, last retrieved 18.01.2010. C. Schiersmann. Zugangsweisen von Mädchen und Frauen zu den neuen Technologien – eine Bilanz vorliegender Untersuchungsergebnisse. Technical Report Jg. 5, H. 1/2., 1978. H. Schelhowe. Interaktivität der Technologie als Herausforderung an Bildung. Zur Gender-Frage in der Informationsgesellschaft. In Forschungsinstitut Arbeit, Bildung, Partizipation (FIAB): Jahrbuch Arbeit, Bildung, Kultur, volume 17, pages 49 –55. 1999. vde. http://www.vde.com/nr/rdonlyres/fa117595-4ca3-4fcc-a337-292551222 fee/2859/vdemonitor2004.pdf, last accessed 18.01.2010. VDMA. http://www.vdma.org/wps/portal/Home/en, last retreived 18.01.2010.
Bringing Problem Based Learning to Academic Engineering Education using Robotics as the Utility Vehicle Nicole Natho, Sabina Jeschke, Lars Knipping, Olivier Pfeiffer, Ursula Vollmer, Marc Wilke
Abstract The Robinson curriculum uses student defined projects in robotics to achieve two basic goals. First, we are introducing students of engineering and the natural sciences to the field of robotics itself, while teaching the problem-solving skills necessary for their future work and career, Second, we aim to increase technological literacy in students of other fields and high school students (in shorter courses) while attracting more students to modern technology and technologically oriented careers through the interdisciplinary popularity robotics enjoys throughout modern society. Based on the experience gathered in these classic project-oriented courses, we propose to expand the concept, using robotics as the basis for a problembased learning (PBL) course, concentrating even more on teaching scientific and engineering skills rather than robotics itself. The interdisciplinary of robotics makes it an ideally suited candidate for such an approach as it incorporates skills and knowledge from diverse fields of engineering, the natural sciences and beyond. Such a course teaches engineering students the skills required in their future work place based on problems similar to those encountered in their professional careers. Keywords Education · Problem-based Learning · Robotics
Solid State Physics (sensor systems, power supplies) Psychology (interaction and communication with humans, HMIs) Computer Science (software and artificial intelligence) Cybernetics
Similarly, applications for robotics range from industrial engineering to microinvasive surgery and caring for the elderly, providing a high degree of social relevance and impact. Second, and as a result of the above variety, a project in robotics will put a high emphasis on team work and communication skills. Students will need to communicate and coordinate their efforts, especially to bring unique expertise, both previous and gained in the course of the project, to bear towards the defined goal. Thus, robotics projects are well-suited for students to learn and train the soft skill required in their future careers as modern day engineers. For our curriculum we use a special teaching method: problem based learning (PBL). Team work and communication skills are naturally furthered with PBL. Furthermore, the differences in the students’ levels of knowledge help to complement these skills and this way of autonomous teaching additionally motivates students in the course. In addition, robotics, in particular mobile robotics, holds a special fascination among all the technological fields: Starting with the first “automatons” of ancient times (Archytas of Tarent, Heron of Alexandria) and da Vinci’s studies of androids to the modern heroes of movies and literature, robots (and in particular, humanoid robots) have always captivated the fantasy of human beings. Probably better than most technological fields, robotics can be used to bring students into engineering and the natural sciences while providing motivation based on already existing interest. Using this combination of interdisciplinary combined with wide spread popularity helps attract students from diverse disciplines, even non-technological fields. Taking advantage of this effect, we propose problem-based learning for engineering and applied science based on robotics projects. Engineering student will learn new skills and acquire new knowledge based on a problem similar to those they might encounter in their future professional careers.
2 Problem-Based Learning Already in ancient times under Socrates it was well-known that the putative ignoramus finds gradual solutions to complex problems starting from a question. This ability is also of crucial importance in today’s knowledge and information society. This style of teaching is called “Problem-based learning (PBL)”. It was developed in the seventies by Howard Barrows et al. [BT80]. He developed a problemoriented curriculum for medical students on the basis of the ideas of David Boud [BF98] and John Dewey [Dew16]. Today this idea is used in numerous technical disciplines also.
Bringing Problem Based Learning to Academic Engineering Education
283
The intention of problem-based learning is the development of skills for precise acting by using problems that were designed to be as authentically as possible in a motivating workflow. In addition, PBL wants to impart social, technical and methodical skills. A further important characteristic is the basic attitude towards learning: learner and educator are equivalent persons regarding technical knowledge and behavioral role. According to Barrows [Bar96] a PBL curriculum has the following characteristics: • • • • • •
Learning is student-centered. Teachers are facilitators or guides. Learning occurs in small student groups. Problems form the organizing focus and stimulus for learning. Problems are a vehicle for the development of problem-solving skills. New information is acquired through self-directed learning.
A problem-based curriculum is arranged for developing the ability of solving problems. Operationalization of learning targets takes place via key skills, and not as in traditional teaching methods in form of bits and pieces of knowledge. According to Weber [Web04], the following key skills are communicated in a PBL curriculum: • • • • •
professional qualifications (expertise), methods and/or media qualifications (methods for the search of approaches), social skills (social behavior in learning environments and teams), personal qualifications (development of the personality), problem solving qualifications and decision-making and responsibility.
In the terms of Barrows [Bar00] the primary educational goals are developed according to the specific competencies of the particular target audience: 1. Acquisition of structured knowledge regarding the problems. 2. Development of efficient problem solving strategies (reasoning) for real world examples 3. Development of effective skills of self-directed learning and team work. 4. Enhancing students’ motivation. Questions and/or problems are developed by stepwise asking (genetically). Moreover the students should be solving problems as autonomous as possible. The lecturer or instructor is only a workflow moderator and not a knowledge transfer moderator. For complex problems, such a procedure is very difficult to realize. Therefore Meril [Mer02] [Mer07] suggests integrating the elements of “instructional design” (c.f. figure 1). In this way, complex problems are reconstructed by sub problems that are subsequently moderated according to PBL.
284
Nicole Natho et al.
Fig. 1 Instructional Design [vMBH04]
3 Defining Robotics-Projects for PBL Robotics offers numerous problems touching upon knowledge from many different fields of science. Based on our experience in a previously taught, project-based robotics course (see chapter 5), we want to adopt the course to PBL. This would give the students a better understanding of how the theory (currently taught in a separate, classic-style frontal lecture) relates to the praxis they will face in their professional careers. Given the broad multidisciplinarity of robotics, it is possible to motivate and teach knowledge and skills from a variety of fields. Let us consider the following example of a wheeled motive system for a robot. Figure 2 shows a robot based on the LEGO Mindstorms’ set of robots [FFA07] [BDD+ 07]. This robot was designed and built by students during the first “Robinson Ing” course taught at the University of Stuttgart. The students had defined their own project: A robot capable
Fig. 2 LEGO Mindstorms Plotter Robot designed by students
Bringing Problem Based Learning to Academic Engineering Education
285
of plotting a vector-graphic onto a sheet of paper. The motive system consists of two independently controlled wheels, each powered by an electric motor and monitored through a rotation sensor. The engine can be controlled by providing an angle of rotation and a power level for the movement. As the students quickly learn, it is not quite easy to move the robot along a given course. Even moving in a straight line becomes a challenge as the robot does not include any navigational system beyond the rotation sensors. Differences in slip and friction between the wheels have to be considered and addressed before the robot can move in any way approaching the precision required for its intended use as a plotter. What knowledge can we teach using this robot and its motive system and how can we refine the problem from a rather basic starting point to an advanced solution following the approach suggested by Merrill [Mer02] [Mer07]? 1. We start with a simple problem: Build the motive system and let the robot move a given distance in a straight line. The students will be given the characteristics of the electric motor and will be able to determine the movement by integrating the equations of motion given the angular momentum and the mechanical and geometrical property of the transmission system, the wheels and the mass of the robot. After implementing the results the students will soon realize that the robot is not moving in a straight line as effects of friction and slip result in the two wheels moving at different speeds. 2. Help the students consider the effects of friction. They will have to study friction, the difference between static friction, rolling resistance, and sliding resistance and how to determine which is applicable using the rotation sensors of the motors. They will also have to expand the originally one-parametrical equation (one, equal angular momentum applied to both wheels resulting in movement in one direction) to include two parameters (the different angular momentums of each motor) and a course in two dimensions. 3. In trying to solve the problem of the uneven movement they will come across (with a little guidance) cybernetics and the idea of controllers. They might implement a PID controller in software to navigate their robot in a straight line. 4. Expand the problem to include more complex navigation and let them solve the problems arising. 5. Let them consider other movement systems such as tracks or maybe even legs and why they might or might not be an improvement over the original wheeled system (Tracks always have slip in a corner, making it harder to follow a precise course). 6. Now expand the problem to include the movement of the pen across the paper dependent on the movement of the wheels. This will lead to studying coordinate transforms, especially if the pen is not mounted in the center of mass or if the (hypothetical) axle connecting the wheels does not intersect the center of mass. As can be seen from this rather basic example we are able to include numerous concepts from mechanical physics, geometry, calculus (to solve the equations of motion), and cybernetics. Depending on their previous experience, software engi-
286
Nicole Natho et al.
neering might be included as well. At this point, we have only been concerned with the movement of the robot and two sensors (the rotation sensors integrated into the electric motors). Further expansion of the project could include: • Mounting additional sensors to improve navigation (magnetic sensors acting as a compass, acceleration or gyroscopic sensors to map the actual movement) or expand on the functionality of the robot (different color sensors to add scanning). Students would study how these sensors work and how they can be used to improve their robot, touching solid state physics and electrical engineering as well as other areas of physics to interpret the output of the sensors. • Different data formats for the images to be plotted, leading into any number of subfields of image processing and geometry. Including popular image formats like JPEG would lead to image compression and related mathematics (Fourier analysis, statistics). • Any number of algorithmic improvements to take better advantage of the limited computational power and memory of the robot’s controlling unit. • Using an external computer to solve more complex computational problems. Students would have to study the bus system provided by the robot’s controlling unit (I2C, USB, and Bluetooth) and different data transfer protocols, possibly including data compression algorithms (s. a.). As the students’ skills and knowledge improve and their motivation is kept up by their interest in improving their robot further and further (experience in teaching robotics show that the problems arising from these improvement do not lessen either), the guidance provided by the instructor will (have to) slowly fade out. In addition to these “hardcore” technical skills, students will be instructed in related skills such as technical writing (preliminary specification of the project, documentation, presentation of the final robot to their peers), project management (distributing the work, quality assurance, learning what is doable and when to cut a feature etc.) and team work related skills (especially communication skills in relating new knowledge, discussing and criticizing ideas). For more on these topics see also chapter “Previous Experience and State of Preparation”.
4 Related Works Problem-based learning was originally developed and applied in medical training for doctors during the late 1960s [SNB90], particularly at McMaster University in Hamilton, Canada. It has since been expanded to other fields. The advantages and challenges of PBL have been widely studied [SvB08] and criticized [KSC06]. The general outcome of these studies shows that PBL has advantages over the classic forms of instruction concerning long-term retention, free recall and application of skills [AM93] [Dea03]. In contrast, PBL proved less effective in immediate postcourse examinations and multiple-choice tests [Dea03]. PBL also tends to overwhelm students at the early stages of study, suggesting that worked examples might be an improved starting point [KSC06].
Bringing Problem Based Learning to Academic Engineering Education
287
The advantages of using robotics as a door opener to attract more students to engineering and other technologically oriented fields was recognized by the “ROBERTA” project of the IAIS Fraunhofer Institute [AM93]. “ROBERTA” uses LEGO Mindstorms’ robots in short courses for high school level female students. The Institute for Personal Robots in Education (IPRE) launched a project similar to Robinson [Bla06] at the beginning of 2007. The project focuses on the use of robots in computer science education. The IPRE, a joint effort between Georgia Tech and Bryn Mawr, developed a special robot for this project. This robot should be cheap enough for every student to buy and use it for learning programming throughout the whole course of studies.
5 Previous Experience and State of Preparation The Institute of IT-Services (IITS) at the – mostly technologically oriented – University of Stuttgart has just finished the first course of a new, project-oriented program for robotics. This course, referred to as “Robinson Ing” was aimed at teaching robotics to 5th semester students of engineering. Currently, the course is projectbased, not PBL as such yet. Engineering students, especially at the beginning of their studies, often lack insight into the working life of an engineer. The students tend to have problems in realizing the importance of the basic courses, which they have to attend, for their chosen subject of studies. The multidisciplinary field of robotics can help in linking basic knowledge with its importance, as well as emphasizing the relevance of looking beyond one’s own nose. Furthermore, the “Robinson Ing” module aims at forming social skills and conveying methodologies for solving complex multidisciplinary challenges. Basic concepts of soft and hardware engineering are taught already at a very early stage of the studies, as well. The course consists of a series of introductory lectures, accompanying the practical part, followed by a presentation task. The lecture topics of this course provide a general overview on robotics, without any special focus. Additionally, the projects are not limited to a specific topic, so that the students have even more freedom in the choice of their projects. The course “Robinson Ing” is primarily aimed at bachelor students of electrical engineering and the computer sciences and is part of their core elective courses. “Robinson Ing” is designed and lectured as an integrated “interdisciplinary robotics laboratory: soft and hardware engineering” for engineering students. It is held each semester during regular term or as a compact course in semester break and consists of three basic components: • a series of introductory lectures, giving an overview to selected principal topics with reference to robotics (2 hours per week, 11 weeks), • a practical training in small groups of 3–4 students designing, building and programming whole robots or robot components (4 hours per week, 11 weeks), and
288
Nicole Natho et al.
• a seminar and presentation part, giving the teams the possibility to demonstrate their project results on a project web page and to introduce it to their lecturer and fellow students during an oral presentation in class (6 hours per week, 2 weeks). After some introductory lectures, students are supposed to form teams and choose their project topic. Before implementation starts, the topics are to be presented by the students. During the implementation phase, lectures focus on different aspects of robotics. Thus, the course includes all components of an “integrated course”, which makes it a complex, but also highly challenging course, of which the students profit in many ways. All components, i. e. the lectures given by the instructor, the hands-on training, and the students’ presentations contribute to the final grades. On the one hand, students acquire important theoretical foundations and knowledge. On the other hand, they have the possibility to practice important soft skills, which they will need later on in their professional life. First practical experiences confirmed the huge potential of robotics to motivate students. “Roberta” courses for girls, held on different occasions, have been met with considerable interest. The participants, even the ones more doubtful at the beginning, were thrilled by the courses. Even before the first “Robinson Ing” course started, the students’ interests could be seen in the number of applications for student instructor positions. The students’ interest in the “Robinson Ing” course taking place at the University of Stuttgart in the summer term 2008 was overwhelming. At the first meeting, far more students than expected wanted to sign up. These students came from different backgrounds, such as computer science, software engineering, technical cybernetics, electrical engineering, mechanical engineering, physics, math, computational linguistics, and technical pedagogics. They were willing to participate in the course, even if they could not get credit points for their studies. The evaluation of an interrogation of the students at the end of the term provided good results (cf. fig. 3). Most of the students stated that the course was clear and well
Fig. 3 Evaluation of the first “Robinson Ing” course
Bringing Problem Based Learning to Academic Engineering Education
289
structured. Questions, covering this aspect included the formulation of goals and requirements, the structure of the course, the relevance of the topics and references to other areas, explanations, media, and methods. In a PBL course, the relevance of the learning topics would be even more understandable for the students. Each learning unit would be embedded in the project at the best fitting point in time. The students attested a good preparation of the lecturers and a good organization of the whole course. Experiences from this course and a restriction in the number of participants will improve the organizational aspects even more. Students felt free to ask questions and place comments. The students declared that their interest in the topic has been brought forward by the course. They were satisfied with the number of topics, that were dealt, but some would have preferred a bit more detail and challenge. PBL would be an ideal method to meet this requirement. In a PBL project, necessary information would be given to the students at the moment most needed and at the exact level of detail needed. The announcement of a follow-up robotics course was taken with a lively interest by the students participating in “Robinson Ing”.
6 Conclusion and Outlook Based on our experiences with project-based robotics courses we propose teaching engineering knowledge and skills in a PBL approach based on robotics. On the one hand, the switch to PBL would conserve the strengths and advantages of our current approach, particularly taking advantage of the popularity of robotics and the handson approach of a project to increase motivation, while training and teaching soft skills throughout the course of the project. On the other hand, PBL would address some of the criticism given by students, the lack of direct application of the lectures to the projects as well as the varying level of detail and challenge. Similarly, we will be forced to limit the degree of freedom the students enjoy in choosing and defining their own project. All students would have to work on the same project or, at best, one project selected from a small range of predesigned projects. Overall, we expect the benefits of changing our paradigm to PBL the disadvantages.
References [AM93]
[Bar96]
[Bar00]
M A Albanese and S Mitchell. Problem-based learning: a review of literature on its outcomes and implementation issues. Academic Medicine: Journal of the Association of American Medical Colleges, 68(1):52–81, January 1993. PMID: 8447896. H.S. Barrows. Problem-based learning in medicine and beyond: a brief overview. In Luann Wilkerson and Wim H. Gijselaers, editors, Bringing Problem-Based Learning to Higher Education: Theory and Practice: New Directions for Teaching and Learning. Jossey-Bass, San Francisco, 1996. H.S. Barrows. Problem-based learning applied to medical education. University School of Medicine, Springfield, Southern Illinois, 2000. Rev. 1994 Ed.
290
Nicole Natho et al.
[BDD+ 07] Martijn Boogaarts, Jonathan A. Daudelin, Brian L. Davis, Jim Kelly, Lou Morris, Fay, Rick Rhodes, Matthias Paul Scholz, Christopher R. Smith, and Rob Torok. The LEGO MINDSTORMS NXT Idea Book: Design, Invent, and Build. No Starch Press, 2007. [BF98] David Boud and Grahame Feletti. The Challenge of Problem Based Learning. Routledge, 2 edition, 1998. [Bla06] Douglas Blank. Robots make computer science personal. Communications of the ACM, 49(12):25–27, 2006. [BT80] Howard S. Barrows and Robyn M. Tamblyn. Problem-Based Learning: An Approach to Medical Education. Springer Publishing Company, 1980. [BT95] R.D. Barr and J. Tagg. From teaching to learning - a new paradigm for undergraduate education. Change, pages 13–25, December 1995. [Dea03] F. Dochy and et al. Effects of Problem Based Learning. Learning and Instruction, 13:533–568, 2003. [Dew16] John Dewey. Democracy And Education. Free Press, Original from The Macmillan Company, 1916. [fAiSA06] St. Augustin Fraunhofer-Institut für Autonome intelligente Systeme AIS. Roberta Grundlagen und Experimente, volume 1. IRB Verlag, 2006. [FFA07] Mario Ferrari, Guilio Ferrari, and David Astolfo. Building Robots with Lego Mindstorms Nxt. Syngress Media, April 2007. [KSC06] P.A. Kirschner, J. Sweller, and R.E. Clark. Why minimal guidance during instruction does not work: An analysis of the failure of constructivist, discovery, problem-based, experiential, and inquiry-based. Teaching. Educational Psychologist, 41(2):75–86, 2006. [Mer02] M. David Merrill. A pebble-in-the-pond model for instructional design. Performance Improvement, 41(7):41–46, 2002. [Mer07] M. David Merrill. A task centered instructional strategy. Journal of Research on Technology in Education, 40(1):33–50, 2007. [SNB90] H. G. Schmidt, G. R. Norman, and H. P. Boshuizen. A cognitive perspective on medical expertise: Theory and implications. Academic Medicine, (65):611–621, 1990. [SvB08] J. Strobel and A. van Barneveld. For What Learning Outcomes is PBL Effective. In Proceedings of the Research Symposium of PBL in Engineering Education, Aalborg, Denmark, 2008. [vMBH04] J.J.G. van Merriënboer, Th. Bastiaens, and B. Hoogveld. Instructional design for integrated e-learning. In Wim Jochems, Rob Koper, and Jeroen Van Merrienboer, editors, Integrated E-Learning: Implications for Pedagogy, Technology and Organization, page 15. Kogan Page, London, UK, 2004. [Web04] Agnes Weber. Problem-Based Learning: Ein Handbuch für die Ausbildung auf der Sekundarstufe II und auf der Tertiärstufe. hep verlag, 2004.
New Media in Education and Research – a Sophomore Lecture at TU Berlin Nicole Natho, Sabina Jeschke, Erhard Zorn
Abstract Information flood is an essential facet of the digital age. Consequently, organizing information of an explicit subject efficiently, without losing the overview is difficult. In education and especially in academic education, information management is a sensitive issue for freshmen and sophomores. Many students are overburdened with the new situation at university having to organize and manage all the new knowledge they obtain in lectures. The use of new media by lecturers and students provides an opportunity to overcome these challenges. Moreover, communication, cooperation and interaction with each other are examples of the social skills, being nowadays’ important factors in education and professional life. We present the application OneNote by the use of Tablet PCs in a sophomore lecture at the Berlin Institute of Technology, aimed at teaching students how to cooperate in their project-work using this collaborative platform. The implementation is described and a first evaluation is presented. Finally, a highly desirable extension for integrating mathematical notation is outlined. Keywords Tablet PCs · Collaborative Work · Cooperative Work · Academic Education · Information Management
Unfortunately, many of appropriate software applications need a lot of training time. In addition, lecturers cannot spend endless time to prepare a perfectly computerized course induced by the complexity of these software applications. What can we do for both parties? Focusing on the social skills like communication, coordination and cooperation as nowadays’ important factors in education, collaborative software applications possess a high potential to support the learning, teaching and research processes at university by the means of the new media and new technologies [CJL07]. As a part of the Galilea project1 , introduced at the Technische Universität Berlin, the new gender sensitive Bachelor of Science program “Natural Sciences in the Information Society” [JNP+ 07], [JEN+ 07], [DEJ+ 08] provides two new innovative lectures, which started in winter term 07/08 and summer term 08: 1. Scientific Information Management (freshmen) and 2. New Media in Education Research (sophomores) This innovative program is designed to be very interdisciplinary while offering a broad spectrum of lectures in natural sciences, mathematics and computer science to a manageable student group of about 30 persons. The lectures “Scientific Information Management” and “New Media in Education and Research” were tailored to complete this program, are part of the compulsory lectures, and are perfectly suited to introduce concepts of Information Management using Tablet PCs in education. To overcome the above described challenges, we use the collaborative platform Microsoft OneNote 2007 [mic], [HKP08] (c. f. figure 1 for a screenshot) in the lecture “New Media in Education and Research” as a software application especially for Blended Learning Scenarios. One Note is a digital notepad with the flexibility of a classical handwritten notepad. Because of its affinity to other well-known of-
Fig. 1 Microsoft OneNote 2007 screenshot 1
http://www.galilea.tu-berlin.de
New Media in Education and Research
293
fice products, it is easy to use after a short training period. The use of OneNote in this context is based on the fundamental assumption that nearly everybody knows how to use office applications. Respectively, this collaborative platform offers the possibility to integrate with several office products. All notes are freely placeable and can be archived by different methods such as tagging or sorting by usefulness. Starting other applications and recording talks or small videos (podcasts) within this digital notepad is also possible. Internet resources can also be attached and tagged to notes. Besides, OneNote supports mobile devices as Tablet PCs, PDAs, and graphics tablets for handwritten notes. Briefly, the benefits of this application are its clarity, usability, and the opportunity to convey soft skills. In the following chapters, we present the preliminary results of our first evaluation. In line with the evaluation, the second chapter describes the theoretical and practical settings of this project in detail, and the comments of the students. Moving on, the third chapter reviews the drawbacks of the use of OneNote against the background of natural sciences and mathematics.
2 Concept The lecture “New Media in Education and Research” is organized as an integrated lecture; i. e. it consists of frontal lectures by the teacher and project work, accomplished by the students in groups. Several projects are currently set up and carried out by the students using Tablet PCs and OneNote.
2.1 Blended Learning Scenario in Higher Education Blended learning is a model of education, combining traditional didactical methodologies with new media technologies for the presentation and distribution of knowledge. Therefore, this method unites the flexibility and efficiency of the new media with social components such as face-to-face communication, which is an important factor in modern higher education [Rei03], [SPW03], [Ker02], [KdWS02]. Done right, blended learning ensures the quality of the academic curriculum [Rei], [HTY06], [Sim], [HMB09]. Moreover, from a psychological point of view blended education brings together different kind of didactical theories such as constructivism [Rei03], [Fos96], behaviorism, and cognitivism , to support different types of learners in their individual learning process [MB05]. However, the main educational presentation for imparting basic knowledge about a field of study is still a lecture. In addition, faculty can concentrate on special topics in tutorials and seminars. However, in all models of educational styles, it is obviously complicated to integrate new media concepts into the curriculum, and additionally difficult to handle. In Germany, the deployment of new media is being promoted since 2000 to augment the professional use of new media for teaching, learning and examinations, yet the outcome of the actual use is sparse [Wer06], [SBHB01]. Of course, many lecturers have attempted to integrate new media in a
294
Nicole Natho et al.
Fig. 2 Students a “New Media in Education and Research” lecture
blended learning approach, but they conclude that the relevant drawbacks are: the support to choose of all appropriate software applications and hardware products, the missing didactical concepts, and time-consuming integration into the curriculum. In this way, the advantages for lecturer to improve educational content are not apparently comprehensible. As a consequence the lecturers’ motivation to use new media is decreasing especially during the initial implementation and in maintenance [EGS02], [EHKS06], [KN07], [HK06], [Mos07]. In conclusion, everybody is aware that the new media requires substantial financial investments and personnel expenditure [EGS02]. Additionally, for a successful introduction of the new media, four different types of lecturers, regarding their character attributes such as motivation, unstableness, readiness to assume risk for innovations respectively their willingness and ability, have to be considered. These four types: entrepreneurs, risk aversive, careerists, and reluctant, defined by Hagner [Ows06], [Hag], [KN07] have to be motivated each in a different way. Entrepreneurs are easy to manage because they are motivated primarily intrinsically using all offers on their own initiative. All other types need external motivation and they must be supported in technical and organizational matters. As a result, a sustainable implementation of new media within the higher education depends on miscellaneous factors, and the complexity of most of the educational innovations obfuscates the positive properties and advantages [EHKS06], [Ows06]. Although individual support for Blended Learning means an additional workload, it appears as the most effective way to encourage lecturers and to implement E-education applications [EHKS06].
New Media in Education and Research
295
2.2 Blended Learning Scenario in Higher Education Which methodologies can be applied in higher education to develop practical expertise by the lectures for Blended Learning? According to [Ows06] the following fundamental requirements have to be provided and adapted to the target audience defined by Hagner [Hag]: development of quality especially for educational scenarios, advisory support, and augmentation of the lecturers’ readiness. One of the most significant issues is the lecturer’s ability to conceive the didactical additional benefit being in opposition with the disadvantages and challenges such as time pressure, complexity of the implementation and missing support. To intercept these primary problems it is advisable to combine one specific didactical methodology with specific communication or information technology to systematize benefits between traditional didactical methodologies and new media concepts. In this investigation, we focus on one specific information technology (office software products) combined with traditional didactical methodologies such as explorative or cooperative learning in classroom scenarios. The deployment of OneNote and akin applications [Wea06] can help resolve some challenges in altering the daily routine within Blended Learning scenarios: the lecturer is able to prepare the lecture directly on the computer similar to traditional preliminary techniques with an integration and execution of different applications and Internet resources. Additionally, media types like pictures, diagrams, notes, audio recordings, etc can be easily collected centrally by OneNote, whilst other content from web pages can be integrated by means of drag and drop. This information can be simply interchanged and shared between the individuals involved and between different teams. Particularly, OneNote supports the integration of inked sketches and comments, which are of great importance for the individual learning process especially in scientific education, and their usage is facilitated by Tablet PCs. In this way, OneNote acts like a control center for all relevant applications. During the lectures, students can be interactively integrated into the workflow in different ways such as explorative learning. The students and lecturers can collect and organize all these information types according to their personal preferences and requirements. However, the technical requirements for such lectures are a drawback.
3 OneNote in HigherEducation 3.1 Exemplary Deployment An exemplary scenario is the use of OneNote as virtual whiteboard and collaborative environment during a course: the lecturer can centrally store every information in the application, presented to the audience by a projector and easily transfer them to the students’ applications or vice versa. In this way, students and lecturer can share and transfer information and data without any difficulties. Within the context of the lecture “New Media in Education and Research” we have deployed OneNote in several educational units of this course. One educational unit is specified exemplary:
296
Nicole Natho et al.
16 students attended the program “Natural Sciences in the Information Society”; about half of them were female students. During the lecture, the students worked in groups of two on Tablet PCs (HP Compaq tc4200, provided as part of a HP Technology for Teaching Grant). The subject of the teaching unit was “Academic Writing with Microsoft Word and Co” with the ambitious aim to work out the subject together with the students. The lecturer prepared the lecture with OneNote by creating a new digital notepad for managing all content i.e. examples, brainstorming notes, presentations, exercises, etc. with the Tablet PC facilitating the use of handwritten notes. The resulting exemplary 90 minutes lecture had following modules: 1. Introduction to “Academic writing in MS Word and Co” (duration: 20 min): brief introduction to common problems of word processing applications in academic documents such as large documents, implementation of figures, images and tables, and the infamous footnotes. 2. Teamwork session “Explorative Examples” (duration: 20 min): students work on some sophisticated examples with the aid of a digital worksheet. 3. Discussion of the results (duration 10 min): exchange of information 4. Teamwork session: “Students’ Examples” (duration 30 min): working student examples on selected problems 5. Discussion of results (duration 10 min) The first component of the educational unit is an introduction in form of a presentation about general problems of word processing programs with numerous examples. In most cases, there are several possible techniques to generate presentations and integrate them in OneNote; a PowerPoint presentation being on possibility, while another possibility is the use of OneNote as a digital board for making handwritten notes similar to a traditional lecture. Sharing this digital notepad with other participants, the students also have the possibility to add notes in this notepad. In addition, various examples can be directly executed in OneNote. For the second and fourth component of the lecture, the lecturer prepared possible exercises (digital worksheets) and attached programs, useful to study. The students also take their own notes in these segments. As the result for components three and five, the students shared their results with the lecturer and other students for discussion purposes. Moreover, the students were able to attach additional examples, or recorded comments. Accordingly, neither the teacher nor the students had “rough papers”, and everything was arranged within the digital notepad (digitalized lecture). Finally, all integrated examples could be transferred to the students’ notebooks.
3.2 Extending the Concept to Mathematics and Natural Sciences When trying to apply the aforementioned techniques to a mathematics lecture we ask too much from the current release of OneNote, as complex formula, sets of symbols, and many technical illustrations still need to be integrated into the
New Media in Education and Research
297
application to make it valuable for mathematic or natural scientific lectures. Thus, the next version of OneNote needs to be extended to handle these particularities. Although OneNote already offers the possibility of exporting handwritten notes, making them available to other office-applications our goal is supporting the recognition of handwritten formula, which shall then be evaluated by the prevalent mathematical tools: Maple, Matlab, and Mathematica. Additionally it is desirable to integrate a conversion tool to export technical outlines to a vector graphic application.
4 Conclusion and Outlook The combination of Tablet PCs and OneNote is used to train students in cooperating in a collaborative information platform. The future integration of handwritten scientific content in OneNote is the next desirable future feature in this context. Our preliminary results show that students perceive a lecture with OneNote as very fascinating, because of the variety of educational methodologies and differences to traditional lectures. Didactical concepts appear to be more similar to school education than to higher education at university. Working with digital examples implies spending more time to impart knowledge. A traditional lecture, as a noninvasive form of the Nuremberg Funnel, is designed to provide students with as much knowledge as possible in the available time. Consequently, the developed concepts are appropriate for selected lectures or seminars and tutorials. An additional challenge emerges from the technical overhead and costs to use notebooks or Tablet PCs, and applications such as OneNote. Yet, Blended Learning scenarios accelerate the progress of educational success in excess of traditional lectures due to the fact of large-scale integration of examples and interactivity. In conclusion, it remains to be examined whether Blended Learning scenarios with OneNote are acceptable and helpful for the lecturers with only sparse technical affinities in the sense of Hagner’s [Hag] four types. It is worth recalling that the benefits of blended learning lectures with OneNote are obvious: lecturers prepare lectures or conduct handwritten lectures on the computer similar to traditional lectures. In summary, after a short training period it can be deployed straightforward because of its affinity to other well-known office applications, based on the fundamental idea that lecturers frequently use these applications. As a collaborative tool, it allows for the interactive integration of students into lectures, and it is an attractive alternative to a traditional lecture. At that time, experienced difficulties arise in lectures of mathematics, engineering, and natural sciences caused by the complex notation and technical outlines. Therefore, it is desirable to integrate special add-ons for scientific notation and conversion tools for technical outlines into a vector graphic application. Acknowledgements The authors from Berlin thank HP for supporting their work on pen-based technology in education with a HP Technology for Teaching Grant.
298
Nicole Natho et al.
References [BH01]
[CJL07] [DEJ+ 08]
[EGS02]
[EHKS06]
[Fos96]
[Hag]
[Hey05]
[HK06]
[HKP08] [HMB09] [HTY06] [JEN+ 07]
[JNP+ 07]
[KdWS02]
[Ker02]
Carole Barone and Paul R. Hagner. Technology Enhanced Teaching and Learning: Leading and Supporting the Transformation on Your Campus: 5. Jossey Bass, August 2001. Sabine Cikic, Sabina Jeschke, and Nadine Sinha Ludwig. Modularization and Integration of Virtual Experiments. In Interactive Computer Aided blended Learning ICBL, Florianópolis, Brazil, March 2007. N. Dahlmann, M. Elsner, S. Jeschke, N. Natho, and Ch. Schröder. Gender Gap in Technological Disciplines: Societal causes and consequences, to appear in 2008. In Proceedings of the Symposium on Technology and Society (ISTAS 2008), Los Alamitos, CA, 2008. IEEE Computer Society. J.L. Encarnacao, H. Guddat, and M. Schnaider. Die Hochschule auf dem Weg in e-Learning Zeitalter. In U. Bentlage, P. Glotz, I. Hamm, and J. Hummel, editors, E-Learning: Märkte, Geschäftsmodelle, Perspektiven, pages 21–55. BertelsmannStiftung, Gütersloh, 2002. D. Euler, J. Hasanbegovic, M. Kerres, and S. Seufert. Handbuch der Kompetenzentwicklung für e-Learning Innovationen: Eine Handlungsorientierung für innovative Bildungsarbeit in der Hochschule. Huber Verlag, Bern, 2006. C. Twomey Fosnot. Constructivism: A Psychological Theory of Learning. In C. Fosnot, editor, Constructivism: Theory, Perspectives and Practice, pages 8–33. Teachers College Press, New York, 1996. Columbia University. Paul R. Hagner. Faculty Engagement and Support in the New Learning Environment. In Paul R. Hagner, A. Barone, A. Carole, and J. Bass, editors, Educause Leadership Strategies, Volume 5, Technology-Enhanced Teaching and Learning: Leading and Supporting the Transformation on Your Campus, volume 5, pages 1–12. Francis Heylighen. Information Overload, Complexity and Information Overload in Society: Why increasing efficiency leads to decreasing control, (submitted to The Information Society). http://pespmc1.vub.ac.be/papers/info-overload.pdf, 2005. J. Hasanbegovic and M. Kerres. Entwicklung von Massnahmenportfolios zur Vermittlung von eLehrkompetenz. In E. Seiler Schiedt, S. Kälin, and C. Sengstag, editors, E-Learning - Alltagstaugliche Innovation? Waxmann Verlag, Münster, 2006. Dagmar Herzog, Nina Koch, and Jan Peters. Selbstorganisation mit Microsoft Office OneNote 2007. Microsoft Press Deutschland, 2008. N. Hoic-Bozic, V. Mornar, and I. Boticki. A Blended Learning Approach to Course Design and Implementation. IEEE Transactions on Education, 52(1):19–30, 2009. T. Hayashi, H. Tominaga, and T. Yamasaki. Blended learning contents for university education. In Information Technology Based Higher Education and Training, 2006. ITHET ’06. 7th International Conference on, pages 499–502, 2006. S. Jeschke, M. Elsner, N. Natho, O. Pfeiffer, and Ch. Schröder. Attractive Universities: New Curricula in Natural Sciences and Engineering. In Proceedings of the Meeting the Growing Demand For Engineers and Their Educators 2010 - 2020 Int. Summit, Los Alamitos, CA, 2007. IEEE Computer Society. S. Jeschke, N. Natho, O. Pfeiffer, Ch. Schröder, and L. Wilke. Challenge Diversity: New Curricula in Natural Sciences, Computer Science and Engineering. Proc. of the Frontiers in Education 2007. IEEE Computer Society, 2007. Los Alamitos, CA. M. Kerres, C. de Witt, and J. Stratmann. E-Learning: Didaktische Konzepte für erfolgreiches Lernen. In K. Schwuchow and J. Guttmann, editors, Jahrbuch Personalentwicklung u. Weiterbildung. Neuwied, 2002. M. Kerres. Online- und Präsenzelemente in hybriden Lernarrangements kombinieren. In A. Hohenstein and K. Wilbers, editors, Handbuch E-Learning. Fachverlag Deutscher Wirtschaftsdienst, Köln, 2002.
New Media in Education and Research [KN07]
[MB05]
[mic] [Mos07]
[oCRL] [Ows06] [Rei]
[Rei03] [SBHB01] [Sim] [SPW03]
[Str08] [Wea06]
[Wer06]
299
H. Kolbe and A. Nikolopoulos. Sustainable Implementation of e-Learning Innovations into Large German Universities. In Proceedings of the 6th European Conf. on e-Learning (ECEL 2007), Copenhagen, DK, 2007. L. McNutt and M. Brennan. Work in Progress - Learning Styles and elearning, what is the Connection?, Frontiers in Education, 2005. FIE ’05. In Frontiers in Education, 2005. FIE ’05. Proc. 35th Annual Conf., 19-22 Oct. 2005, pages F1H–27 – F1H–31, October 2005. Microsoft OneNote Homepage. http://office.microsoft.com/de-de/onenote/default. aspx. Franziska Zellweger Moser. The Strategic Management of E-Learning Support: Findings from American Research Universities. Waxmann, 1., aufl. edition, January 2007. Association of College & Research Libraries. Information Literacy Competency Standards for Higher Education. www.ala.org/ala/acrl/acrlstandards/standards.pdf. R. Owston. Evaluation of a blended learning professional development program for middle-school mathematics and science teachers. AERA, San Francisco, 2006. T. Reichlmayr. Enhancing the Student Project Team Experience with Blended Learning Techniques. In Proceedings Frontiers in Education 35th Annual Conference, pages T4F–6–T4F–11, Indianapolis, IN, USA. Gabi Reinmann-Rothmeier. Didaktische Innovation durch Blended Learning: Leitlinien anhand eines Beispiels aus der Hochschule. Huber Verlag, Bern, 2003. Sabine Seufert, Andrea Back, Martin Haeusler, and Sonja Berger. E-Learning Weiterbildung im Internet. SmartBooks Publishing AG, 2001. B. Simington. Blended Learning - Neither Shaken nor Stirred. In The 17th Annual SEMI/IEEE ASMC 2006 Conference, pages 358–361, Boston, MA. K. Schweizer, M. Paechter, and B. Weidenmann. Blended learning as a strategy to improve collaborative task performance. Journal of Educational Media, Special Issue on Blended Learning, 28(2-3):211–224, 2003. Special Issue on Blended Learning. Randall Stross. Struggling to Evade the E-Mail Tsunami. New York Times, April 2008. A.C. Weaver. Does Classroom Presenter Software Change Learning Outcomes? In Proceedings of the 36th ASEE/IEEE Frontiers in Education Conference, pages 10– 14, San Diego, CA, USA, October 2006. B. Werner. Status E-Learning an deutschen Hochschulen. Technical report, Inst. f. Wissensmedien, 2006.
Supporting Collaboration in Professional Soft-Skill Training Courses Olivier Pfeiffer, Sabina Jeschke, Lars Knipping, Nicole Natho
Abstract More and more employers qualify their employees in soft skills or make soft-skill knowledge a decision criterion when assigning new jobs. Soft-skills courses are typically highly interactive as soft skills are no factual knowledge and cannot be acquired by simple drilling. The course instructors, who very often are external experts, have to face the challenge to adapt to new media and course styles since new technologies not only shape everyday working life of the course attendees, but also demand for new soft skills due to changed communication practices. This paper proposes a community-oriented approach for professional soft-skill courses using a room-based collaboration platform. Keywords University Education · Collaborative Learning · Collaborative and Communication Software · Educational Software · Curriculum Design and Development
1 Introduction When it comes to job applications and vacancies respective journals report that times are changing. Personnel managers no longer only prefer the applicants that are the best in their area of expertise. The appeal of future colleagues should also lie in socalled “soft skills”. That is why such professional trainings are continuously rising, that are focused on further education in these soft skills. On the other hand, new media based courses face the bias of being anonymous, non-intuitive and purely theoretical and therefore inappropriate to impart soft skills. This article describes the special characteristics of soft-skill courses performed as blended learning courses and how these characteristics can be supported by using a room-based cooperative knowledge space.
O. Pfeiffer (B) MuLF, TU Berlin, Straße des 17. Juni 136, 10623 Berlin, Germany e-mail: [email protected]
2 Types of Soft-Skill Trainings As the awareness of the importance of soft skills has risen in the last 15-20 years [KBGM90], many companies, but also university facilities have started providing professional training to their employees (cf. [Aro04], [Hed07] for two of many examples). In some soft-skill courses in a professional context it is preferred to train the participants within their familiar surroundings. But in most cases the courses are guided by external coaches and the participants come together from different companies, institutes and backgrounds that enforce the use of an abstract common environment. This article and the concepts resulting from it are aimed at the second type of soft-skill courses, namely those with new media support. These trainings often consist of multiple modules with focus on thematic constellations, such as: • • • • • •
self and time management work-life balance communication and leadership skills conflict management cross-cultural communication professional networking
3 Special Characteristics in Soft-Skill Courses The soft-skill courses in question are attended by persons with diverse personality structures who all are detached from their familiar professional environment and who have come together to update their soft skills and improve their own work and the performance of their teams. Some of the participants only need some theoretical instruments to improve their soft skills and achieve their ambitious aims. But others actually have conflicts with colleagues, supervisors or themselves and are very shy and need support and advice to solve their problems. A main goal of soft-skill courses is to unite these personalities and strengthen them all in the different needed achievement. Soft-skill courses are usually held in groups of 10 to 15 people and most of the time the group members work together and discuss the topics. A typical scenario for such a course is to have each participant introduce their respective neighbour. First they introduce themselves to each other in groups of two, and then they are introduced to the rest of the whole group by one another. Typically such classes are taught in one or two day courses. Theoretical lessons given by the course lecturer alternate with group work in which the participants mostly deal with “real-life” problems to apply and internalize the theoretical concepts. Sometimes these group constellations remain the same throughout the whole course. In other cases the lecturer encourages the participants to change partners in the group in order to get to know each other, because most of the times the participants are not acquainted to each other. Such soft-skill courses demand a certain level of trust among the participants, because most participants attend these courses to
Supporting Collaboration in Professional Soft-Skill Training Courses
303
solve certain currently existent problems and conflicts with themselves or colleagues and they need to “open up” to the group of participants to tell their story. A further characteristic of soft-skill courses is the complexity and poor structure of the provided content. A conflict management course will also cover topics from self management and presentation or communication skills. That means a variety of information will be provided to course participants because the different topics of soft skills are not clearly definable. Figure 1 shows some content from a typical soft skill course performed as a blended learning course, i. e. partly as presence course and partly as an online course provided using the Moodle platform [Dou]. In addition to information documents and references to external work regarding the topic, the course consists of a variety of tasks (with deadlines), different forums about the different topics and tasks, chat appointments and other events in the calendar. Most of the tasks are “virtual group work” tasks. This implies that almost every task has its own forum and information sites. Obviously this is a lot of poorly structured information where cooperative work aspects and content elements are mixed up in an extensively incomprehensible way.
Fig. 1 Screenshot of a soft-skill online course in a content-oriented platform
4 Collaboration in a Room-Based Environment In the previous section we have outlined some major downsides of a purely contentcentred view of soft-skill learning. Content-orientation may be appropriate for factual learning where the focus actually is on the learning content. Soft skills, however,
304
Olivier Pfeiffer et al.
are not only about “knowing that”, but much more about “knowing how”. More precisely, soft skills are about interacting with other people which obviously is a practical task. Moreover, most of the skills listed above are themselves related to collaboration. Communication, conflict management and coordination skills are fundamental prerequisites to collaborative work. Following [RT95], we define “collaboration” as “a coordinated, synchronous activity that is the result of a continued attempt to construct and maintain a shared conception of a problem.” In a collaborative environment, these soft skills can be trained and practiced, making the ability to collaborate a learning objective in its own right. Thus, it is desirable to rethink web-based soft-skill courses from a collaboration-centred point of view, and, when offering the course on the web, to deploy them in a community-oriented, rather than a content-oriented environment.
4.1 Content Orientation vs. Community Orientation Most eLearning environments currently deployed, however, follow the contentoriented approach. Their architecture focuses on content management, content authoring, and content provision, whereas communication and collaboration facilities are considered add-ons, if provided at all. In contrast, the community-oriented approach focuses on the collaborative processes of communication and knowledge construction. Content objects are means to foster these processes — e. g. a theoretical introduction or a group work assignment — and/or results (one might even say “by-products”) of these processes, codifying knowledge constructed cooperatively. Figure 2 illustrates the two approaches: in a content-oriented approach, communication and collaboration tools are grouped around content elements (top), whereas in a community-oriented approach, they are embedded in a collaboration infrastructure (bottom). A related discussion on content-orientation vs. collaboration can be found in the first section of [SKS06]. It should be noted that purely community-oriented plat-
Fig. 2 Content-oriented (top) vs. community-oriented (bottom) approach. The purple boxes symbolize content objects
Supporting Collaboration in Professional Soft-Skill Training Courses
305
forms in reality are very rare, and that developers of content-oriented platforms strive to complement them with collaboration tools. These tools, however, are often isolated, cf. the example above.
4.2 Structuring Content and Collaboration in Virtual Rooms Quite a few CSCW and CSCL systems follow a room-based approach which is often referred to as the “room metaphor”. In [GR03], the authors give an exhaustive list of features which both real and virtual rooms are comprised of. A major advantage of environments based on the notion of virtual rooms is their support not only for “real” collaboration which by definition is synchronous, but also for cooperative (potentially asynchronous, with more individual actions) and mixed scenarios of group learning and group work [Wes05]. Collaborative learning platforms based on the room metaphor are often referred to as “cooperative knowledge spaces” [JCL+ 07]; examples of such platforms are CURE [HSH+ 04] and open sTeam [HK01]. A key feature of room-based systems is their ability to partition the “world” of people and objects which has a structuring effect on both objects and discussions. Thus discussions can take place in close proximity to the objects being discussed, while avoiding disturbance by other discussions which regularly occurs if the only communication tool is one single, course-wide forum. Access to a room can be controlled and limited, enabling users to share privileged access to objects in a smaller group. Both these features considerably facilitate finding a shared context for group work. Moreover, as we have pointed out earlier, discussions in a soft-skills context tend to have sensitive subjects as well. A more confidential environment can foster these discussions, which are an important part of a soft-skills course. Since virtual rooms can be arranged and connected by doorways freely, semantic relationships between objects in different rooms can be visualized and made perceivable to users. Room-based systems also support awareness of other users” presence. Obviously, a person present in the same room is a potential communication partner, making awareness a key prerequisite for communication and (synchronous) collaboration. Sensing that person”s actions enables coordination and helps to retrace the actions. The ability to coordinate is a key feature to leadership and traceability of actions is helpful in conflict management. Thus, awareness can give substantial support for real-world training scenarios. To sum up: structuring and awareness features of room-based collaborative platforms can enhance the learning experience in soft-skill courses fundamentally.
5 Outlook We have outlined the downsides of a mainly content-oriented approach to online professional soft-skill training and how they may be overcome by switching to a community-oriented approach. Providing information to the participants is not the
306
Olivier Pfeiffer et al.
main goal of a soft-skill course as described here. Soft skills need to be practiced, and we argue that a collaborative environment makes for a good training ground. It should be observed that simply putting content into a cooperative knowledge space will not solve the issue. Rather, the whole course will have to be rethought focussing on collaborative scenarios and a thematic structure that is reflected by the concept of virtual rooms. We expect to improve the quality of the courses the employees attend, thereby improving their soft skills as well.
References [Aro04] [Dou]
Shipra Arora. Soft skills training: Outsourced vs in-house. IT People, April 2004. Martin Dougiamas. Moodle.org: open-source community-based tools for learning. http://moodle.org/, last visited 2008-04-18. [GR03] S. Greenberg and Mark Roseman. Using a room metaphor to ease transitions in groupware. In M. Ackerman, V. Pipek, and V. Wulf, editors, Sharing Expertise: Beyond Knowledge Management, pages 203–256. Cambridge, MA, MIT Press, January 2003. [Hed07] Karyn Hede. SCIENCE CAREERS: Managing Scientists. Science, 318(5852): 993–995, November 2007. [HK01] T. Hampel and R. Keil-Slawik. sTeam: structuring information in team-distributed knowledge management in cooperative learning environments. ACM Journal on Educational Resources in Computing, 1(2):1–27, 2001. [HSH+ 04] J.M. Haake, T. Schummer, A. Haake, M. Bourimi, and B. Landgraf. Supporting flexible collaborative distance learning in the CURE platform. In Proceedings of the 37th Annual Hawaii International Conference on System Sciences. IEEE Computer Society, 2004. Sabina Jeschke, Sabine Cikic, Nadine Ludwig, Olivier Pfeiffer, Uwe Sinha, and [JCL+ 07] Christian Thomsen. Virtual Room Concepts for Cooperative, Scientific Work. In Proceedings of the 2nd International Conference on Interactive Mobile and Computer Aided Learning (IMCL) 2007. Kassel University Press, April 2007. [KBGM90] M. Kane, S. Berryman, D. Goslin, and A. Meltzer. Identifying and Describing the Skills Required by Work. Technical report, U.S. Department of Labor, September 1990. [RT95] J. Roschelle and S.D. Teasley. Construction of shared knowledge in collaborative problem solving. In C. O’Malley, editor, Computer-supported collaborative learning, pages 69–197. Springer, 1995. [SKS06] G. Stahl, T. Koschmann, and D. Suthers. Computer-supported collaborative learning: An historical perspective. In R.K. Sawyer, editor, The Cambridge Handbook of the Learning Sciences, pages 409–426. Cambridge University Press, Cambridge, 2006. [Wes05] Martin Wessner. Kontextuelle Kooperation in virtuellen Lernumgebungen, volume 8 of Schriften zu Kooperations- und Mediensystemen. Josef Eul Verlag, Lohmar-Köln, 2005.
LiLa: A European Project on Networked Experiments Thomas Richter, David Boehringer, Sabina Jeschke
Abstract The LiLa project – short for “Library of Labs” – is a European Community funded project to network remote experiments and virtual laboratories. The goal of this project is the composition and dissemination of a European infrastructure for mutual exchange of experimental setups and simulations, specifically targeted at undergraduate studies in engineering and science. This article discusses the architecture of the project, introduces its components and sheds some light on our motivation and background. Keywords Remote Experiment · Virtual Laboratory · Automated Course System
1 Introduction Hands-on courses are – in addition to lectures – one of the fundamentals of engineering education. Students learn here how to solve practical problems, and delve into experimenting with real equipment. Besides theory and practical experiments, simulations also become relevant in science and engineering; increasing costs force engineers to substitute expensive or complex experiments by simulations – sometimes not even to cut costs, but also to gain insights not or only hardly achievable due to physical constraints otherwise. However, increasing complexity also limits the ability of universities to fund courses on such matters, and often we find the situation that only a limited number or a limited corpus of experiments can be made available for student labs. To address this problem, universities setup – now and in the past - Remote Experiments and Virtual Laboratories. While the former are “real” experiments that are, however, remotely controlled over the internet, the latter are highly flexible environments to run simulations. Both enable students to make use of the equipment 24 h/7 days
T. Richter (B) RUS Computing Center, University of Stuttgart Allmandring 30a, 70550 Stuttgart, Germany e-mail: [email protected]
a week, making them independent on the opening hours of the lab and the work schedule of the staff. Up to now, most of such solutions were restricted to single universities or institutions – for a couple of noteworthy exceptions, see the next chapter. That is, the equipment and software available has been limited by the abilities and funding of the controlling institution. However, since said experiments are by construction available over the internet, an obvious improvement of this situation is to found a federation of supporting institutions, and allow students mutual access to the equipment available in this federation in total: This is the major goal of the LiLa project, namely to setup a Library of Laboratories across Europe, and to share resources available in this network. LiLa is funded by the European Community by its eContentplus program. However, goals of this project go beyond generating the software necessary to setup said network: its aim is not only to share and increase the utilization of the equipment, but also to help students to find the experiments they need, to integrate the experiments into electronic library catalogs, to link them to “traditional” media as for example lecture notes, and to equip and extend the experiments by courses build from this media; last but not least, an important further goal of LiLa is to integrate such interactive courses into the curricula of universities. Clearly, to make this vision to become true, contracts and legal side-conditions must be worked out and understood, and LiLa attempts to provide contracttemplates to make it as easy as possible for interested partners to join the consortium. This article is structured as follows: We will present some related projects in the next section and discuss similarities and differences between them. Following that, we introduce the architecture of the LiLa project and give some examples of the existing content. We close by a conclusion.
2 Related Works A project similar to that of LiLa is driven by the MIT in the USA: Similar to LiLa, the iLabs project [ila, HdAL+ 08] supported by Microsoft aims at making experiments remote-controlled, and thus having them accessible by web-services for other members over the internet. In fact, one of the LiLa members, the University of Cambridge, is already participating in the iLabs project, and we aim to cooperate with iLabs in a long term perspective. Very similar to iLabs, LiLa is not focused on a specific topic, but addresses bachelors in all engineering and scientific studies. Similar to iLabs, we aim at a “single sign on” process to gain access to our resources. This process will be integrated into the Wonderland architecture (see below). The Blekinge Institute of Technology in Sweden started in 2007 the VISIR project [GZH+ 07], which is, however, unlike LiLa not supported by the European community. Quite similar to LiLa and iLabs, VISIR tries to increase laboratory utilization by sharing remote-controlled equipment across universities. Another similarity is that this, and many other projects [BI06, JRS+ 05] deploy the LabView software by National Instruments to wire experiments to the internet.
LiLa: A European Project on Networked Experiments
309
2.1 Differences and Concepts Despite all similarities, some major differences exist between the projects introduced above and LiLa: the driving motivation of LiLa is to establish remote experiments and virtual laboratories as a core component of engineering education, to integrate them as such into the curricula and to share those materials with others all over Europe - and even beyond. For the achievement of this goal, didactical guidelines are developed, a portal for the access to remote and virtual experiments including an access control and scheduling system is set up, and the experiments are enriched with “traditional learning material” like lecture notes and recordings, exercises and assessments, and content the libraries can provide. As learning assistance, a tutoring system allowing individual learning paths is guiding students through the material. Last but not least, the virtual world Wonderland supports the LiLa architecture by providing a framework for collaboration: fostering the cooperation between students working on experiments is of special concern to our project, and we see it as one of the most important non-technical skills trained in “traditional” laboratory courses. We believe that LiLa goes beyond comparable projects due to this comprehensive approach.
3 The LiLa Architecture The LiLa architecture is structured in four tiers or layers, see Fig. 1: The lowest tier consists of the content, where we understand the content to include remote experiments and virtual laboratories – and not only to consist of static documents. All content modules are to be annotated by suitable metadata to make them locatable and available, and to allow their integration into library catalogs. Of course, we cannot expect that a metadata system like the DDC notation originally developed for
Fig. 1 Architecture of the LiLa Project
310
Thomas Richter et al.
static documents to be suitable for interactive components; to this end, the design or extension of a metadata system is one of the tasks within LiLa. The second tier ensures accessibility of the content: On one hand, content (i.e. experiments and static data) must be locatable within the pan-European LiLa network. That is, we want to enable students and researchers to search for content available at participating members that satisfies their needs, similar to the way how a library catalog is used to locate suitable books. On the other hand, remote experiments – unlike simulations – are also a resource with limited availability that needs to be shared fairly between users and for which varying access policies must be established. The second tier thus includes a booking system granting access to the content, and controlling the privileges of the users. The third tier is responsible to integrate experiments and documents to interactive courses and implements an interactive and intelligent course system that guides students through an – real or virtual – experiment. A prototype of a corresponding system has already been developed at the TU as the outcome of earlier projects, and is currently rolled out at the University of Stuttgart. Further details on Marvin and on its learner- and course model are found in chapter IV and in [JJP+ ]. The user interface as forth layer will be represented by and in the virtual world of Wonderland by Sun Microsystems. Wonderland has been originally developed to ease the cooperation of home-office workers; its purpose in LiLa is to integrate the experiments, courses and documents into a consistent virtual environment, and make equipment, represented by virtual objects, available to the users’ avatars. Due to the rather tight development plan, a two-dimensional projection of a traditional user-interface must be sufficient in the first project phase. Alternatively, we plan a traditional web-front end for experiments, thus the web-browser becomes the user interface.
4 Examples and Components A project of the given size cannot be constructed from scratch; instead, our target is to use existing components as far as possible, and to reach the project goals by merging existing solutions. This is also required by the eContentplus programme of the European community, which targets at making existing content more accessible.
4.1 Remote Experiments The first type of content made accessible by the LiLa network is that given by “Remote Experiments”: A remote experiment is an experimental setup controlled by a common PC; by appropriate software – typically LabView [lab] by National Instruments – sensor data and control parameters of the experiment are made accessible over the internet. Remote experiments are already deployed at our project parters in Berlin, Cambridge and Basel. An experiment from thermodynamics run at the TU Berlin may aid as an example: A motor controls the position of a piston in a glass cylinder, expanding or
LiLa: A European Project on Networked Experiments
311
Fig. 2 A remote experiment on phenomenological thermodynamics. Left: A piston compresses air in a glas cylinder. A heater (below) controls the temperature of the air. Pressure and temperature are measured by the PC. Right: The LabView user interface of this experiment.
compressing the gas contained within. An electrically controlled valve enables users to fill the cylinder with air, and a heater can be turned on to change the temperature of the gas volume. A pressure sensor measures the pressure of the gas. All actors – the motor, the valve and the heater – are steered by LabView; similarly, all sensors are wired to the PC making sensor data available to the user. The LabView front-end bundles sensor and actor data, and an additional web-cam allows users to observe the movement, cf. Fig. 2. In a typical experiment, the student first opens the valve and fills the cylinder with air; afterwards, the valve is closed and the moter is run to compress the isolated gas volume - by that, pressure increases. By turning the heater on, temperature is increased as well. If the contained air is now allowed to expand again by moving the piston out of the cylinder, pressure decreases again, but will not reach exactly the same value as it had before compression - this only happens after allowing the gas to cool down. By plotting the pressure over the volume in a so-called pV diagram, one observes that the pressure-volume state of the gas in this process forms a closed loop, and one can now show theoretically that the area enclosed by this loop is proportional to the amount of work performed on the gas – and is thus given by the electrical work of the heater. Experiments similar to the one described above are part of the experimental corpus of every undergraduate engineering course. At the TU Berlin, it is not unusual that more than 1000 students per semester participate in courses on experimental physics.
4.2 Virtual Laboratories Unlike remote experiments, virtual laboratories are simulation frameworks that run on the computer only and do not interact with a physical experiment; even though one could simply ask students to install the simulation software on their home PC,
312
Thomas Richter et al.
Fig. 3 A virtual experiment on the Second Law of Thermodynamics. Left the initial state with the gas (yellow) in one corner of the container (red). Right the final state after running the experiment for a while, also showing the entropy over time.
we prefer to use a client-server model here as well. First of all, the amount of work for installation is minimized, but second – and even more important – the server architecture allows students to interact with each other and collaborate on one simulation. An example for a virtual laboratory is “VideoEasel”, currently deployed at the University of Stuttgart, a simulation framework for experiments in multiparticle physics. Further virtual laboratories are provided by our partners in Basel, Cambridge and Linköping in Sweden. A virtual counterpart of the above experiment on thermodynamics can be realized in VideoEasel, cf. Fig. 3: The laboratory simulates here an idealized and simplified gas, a so-called lattice gas [HPdP73, HPdP76]. In a typical experiment, the student first draws a gas container, and fills one edge of this container with particles. A sensor measuring the entropy of the gas volume is then added to the simulation – the entropy is a physical quantity measuring the amount of disorder of a multi-particle system. When running the simulation, the gas spreads out into all of the container, and by that increases the entropy – following the Second Law of Thermodynamics. While this is clearly to be expected, one can now extend this experiment in a way that is not possible in reality: by inverting the velocities of all gas particles, the movement is inverted and the gas flows back to its initial location by decreasing the entropy. It is now interesting to let students discuss why this movement, while possible by the laws of physics, cannot be observed in reality.
4.3 Tutoring Software and Intelligent Assistants Running complex experiments usually requires students to consult an experienced fellow student or an assistant. However, in order to help students to experiment outside of the regular opening hours of labs, simulations and remote experiments are equipped and extended by electronic course systems that guide users through an experiment. An example for such a course system is given by the “Marvin” system, originally developed for VideoEasel [JJP+ ].
LiLa: A European Project on Networked Experiments
313
Fig. 4 Path model of a course, consisting of learning units. Arrows indicate dependencies between units, from the goal to the initial node. Students follow the course in opposite direction, resolving prequisites first. On the left: A decision point (green).
A course within this system is understood to be built from “learning units”, where each learning unit encodes both its contents and the perquisites a student must have to be able to follow it, cf. Fig. 4. The course system is now responsible to resolve all dependencies by constructing a suitable learning path – defined as a sequence of learning units of a course. To evaluate the learning success of a learning unit, Marvin uses a plug-in mechanism that links code at run-time to the experiment to evaluate the user behavior there, and to feed back the results into the course system. In addition, Marvin uses a statistical approach to optimize the learning strategy: This mechanism is used whenever more than one learning path connects the current node with the final learning goal, i.e. whenever the system reaches a decision point (the green node in Fig. 4). At those points, Marvin recommends either one or the other node depending on which maximized the learning success in the past on average. For more details, see [JJP+ ].
4.4 Cooperation in Virtual Worlds The user interface of most experiments and simulations remains necessarily abstract and reassembles the look and feel of a real laboratory only to a very minor degree. Specifically, a real lab allows students to communicate to each other and to cooperate in performing a course. Unfortunately, this important element often under-estimated in similar projects, even though it is very relevant for the working engineer or scientists: Teamwork is the common practice given the complexity of today’s technology.
314
Thomas Richter et al.
Fig. 5 A screen shot of the virtual laboratory with the tutoring system in a separate window on top. Here an experiment on reflection.
Fig. 6 Screen shot of Wonderland: Avatars in front of a projection surface of an embedded web browser (Source: Sun Microsystems).
LiLa: A European Project on Networked Experiments
315
Sun Microsystems “Project Wonderland” [pro, Wat09] has been developed with quite similar applications in mind, namely to allow and foster the cooperation of teams of home-working employees in companies. Within the LiLa project, Wonderland will be user interface and primary access tool for the experiments provided by the network. Fig. 6 demonstrates a typical look of the virtual world: Colleagues, represented by avatars, move around in the virtual world and can talk to each other, by using the built-in voice over IP functionality. Simulated projection surfaces allow running presentations or embed “flat” desktop programs into the world. In the LiLa project, we will aim at constructing virtual three-dimensional representations of real experiments that are close to look and feel, but due to time constraints, only flat two-dimensional representations as those shown in Fig. 6 will become available in the first project phase.
5 Partners and Roles The LiLa consortium consists of eleven institutions from seven European states: the University of Stuttgart, the initiator and coordinator of the project; the Institute of Technology (TU-Berlin), the Universities of Basel, Cambridge, Link¨’oping and the Aristotle University of Thessaloniki contribute their remote experiments and their virtual laboratories – namely content – to LiLa. Thessaloniki is furthermore responsible for the evaluation of the project results. The University of Delft is our key partner for didactics: they are responsible for the didactical preparation and evaluation of the project results. The Universidad Politénica de Madrid is responsible for developing and implementing the access control system, i.e. tier 2 in Fig. 1. The second activity in that tier is the development of the localization services and the embedding of the content into library catalogs by link-resolver technology, which is the responsibility of the Library of the University of Stuttgart. Three non-academic partners contribute their technology to the project: First, Sun Microsystems (now Oracle) support us in enhancing their Wonderland technology (see section IV.D) for the needs of LiLa; Computational Modeling Cambridge, a spin-off of the University of Cambridge, develops simulation software for virtual laboratories and will contribute their developments to the LiLa network; and finally MathCore Engineering from Linköping, which is our 3D expert, will develop the virtual 3D models of the experiments and will design the virtual world of LiLa.
5.1 Work Packages and Partners The LiLa project is structured in six work packages, each of them consists again of several tasks; every work-package is managed by one of the consortium members, even though the tasks within a work package might be handled by members different from the work package leader.
316
Thomas Richter et al.
In detail: The first work package is the management and monitoring of the project as a whole, which is in the responsibility of the University of Stuttgart; the virtual world and its interfaces, the access control and booking system are designed in the second work package by the University of Madrid. Didactical models, the design and evaluation of interactive courses and course materials is in the responsibility of the Technical University of Delft, where some components are also created in Stuttgart, namely the course system described in section IV.C. The fourth and largest work package is managed by the Institute for Technology Berlin (TUB): This work package integrates existing content into the LiLa architecture, where content is not only understood to consist of interactive experiments, but also includes “more traditional” material like lecture notes and scientific publications. In this work package, we will also design a meta-data set suitable for the annotation of interactive material; this specific work lies in the hands of the Library of the University of Stuttgart. The fifth work package monitors and evaluates the project results from the scientific and didactical view point; the work package leader is the University of Thessaloniki, even though Delft will run the didactical evaluation as part of this work package. Last but not least, the sixth work package disseminates project results, is responsible for making project results visible by publications, and will also offer training courses and material for lecturers, students and university administrations to aid them in setting up, running and using remote experiments and virtual laboratories. An important task of this package is also the development of a contract template to ease interesting parties joining the network; legal constraints for exchanging resources between universities need to be identified, and a legal framework for shared resources needs to be worked out here. It is interesting to note that we found in a preliminary questionnaire that most universities do have as much trouble in earning money by sharing experiments as they have in spending money for it. Thus, mutual exchange of resources is likely the model to be implemented.
6 Conclusions Even though LiLa is an ambitious project for constructing an infrastructure for “virtual” experiments, we want to stress that it is not our aim to substitute traditional hands-on training; the value of such courses lies beyond the gained scientific insight, namely in getting acquainted to laboratory equipment, and in training the social communication skills with fellow students, colleagues and tutors. Even though we try to mimic these structures as far as possible in the virtual world, their replication remains necessarily incomplete. Instead, we hope that LiLa makes the best out of the financial problems universities have to face – and establishes a strong federation of universities which, as a group, are able to support their students better than a single isolated institution could.
LiLa: A European Project on Networked Experiments
317
References [BI06] [GZH+ 07]
[HdAL+ 08]
[HPdP73]
[HPdP76]
[ila] [JJP+ ]
[JRS+ 05]
[lab] [pro] [Wat09]
H.A. Basher and S.A. Isa. On-campus and Online Virtual Laboratory Experiments with LabVIEW. In Proceedings of the IEEE SoutheastCon, pages 325–330, 2006. I. Gustavsson, J. Zackrisson, L. Håkansson, L. Claesson, and T. Lagö. The VISIR project – an Open Source Software Initiative for Distributed Online Laboratories. In Proceedings of the Annual Int. Conf. on Remote Engineering and Virtual Instrumentation, 2007. V. J Harward, J. A del Alamo, S. R Lerman, P. H Bailey, J. Carpenter, K. DeLong, C. Felknor, J. Hardison, B. Harrison, I. Jabbour, P. D Long, Tingting Mao, L. Naamani, J. Northridge, M. Schulz, D. Talavera, C. Varadharajan, Shaomin Wang, K. Yehia, R. Zbib, and D. Zych. The iLab Shared Architecture: A Web Services Infrastructure to Build Communities of Internet Accessible Laboratories. Proceedings of the IEEE, 96(6):931 –950, June 2008. J. Hardy, Y. Pomeau, and O. de Payssis. Time evolution of two-dimensional model system I: invariant states and time correlation functions. J. of Math. Physics, 14:1746–1759, 1973. J. Hardy, Y. Pomeau, and O. de Payssis. Molecular dynamics of a classical lattice gas: Transport properties and time correlation functions. Phys. Rev. A, 13:1949– 1961, 1976. iLabs: Internet access to real labs - anywhere, anytime. http://icampus.mit.edu/ iLabs/. M. Jeschke, S. Jeschke, O. Pfeiffer, R. Reinhard, and Th. Richter. Intelligent Training Courses in Virtual Laboratories. In Proc. of ED-Media 2006 (Orlando), pages 2069–2074, VA, USA. Association for the Advancement of Computing in Education (AACE). Association for the Advancement of Computing in Education, Norfolk. S. Jeschke, Th. Richter, H. Scheel, R. Seiler, and C. Thomsen. Das Experiment und die eLTR-Technologien: Magnetismus in Virtuellen Laboren und RemoteExperimenten. Bonner Köllen Verlag, 2005. LNI. Labview by National Instruments. online document available at http://www.ni.com/ labview. Project Wonderland website. https://lg3d-wonderland.dev.java.net/. John K. Waters. Sun Makes Its MUVE, sidebar in Ä ’Second Life’ For Educators. T.H.E. Journal, January 2009.
VIDEOEASEL - A Flexible Programmable Simulation Environment for Discrete Many Body Systems Thomas Richter, Sabina Jeschke, Olivier Pfeiffer
Abstract In this work, we present a Virtual Laboratory providing a simulation framework for discrete many-body systems. Programs defining the dynamics of the system and instruments measuring on the simulation can be easily implemented within its own programming language, and can be linked and edited at run time. The system class that can be covered within this framework reaches from discrete difference equations over classical many-body problems is physics to research problems in image processing, allowing us to apply this laboratory in education and research. Keywords Virtual Laboratory · Explorative Learning · Mathematics · Cooperative Virtual Knowledge
1 Introduction Many particle systems show a rich and sometimes even surprising set of phenomena; the interaction of many small systems can create unexpected behavior at macroscopic level that is not easily derived from the dynamics of its individual participants. In physics, typical many-body phenomena are phase transitions and nonreversibility of macroscopic dynamics, the models used for studying these effects are true classics of physics: The Ising model describing the ferromagnetic effect, and the lattice gas model as a very simplistic description of an ideal gas. But even beyond physics, many-body dynamics are an important field to be studied: In biology, the almost unpredictable population change of a predator-prey system, and in chemistry patterns seen in oscillating reactions are all phenomena that can be understood as the outcome of the interaction of very many, but very simple constituents at small scale.
T. Richter (B) RUS Computing Center, University of Stuttgart Allmandring 30a, 70550 Stuttgart, Germany e-mail: [email protected]
In this work, we introduce a framework that allows studying all the above and related systems, in a quantitative matter, giving hands-on access to a surprisingly large and rich system class in an extremely comfortable, flexible and powerful manner. This work is organized as follows: In the next section, we introduce the common mathematical framework on which our models are based and give a review of related works. The next section describes the design of our implementation. We then present example systems as use-cases for the laboratory, and then give some insight into the capabilities of the laboratory and its programming language. We finally describe our experience of deploying this framework in higher physics education and conclude with an outlook on future work.
2 Mathematical Framework The simulation environment presented in this paper is based on the idea of Cellular Automata (CAM) [TM87] already deployed much earlier for similar purposes; a Cellular Automaton is defined by a regular grid of cells, where each cell can be in one out of finitely many states. The dynamics of the system is defined by assigning a finite state machine to each cell, which, depending on its own state and the state of its neighbors, computes the next state of the cell in the next time step. This cell grid is often visualized by assigning to each cell state a color, and representing the cells as colored pixels of a computer display. Historically, Conway’s ”‘Game Of Life”’ was one of the first popular mathematical “games” [Gar70], even though the concept of a cellular automata goes back to earlier times, e.g. Zuse [Zus69] already experimented with this idea. In 1978, Toffoli and Margolus built a hardware implementation that was programmable in Forth [TM87] and could execute programs in real-time; many interesting applications can be found in their book. Wolfram studied cellular automaton to demonstrate that for many, even very simple automata, the most effective description of the patterns they generate is the automaton rule itself [Wol02]. Thus, for most setups, the complexity of the complete cellular automaton machine cannot be reduced further by any mathematical description. Recently, interest in this field is diminishing since writing a suitable automaton to model a specific multi-particle system is often hard and not straight-forward, the programming model is often too complex, and quantitative measurements required for any scientific analysis are also easily neglected. Furthermore, it is often hard to establish strict mathematical proofs in this field. Many early works also concentrated on purely qualitative experiments. We still believe that CAMs have lots to offer, they provide a suitable framework for many experiments that have important lessons to tell, especially for students in an educational environment. We will provide some examples in the next sections, but also point out that many of the early problems can be resolved today: Our automaton is freely programmable in a language closely related to Java or C that provides more flexibility and features than earlier implementations. To allow qualitative measurements, the necessary instruments can be designed and implemented in the same language along. The code also integrates
VIDEOEASEL - A Flexible Programmable Simulation Environment
321
into mathematical toolkits like Maple [Map08], or learning management systems like moodle [moo08].
3 Implementation A VideoEasel program describes n coupled cellular automata, each of them can be understood as a two-dimensional plane of cells representing a cellular automaton in the traditional sense. However, cell planes can also interact to its nearest neighbor planes, and thus coupled automata or three-dimensional setups are possible. Each cell plane is defined by a set of parameters and the automaton itself, specified in a C-like programming language. This program is conceptionally executed on all cells of a plane in parallel, computing the state of the plane to the next time stamp given its and the neighbor plane states to the current time. Parameters are exposed to the user in the form of GUI elements like sliders or checkboxes, and even the program can be manipulated at run-time. Measurement instruments are also represented as cellular automatons, though they do not modify the cell configuration, but rather modify output parameters collecting the result value of the measurement. For example, a very simple measurement instrument counts all cells in a given state, and computes from that the relative density of these cells in the overall configuration; the magnetization of the Ising model, described in the next section, can be computed in this way. However, the programming language is flexible enough to allow operations that are more complex in the measurement process. The laboratory software is split into two separate functional blocks: First, the number crunching core which compiles the CAM from the VideoEasel programming language into machine code and performs the simulation itself, and one or several clients that connect to this server over the internet. The purpose of the clients is to either visualize the simulation and provide access to the experiment and its parameters, or to act as an interface to additional software, for example to Maple or LabView. This flexibility allows not only several users to share the same experiment, but it also integrates mathematical software seamlessly into the laboratory. The server, due to its computational demands, is written in C++, whereas most of the clients have been implemented in the Java language, offering platform independence. Java has the additional advantage that it allowed us to provide an applet version of the client that integrates nicely into browsers, and also makes the laboratory available as a learning element in a SCORM compliant learning management system.
4 Examples In this section, we discuss three typical example systems to demonstrate the capabilities of our laboratory. First, a classical lattice gas system first introduced by Hardy et all [HPdP73, HPdP76] is described; this is a reversible and deterministic
322
Thomas Richter et al.
system, but provides a lot of insight into the nature of the second law of thermodynamics. The second example is the Ising model of Ferromagnetism [Isi25], a non-deterministic automaton allowing the discussion of phase transitions in manybody systems. Our third example is a stochastic image-denoising algorithm first considered by Geman & Geman [GG84]. The examples present important properties of the VideoEasel system: First, the automaton rules can be formulated in a way that avoids otherwise typical problems in this area, e.g. the formulation of rules that conserve the particle count, second, quantitative measurements are formulated similarly to the dynamics and are intrinsically part of the framework. Further, the Ising model shows that built-in tunable random generators allow the construction of stochastic systems. Last but not least, the Geman & Geman example will demonstrate that it is important to consider the case of coupled automaton systems. We will provide code to demonstrate how to setup these experiments, show the corresponding source code, and will demonstrate some of the outcomes.
4.1 The Lattice Gas Model In this model, cells can be in one of three states: They can be in a zero state, representing empty space, in a particle state, representing a space volume filled by one particle of the lattice gas, and in a “wall” state, representing the container wall confining the lattice gas. The automaton has now to be defined in such a way that the particle movement follows the laws of physics, i.e. particles move freely, reflect on the container and collide under preservation of energy and momentum. One of the challenges of this system is to define the automaton rule now in such a way that the number of particles represented by the configuration remains constant. It turns out that this problem is resolved best by using an alternative description of the automaton called the "‘Margolus"’ rule [TM87]: Here the cells of the lattice are first grouped into a super-lattice of 2x2 cell-blocks, and the elementary automaton rule is formulated in these super-blocks such that the number of cells representing particles in a super-block remains constant. Then, the division of the lattice into super-blocks is altered every other time stamp such that on even time stamps the origin of the superblocks is at even-even locations and in odd time-stamps is at odd-odd locations. This yields a particle movement into four diagonal directions. An interesting aspect of this simple system especially for educational purposes is that even though it is easily verified that this automaton obeys time-reversal symmetry, the second law of thermodynamics holds, i.e. entropy increases over time. In our laboratory, unlike in many earlier works, this can be quantitatively verified: A measurement instrument computing the 4x4-block entropy is easily established by computing the statistical distribution of all 4x4-block configurations. Even though this still an approximation of the abstract entropy function, the plot of the approximated entropy over time is observed to be almost monotonically increasing, establishing the second law of thermodynamics. The resolution of the antinomy between
VIDEOEASEL - A Flexible Programmable Simulation Environment
323
the second law and the reversibility of the system is also an interesting section of the history of physics.
4.2 The Ising Model Ernst Ising defined this model in 1921 in his thesis [Isi25] to establish a simple mathematical system to study phase transitions in many-particle systems. Even though this model seems simple, it took more than 30 years to prove that the two-dimensional version of this model is indeed capable of reproducing this effect [Ons44]. In this system, cells are in two possible states, representing elementary magnets — also called “spins” in the following — that can be magnetized in two different directions: Spin up and spin down. Spins couple to their four nearest neighbors by a local energy function that is lower for spin pairs that point in parallel direction than for spin pairs pointing in anti-parallel direction, and they further couple to an external field H by energetically favoring spins adjusted in parallel to H. The simulation in our laboratory computes now the Metropolis dynamics of this system [MRTT53]: The spin orientation is reversed whenever the local energy contribution can be lowered by a spin-flip, or if a heat-bath can provide the energy for the flip. The heat-bath is represented by a random generator, providing a random distribution proportional to exp(-1/T), where T is the temperature of the heat bath. By adjusting the parameter T while the simulation is running and observing the system, students can now determine the temperature at which a phase transition occurs. Similar to the first model, more quantitative experiments are available as well: It is not difficult to measure the inner energy U and the entropy S, and from that derive the Helmholtz Free Energy F = U − T ∗ S. Further, one can measure the magnetization μ of the system. It is now interesting to compare the plots of F and μ over the external field H to find that μ is nothing but the (negative) derivative of F by H. This relation can also be verified from the Gibbs state of the model.
4.3 Stochastic Image Denoising The idea of using methods of statistical mechanics for image denoising purposes goes back to an article of Geman & Geman [GG84]. The image is here understood as a two-dimensional Ising-like model where the grey-levels of the image pixels are represented by the orientation of the pins. Unlike in the Ising model, more than two grey-levels, and thus a larger number of spin orientations are possible. As in the Ising model, an energy functional is defined by the quadratic distance of grey-levels of neighboring pixels, and a Metropolis [MRTT53] algorithm tries to minimize the total energy by adjusting pixel grey-values at random. In addition, however, the pixel-pixel interaction between neighboring pixels can be interrupted by a second
324
Thomas Richter et al.
process operating on the edges of the pixel lattice. The aim of this process is to find edges in the image, and disrupt pixel interactions across those edges to protect the edges from blurring by the Ising process. The edge process also contributes to the energy functional by penalizing edge configurations that are unlikely to occur in natural images, e.g. T-junctions are less likely than straight edges. Both processes are, similar to the Ising model, coupled to a heat-bath. It can now be shown [GG84] that pixel configurations of the coupled processes now converge under simulated annealing, i.e. when the temperature is decreased slowly, to images consisting of separate areas of constant grey-level. In VideoEasel, pixel and edge process are represented by two cell planes, i.e. two cellular automata that are coupled to each other. The pixel process computes the pixel-pixel interaction energy by first checking for interrupted pixel bonds in the edge process, and the edge process has to check the pixel values to separate unlikely from likely edge configuration. The temperature of the heat bath becomes an external parameter users can adjust, and thus can control the annealing process. The gradual improvement of the image quality is then easily observed. However, similar to the above examples, the improvement in image quality can also be measured quantitatively: For that, a non-distorted reference image is loaded into the simulation, and a measurement instrument measures the mean square error between the reference and the reconstructed image. The logarithm of this function, also called PSNR, is the most popular image quality metric, and can be seen to increase over time under the Geman & Geman process.
5 Evaluation The focus of the virtual laboratory introduced above is to provide experiments to demonstrate abstract phenomena to graduate students of physics, mathematics, and engineering courses at university level. However, applications in research, undergraduate, or high school education are also possible. We first deployed the virtual laboratory in the graduate course “Mathematical Physics II” the TU Berlin. This three-semester course covers in its second semester models of statistical mechanics, specifically the Ising model and the lattice gas model. Even though this course is taught at the institute for mathematics, the majority of participating students are typically physicists; the group size is typically between 15 and 20 students. While the lecture covers the theoretical aspects and the mathematical background of the models, we used the practice group of the course to guide the students to experiments on the theory discussed in the lecture before. Specifically, the Metropolis dynamics of the Ising model got introduced, the phase transition was measured on the virtual experiment and the relation between magnetization and free energy was derived experimentally. Students were asked to carry out the experiments, perform the measurements, and collect all necessary data, and were requested to put this data in relation to the material learned in the lecture. To
VIDEOEASEL - A Flexible Programmable Simulation Environment
325
our delight, students did find the requested relation between the two quantities, and were able to derive them by their own from the mathematical model. At the end of the course, we requested students to fill out an anonymous query form on the lecture and the practice group. This standard form is kindly provided by the student union of the TU Berlin, and used consistently for the evaluation of all lectures of the institute. According to this evaluation, the lecture received an average grade of 1.7 on a 1 to 6 scale, 1 being best and 6 being worst, thus placing this lecture in the best third fraction. Students appreciated mostly that they could relate the theoretical material and models to practical experience and could gain some hands-on approach on the abstract definitions learned in the mathematical course. Interestingly, the course also triggered some interest in the actual implementation and infrastructure of our virtual laboratory which we couldn’t delve into in the group due to time constraints. Even though experiments on the lattice gas have often been demonstrated to students with great success, we did not yet had the chance to discuss the model in a similar experiment, unfortunately.
6 Conclusion and Outlook VideoEasel is a comprehensive and elaborated tool for studying statistical mechanics and systems from related fields. Its flexible architecture allows deployment in both education and research: For the former, a SCORM compliant applet version eases the integration into learning management systems like moodle or Ilias; for the latter, process control and analytical or numerical evaluation of quantitative results can be provided by Maple or similar toolkits. Let us emphasize that Virtual Laboratories as described can play an important role in education: Conceptionally located between theoretical concepts and practical applications, they can act as pedagogical bridge to make students “see how the theory works”. Thus, they offer the opportunity to experience mathematics in action, an argument supported by our classroom experiment. Being available 7 hours / 24 days a week, Virtual Laboratories, however can also help to extend the experimental capabilities of real laboratories. They can support or complement existing experiments or provide experiments not possible otherwise due to spatial, security or other constraints. Especially the combination of remotely controlled real experiments with virtual laboratories offers great opportunities: It allows students to compare the outcome of a real experiment, compare that with a similar simulation, and see the results in relation to the prediction of the mathematical theory. Comparing all three side-by-side eases students to understand the role of experiments and the unavoidable limitations of every model in physical theories [JRTS07]. Our current focus is to establish VideoEasel as learning resource for graduate and undergraduate studies at the University of Stuttgart and, by that, collect additional experiences from its application in an educational environment. Furthermore, we
326
Thomas Richter et al.
Fig. 1 The Ising Model SCO of the virtual lab VideoEasel running in the browser.
equipped VideoEasel with an intelligent tutoring system [JJP+ 06] that allows us to run automated training courses for students where the courses existing so far focus on selected aspects of statistical mechanics. One of our future research works is to collect more experience with this system as well.
References [Gar70]
M. Gardner. The fantastic combinations of John Conways new solitaire game life. Scientific American, 223:120–123, 1970. [GG84] S. Geman and D. Geman. Stochastic relaxation, gibbs distributions, and the bayesian restoration of images. IEEE Trans. on Pattern Analysis and Machine Intelligence, 6:721–741, 1984. [HPdP73] J. Hardy, Y. Pomeau, and O. de Payssis. Time evolution of two-dimensional model system I: invariant states and time correlation functions. Journal of Mathematical Physics, 14:1746–1759, 1973. [HPdP76] J. Hardy, Y. Pomeau, and O. de Payssis. Molecular dynamics of a classical latticegas: Transport properties and time correlation functions. Physics Review A, 13:1949–1961, 1976.
VIDEOEASEL - A Flexible Programmable Simulation Environment [Isi25]
327
E. Ising. Beitrag zur Theorie des Ferromagnetismus. Zeitschrift für Physik, 31:253– 258, 1925. [JJP+ 06] M Jeschke, S. Jeschke, O. Pfeiffer, R. Reinhard, and Th. Richter. Intelligent Training Courses in Virtual Laboratories. In Proceedings of World Conference on Educational Multimedia, Hypermedia and Telecommunications (ED-MEDIA), volume Volume 1, pages 2415–2422, 2006. AACE. [JRTS07] S. Jeschke, T. Richter, C. Thomsen, and H. Scheel. On remote and virtual experiments in eLearning in statistical mechanics and thermodynamics. In Proceedings of the Fifth Annual IEEE International Conference. Pervasive Computing and Communications Workshops, pages 153–158, 2007. Times Cited: 0 5th IEEE International Conference on Pervasive Computing and Communications MAR 19-23, 2007 White Plains, NY. [Map08] Maplesoft. Maple, 2008. http://www.maplesoft.com/. [moo08] Moodle, 2008. http://moodle.org/. [MRTT53] N. Metropolis, A. Rosenbluth, M. Teller, and E. Teller. Equations of state calculations by fast computing machines. J. Chem. Phys, pages 1087–1091, 1953. [Ons44] L. Onsager. A two-dimensional model with an order-disorder transformation. Physial Review, 65:117–149, 1944. [TM87] T. Toffoli and N. Margolus. Cellular Automata Machines. MIT Press Cambridge, 1987. [Wol02] S. Wolfram. A new kind of science. Wolfram Media, 2002. [Zus69] K. Zuse. Rechnender Raum. Vieweg, 1969.
An Intensive Course in Mathematics for Engineers: Experiences and Prospects Mike Scherfner, Sabina Jeschke, Matthias Plaue
Abstract One main concern – especially after the implementation of the bachelor program in Germany – is how to manage the great number of lectures and how to give above-average students (but not only for them!) the opportunity to complete their course of studies quickly, but without loss of content or quality of teachings. In order to attack these problems we started an intensive course for selected students in order to offer them a unique learning experience by employing a special teaching concept, with appropriate training and exercises. Keywords Bologna Process · mathematics education for engineers · intensive courses
1 Introduction The bachelor’s degree was introduced at the Department of Mathematics at the Technische Universität Berlin (Berlin Institute of Technology) as well as at other German and European universities in the course of the Bologna Process, as the first level in a three-tier (BSc, MSc, PhD) graduation system. Some of the main goals are: Encourage students to be more mobile (within the European Higher Education Area), generate a greater convergence between the academic education in the United States and Europe, shorten the overall duration of studies. After the introduction of the bachelor program in Germany there arose the necessity for new concepts in mathematics education for engineers. In order to tackle this point the Department of Mathematics at TU Berlin has conceived a novel concept enabling selected engineering students to complete the better part of their mathematical studies in the condensed timeframe of one instead of three semesters. This course has already been employed in 2004/05 with a different approach by Lutz [LL05]. M. Scherfner (B) Institute of Mathematics, TU Berlin, Straße des 17. Juni 136, 10623 Berlin, Germany e-mail: [email protected]
To qualify for attending the course students had to participate in a written examination with school knowledge being the sole requirement. Subsequently, the 30 students with the best grades were admitted to the course. To this end, the content of the regular traditional lectures “Linear Algebra” and “Analysis I & II” has to be communicated and a special approach concerning the lectures and exercises must be employed. For each part of the course, students have to pass a written examination. Although the tutorials are small (10 students), because of the condensed timeframe a higher staff requirement does not occur. The intensive course program enables students with broader previous knowledge to immediately take advantage of their higher qualification. Thus, this model provides an interior differentiation that can be roughly compared with the honors class system in the U. S. The advantages of the new teaching concept showed in significantly improved pass-rates. The prospects resulting from our experiences will be presented in a way that could function as a guide for the employment of the approach with similar courses at other universities.
2 Requirements for Attending the Course We have restricted the number of students for the intensive course to a maximum of 30. The requirement for attending the course was the successful participation in a written qualifying examination. To successfully solve the problems given in this exam, only knowledge from school education was required, as well as the ability to apply that knowledge to standard exercises we took from usual class books. Additionally, we took one exercise from PISA (Programme for International Student Assessment, established by the OECD) for grammar school pupils in their third year. Any of about 1500 newly enrolled students could participate, approx. 100 students did so. The 30 students with the best results were allowed to attend the course. We want to point out that 40 percent of the students had no advanced courses for mathematics in school and not the highest grades. So with the qualifying examination we only checked the basic knowledge coming from the courses every pupil in Germany has to pass during gymnasium.
3 Requirements for Successfully Completing the Course Every first lecture of the week, a problem sheet was handed out that had to be solved in small groups of two students and returned the week after. The problems on each sheet were divided in three mandatory and five optional exercises. The students had to solve all of the three mandatory exercises and two of the optional exercises, the latter of which could be freely chosen. At the end of the semester, a problem sheet score of 60 % was required to attend the final written examination. Additionally to this requirement, the students had to participate in a total of three written exams: Two exams in the middle of the semester to verify the students’ skills in the subject
An Intensive Course in Mathematics for Engineers
331
matters “Analysis 1” and “Linear Algebra”, and one final exam about “Analysis 2” that was taken together with the other students attending the standard course “Analysis 2”. In this way, we could directly compare the pass-rates. Finally, we demanded active and regular attendance of the tutorials throughout the course.
4 Problem Sheets In the following, we exemplify our problem sheet design by presenting the exercises of the 11th sheet of a total of 14 handed out during the course, which had vector calculus and line integrals as a topic. As we discussed in the previous section, we divided every problem sheet in mandatory and optional exercises.
4.1 Mandatory Exercises Determine a potential for the vector field E(x, y, z) and calculate the line integral E ds, where k ⎞ 2x y + z 3 E(x, y, z) = ⎝ x 2 + 3z ⎠ , 3z 2 x + 3y ⎛
Compute the line integral
x
⎞ t 7 et = ⎝sin(πtet 2 )⎠ with 0 ≤ t ≤ 1 . k(t) t3 ⎛
where vds,
⎛
⎞ y−x v(x, y, z) = ⎝ −y ⎠ , 1
⎛ ⎞ 2t x(t) = ⎝4t ⎠ with 0 ≤ t ≤ 1 . t2
The parametrization of a cardioid is given by a(1 + cos φ) x(φ) = with 0 ≤ φ ≤ 2π . a(1 + sin φ) Calculate the arc length of this curve!
4.2 Optional Exercises How long is the spiral groove of a long-playing gramophone record? Model this problem mathematically and calculate the solution with realistic data. Let f : R2 → R be a twice differentiable function, and k : [0, 1] → R2 a differ entiable curve with the property f (k(t)) = c for some fixed c ∈ R (i.e. k is part of = 0? Justify your answer. a level curve of f ). Is it true that k grad f ds
332
Mike Scherfner et al.
Consider the vector field 1 v(x, y) = 2 x + y2 a) b) c) d)
−y with (x, y) = (0, 0) . x
Compute the line integral of v along the unit circle. Confirm that v = grad arctan xy for x = 0 and v = −grad arctan xy for y = 0. Does v have a potential? Is the line integral path independent? Draw a picture of the vector field and the computed line integral. A vector field v is said to have a vector potential A if v = rot A.
a) Prove the following: If v has a vector potential, then div v = 0. b) Consider the vector field ⎞ ⎛ 2 3x − 3y 2 + 6x z ⎠. −6x y v(x, y, z) = ⎝ 2 2 3x − 3z Determine a vector potential for v. Hint: One component of the vector potential can be chosen arbitrarily, e.g. A1 = 0. Determine the functions f : R3 → R for which the vector field v : R3 → R3 with ⎛ ⎞ f (x, y, z) v(x, y, z) = ⎝ x 2 + yz 2 ⎠ y2z has a potential.
5 Lectures There were four hours of lecturing every week, divided into two lectures. The original standard courses – “Linear Algebra”, “Analysis 1 and 2” – demand 2+4+4 hours instead, and we were thus able to significantly reduce the total number of lecturing hours. To achieve this goal while maintaining content and quality, it was essential to reorganize the lectures. This was done in the following way. We started with the usual preliminaries (sets, maps, elementary logic), and directly proceeded with multivariable calculus as follows: • • • •
Open, closed and compact sets, sequences and limits, continuity, differentiability.
An Intensive Course in Mathematics for Engineers
333
At this point, the derivative was introduced as a means of describing the tangent line to the graph of a real valued function in one variable. We then interpreted this ˜ − x) ˜ + f (x). ˜ tangent as a linear approximation, i. e. f (x) ≈ f (x)(x In this way, we made it clear that a special kind of generalization for the expres˜ was needed in the case of multivariate functions. This was the starting sion f (x) point for introducing linear algebra: to understand and apply the differential (functional matrix) in the n-dimensional case. Although the approach to begin analysis with the more general multidimensional case seems unusual, it has already been successfully employed for example in the undergraduate textbook by Hellwig and Wegner [HW92]. Linear algebra included: • • • • • • • •
Vector spaces and subspaces, systems of linear equations, linear mappings and determinants, inner products and norms, coordinate change, eigenvalues and eigenvectors, diagonalization, ordinary differential equations. Then we proceeded with
• • • • • •
differentiation rules, extreme values, parametrizations, line and volume integrals, theorems of Gauss, Stokes and Green, Taylor and Fourier series.
The above curriculum seems to be hard for the beginning student, but it turned out that this natural successive way of asking “What is the question” and “How can we give the solution” – especially at the point where the differential was introduced – led to interest and insight among the students. The typical topics for 1-dimensional problems – like rules for integration, intermediate value and mean value theorem – were treated especially in the tutorials and when presenting explicit calculations during the lecture. For example, the usual 1-dimensional integral is crucial while applying Fubini’s theorem. In order to achieve this tour de force, it was vital to give solid motivation for all the topics, coming mainly from the natural sciences, and to include many drawings in order to speed up the understanding.
6 Tutorials The tutorials were divided into two parts: (1) The first part included the main tutorials for two hours a week with 15 students. (The complete class was divided in order to reduce the number of students for
334
Mike Scherfner et al.
an improved instructor/student relation.) In these tutorials, the students were provided with additional explanations and examples for the topics presented in the lectures. (2) The second part included tutorials for one hour a week with five students, in which the participants solved exercises with the help of an instructor. This form of tutorials is very intensive, and there was no need for more instructors than for the usual courses. In fact, only two assistants needed to be employed as instructors for the tutorials – in general, the combined courses “Linear Algebra”, “Analysis 1 and 2” need one assistant for administrating each course, and one additional assistant for the tutorials (with 15-40 students) which are similar to the tutorials described under item (2) above.
7 Evaluation At the influential nation-wide evaluation website www.meinprof.de, 18 of the 30 students gave ratings (possible from 5.0 to 1.0, 1.0 being the best mark): Fairness Support Material Comprehensibility Fun Motivation Overall rating
1.1 1.0 1.2 1.0 1.0 1.1 1.1
The course is recommended by 100% of the students that participated in the evaluation. Student comments include: • “Definitely recommended. I did not learn this much from one year of studying mathematics than while attending this course.” • “I never had so much fun learning. Challenging, but excellent!” • “Certainly the most motivating course I have ever attended at the TU Berlin.” • “Very nice course if you are interested in math! (Recommended: Having had the advanced math course in school)” It should be mentioned that the intensive course received the best overall rating of all mathematics courses registered with this website (which are more than 1000). In comparison, the usual analysis lectures (combined in the evaluations) together with the linear algebra course, also taught by M. Scherfner, received the following average grades by 42 students:
An Intensive Course in Mathematics for Engineers
Fairness Support Material Comprehensibility Fun Motivation Overall rating
335
1.25 1.2 1.45 1.05 1.25 1.45 1.25
This shows – with the same lecturer and the same topics – that even a top evaluation can be surpassed by changing the methodology. We like to emphasize that www.meinprof.de is not optimal to make final decisions with respect to the merit of a course, but at the moment it is the only data available.
8 Conclusion and Summary Within the intensive teaching framework described above, the subject matter is presented in a way different from the standard courses, and the number of tutorial participants is held low. For each part of the course the students have to pass a final written examination. Although the tutorials are small, a higher staff requirement does not occur. The advantages of the new teaching concept showed in significantly improved pass-rates – the last course taught by M. Scherfner was successfully completed by even 100 % of the students. The intensive course program enables students with broader previous knowledge to immediately take advantage of their higher qualification. Thus, this model provides an interior differentiation that can be roughly compared with the honors class system in the U. S. Based on the results of this course, another idea was implemented: Early Bird. This course attracts the highly motivated students who want to use the free time between semesters in order to speed up their studies in mathematics for engineers. There “Analysis I” and “Linear Algebra” are presented in a condensed schedule with everyday lectures and tutorials. It turned out that the pass-rates ratios “Early Bird vs. usual courses” were: 1.33 (2006), 1.21 (2007), 1.2 (2008).
References [HW92] Karl-Eberhard Hellwig and Bernd Wegner. Mathematik und Theoretische Physik, volume 1. de Gruyter, 1992. [LL05] Frank H. Lutz and Brigitte Lutz-Westphal. Schnellkurs Ingenieurmathematik – Pilotprojekt an der TU Berlin. DMV-Mitteilungen, 3:188–191, 2005.
Moderne Studienform: Galilea und der Bachelorstudiengang „Naturwissenschaften in der Informationsgesellschaft“ Christian Schröder, Sabina Jeschke, Nicole Natho, Olivier Pfeiffer
Zusammenfassung Die durch den Bolognaprozess [oERCtAoEUC] voran getriebene Reform der Hochschullandschaft zur Schaffung eines einheitlichen europäischen Hochschulraums bedeutet für Deutschland eine tief greifende Systemänderung. Die Diplom-/Magisterstudiengänge müssen in die neue Abschlussform Bachelor/Master inklusive aller dazu gehörenden weiteren Umstellungen überführt werden. Dabei ist eine völlige Neukonzeption eines alten Studiengangs selten, obwohl dies forciert wird. Die Evaluation der alten Studiengänge und die modernen Anforderungen führen häufig zu dem Ziel, neue methodische Ansätze in die Lehre zu integrieren. [(2008, Pan08, pro07, TBLF06] Diese schließen sowohl moderne Lehr- und Lernformen als auch eine gendersensible Ausrichtung besonders der technischen Studiengänge mit ein.1 Der erste Im Rahmen des Studienreformprojektes Galilea [DJTW06] wurde ein erster Bachelorstudiengang, „Naturwissenschaften in der Informationsgesellschaft“, konzipiert, der die oben genannten Aspekte verbindet. Ein Hauptziel ist die Erhöhung des Anteils weiblicher Studierender in einem naturwissenschaftlichen Studiengang.
Schlüsselwörter Curriculum Development · Gender · Academic Education
C. Schröder (B) MuLF, TU Berlin, Straße des 17. Juni 136, 10623 Berlin, Germany E-Mail: [email protected] 1 Der erhöhte Bedarf an IngenieurInnen und NaturwissenschaftlerInnen wird von Verbänden wie z.B. dem VDI und der Tagespresse in zunehmendem Maße öffentlich geäußert. Um den erwarteten Mehrbedarf decken zu können müssen neue Strategien entwickelt werden um die Attraktivität technischer Ausbildungswege zu erhöhen. Eine große Zielgruppe bilden dabei Frauen, da sie in natur- und ingenieurwissenschaftlichen Studiengängen häufig deutlich unterrepräsentiert sind (http://www.hrk.de/de/download/dateien/HRK-Statistik_SoSe_2008_komplett.pdf).
1 Das Projekt Galilea 1.1 Der Ansatz von Galilea an der TU Berlin Trotz umfassender gesellschaftlicher Veränderungen und politischer Anstrengungen zur Schaffung der Chancengleichheit bleiben Frauen in naturwissenschaftlichen und technischen Studiengängen und den entsprechenden Berufsfeldern eine Minderheit. Unter bildungsökonomischer Perspektive und unter dem Aspekt der Qualitätssicherung und Innovationsfähigkeit hat die Forderung nach Chancengleichheit der Geschlechter gerade auf Grund der Unterrepräsentanz von Frauen in der Wissenschaft ¨ ein erhebliches Gewicht gewonnen. [Kom99, Kom01, Sch03, W98] In einer vorwiegend technologisch ausgerichteten Universität wie der TU Berlin macht sich ein geringer Frauenanteil besonders bemerkbar. Die Universitäten müssen die Möglichkeit nutzen, entsprechende Verbesserungen innerhalb von Studiengangs-Curricula zu initiieren. Ein wichtiger Aspekt ist dabei die Förderung von modernen Lehr- und Lernformen, die geschlechtsspezifische Interessen, Ziele und Ansprüche berücksichtigen. Die Galilea-Studiengänge sollen koedukative, technologisch ausgerichtete Studiengänge sein, in denen Frauen 50% der verfügbaren Studienplätze belegen. In die Lehrpläne werden umfangreiche Projekt- und Teamarbeit, Praktika in Wirtschaft und Industrie sowie internationale Austauschprogramme integriert. Sie adressieren die Schulung von Schlüsselqualifikationen wie interdisziplinäre Fähigkeiten, soziale und strategische Kompetenzen, aber auch Führungs- und Managementqualifikationen. Ein besonderer Schwerpunkt liegt auf dem Einsatz neuer Medien in der Lehre. Interaktive Lehr- und Ausbildungskomponenten wie kooperative Wissensräume [Ham03] werden in mehreren Pflichtmodulen des Studiengangs angewandt. Sie ermöglichen einen hohen Individualisierungsgrad und unterstützen so die Umsetzung eigener Lernstile. Es wird erwartet, dass sich der oft als sehr verschult und „technokratisch“ wahrgenommene Lehransatz in den techniknahen Fächern flexibilisieren lässt. So ist es nach einer erfolgreichen Einführung z.B. einer e-Learning Plattform sehr viel einfacher, alternative Inhalte zur Verfügung zu stellen und so individuelle Interessen stärker zu berücksichtigen. Das Galilea-Konzept bildet die Grundlage des Bachelorstudiengangs „Naturwissenschaften in der Informationsgesellschaft“ an der Fakultät II – „Mathematik und Naturwissenschaften“. Die Entwicklung weiterer Studiengänge an anderen Fakultäten ist derzeit noch in der Planungsphase. Durch die Entwicklung neuer Lehr- und Lernformen, eines gezielten Mentoringprogramms und die Erstellung neuer Hybridstudiengänge in den Natur- und Technikwissenschaften sollen einerseits moderne Anforderungen von Gesellschaft, Wissenschaft und Wirtschaft Eingang in die Curricula bekommen und gleichzeitig der Anteil von Frauen in diesem Bereich stark erhöht werden. Berücksichtigt wird dabei, dass es durchaus viele Frauen gibt, die sich sehr für Technik interessieren, das Image klassischer Ingenieursfächer wie Maschinenbau oder Elektrotechnik aber den Interessen einer großen Zahl von Frauen widerspricht. Durch die stärkere
Moderne Studienform
339
Integration und Betonung von interdisziplinären Fragestellungen sowie der Anwendung moderner Medien in Lehre und Lernen wird dem entgegengewirkt.2
2 Das Projekt Galilea 2.1 Die Ziele des Studiengangs Wirtschaft und Gesellschaft haben einen zunehmenden Bedarf an interdisziplinär ausgebildeten und flexibel einsetzbaren Absolventinnen und Absolventen, die über ein breites naturwissenschaftliches Fachwissen verfügen. Der interdisziplinär, anwendungs- und gleichzeitig forschungsorientiert angelegte Studiengang vermittelt Methoden und Grundlagen der Informatik, Mathematik und Naturwissenschaften. Die Zusammenführung dieser Gebiete bildet, ergänzt durch Wahlmöglichkeiten aus weiteren technischen und nicht-technischen Fächern, die Basis für die Entwicklung einer umfassenden naturwissenschaftlichen Methodenkompetenz. Durch das Bachelorstudium machen sich die Studierenden mit den fachspezifischen Methoden zur Behandlung und Lösung von Problemen der Naturwissenschaften vertraut. Dies ermöglicht im Anschluss sowohl den übergang in die berufliche Praxis, legt aber auch die Grundlage für eine weiterführende universitäre Ausbildung (Masterstudium). Die Struktur dieses Studiengangs mit seinem multidisziplinären Inhalt kommt besonders den Präferenzen von Frauen entgegen. Die enge Verbindung zwischen Theorie und Experiment wird klar herausgestellt und zusätzliche Möglichkeiten zum Experimentieren werden angeboten, z.B. mit Online remote Experimenten3 . Der Studiengang „Naturwissenschaften in der Informationsgesellschaft“ steht für die Modernisierung von Studiengängen, in denen das Verhältnis von Frauen extrem niedrig ist sowie für andere naturwissenschaftliche Studiengänge als Grundlage zur Verfügung. Für Absolventinnen und Absolventen, die nach dem Bachelor den Einstieg in die Berufstätigkeit wählen, ergeben sich verschiedene Aufgabenstellungen und Einsatzmöglichkeiten in solchen Bereichen, die ein breites naturwissenschaftliches Grundverständnis und Methodenwissen erfordern, in denen spezifische Fähigkeiten und weiterführende Kenntnisse jedoch weitgehend in der beruflichen Praxis erworben werden. Beispiele hierfür können sein: Wissenschaftsjournalismus, Tätigkeiten in Wissenschaftsverlagen, wissenschaftliches Bibliothekswesen, Referententätigkeit in Politik/ Ministerien/ Behörden im nationalen und internationalen Umfeld,
2
Sowohl in den Ingenieurwissenschaften als auch in den Naturwissenschaften werden steigende Studierenden- und Absolventenzahlen erwartet, die zu betreuen die Aufgabe der Hochschulen und die zu erreichen die Aufgabe einer hochtechnisierten Bildungsgesellschaft ist. Verfügbar unter: http://www.kmk.org/statist/fachspezprog_text.pdf [31.10.08]. 3 http://remote.physik.tu-berlin.de/farm/ [31.10.08]
340
Christian Schröder et al.
Projektmanagement in naturwissenschaftlich-technischen Gebieten, Wissenschaftsmanagement an Hochschulen und Forschungsinstituten, Tätigkeiten in Finanz- und Versicherungsunternehmen, u.a. Für einen ausführlicheren überblick über den Bachelorstudiengang „Naturwissenschaften in der Informationsgesellschaft“ siehe Kapitel [?] Anhang.
2.2 Aufbau des Studiengangs Mit dem Bachelorstudiengang „Naturwissenschaften in der Informationsgesellschaft“ bietet die Technische Universität Berlin ein Studium Generale der Naturwissenschaften an, das auch die Informatik und vor allem die Mathematik einbindet. Grundlage des Curriculums für den Studiengang bildet eine enge Verbindung zwischen Theorie und Praxis. Dieser erste entwickelte Studiengang, ist zum Wintersemester 2007/2008 mit 16 Studierenden (9 Frauen, 7 Männer) gestartet. Im Wintersemester 2008/2009 begannen 30 Studierende (14 Frauen, 16 Männer) das Studium. Die Grundlage des Studiums (Anteil von ca. 59%) wird durch die jeweils fächerübergreifende Module Mathematik für PhysikerInnen, Numerische und Computerorientierte Mathematik, sowie Experimentalphysik gebildet. Dadurch werden die theoretischen, informationstechnischen und methodischen Grundlagen vermittelt, die für jedes naturwissenschaftliche Studium benötigt werden. Zusätzlich werden die zwei neuen Pflichtmodule „Wissensmanagement in der Informationsgesellschaft“ und „Neue Medien in Forschung und Lehre“ angeboten, in denen Grundlagen des modernen Wissensmanagements und multimedialer Lehr-, Lern- und Präsentationstechniken vermittelt, bzw. angewandt werden. Diese Module werden durch die Universitätsbibliothek und das Zentrum für Multimedia in Lehre und Forschung (MuLF) durchgeführt. Beide Veranstaltungen finden im ersten Studienjahr statt und bilden neben der fachlichen Ausbildung vor allem eine methodische Grundlage für das weitere Studium. Die Anwendung von IuK-Technologien von Seiten der Lehrenden und der Lernenden steht dabei im Vordergrund. In diesen Kursen erwerben die Studierenden notwendige Schlüsselqualifikationen für ihre erwarteten zukünftigen Arbeitsbereiche. Zusammen mit dem Modul Computerorientierte Mathematik vermitteln die beiden neuen Module Fähigkeiten und Kompetenzen im Umgang mit IuKTechnologien. Die Studierenden sind anschließend befähigt, die erworbenen Kenntnisse in ihrer fachspezifische Arbeit anzuwenden. Neben den fachspezifischen Ausrichtung im Wahlpflicht- und Freien Wahlbereich bilden diese drei Module den Hauptanteil der Berufsbefähigung des Bachelorstudiengangs. Die in den Naturwissenschaften größer werdenden Anforderungen zum Umgang mit IuK-Technologien wird damit Rechnung getragen. Der Wahlpflichtbereich (ca. 21%) besteht aus einem Katalog von derzeit etwa 70 Modulen, aus den Bereichen der technischen Biologie, Chemie, Informatik, Mathematik und Physik. Die fachliche Vertiefung in einem oder mehreren dieser Bereiche steht hier im Vordergrund.
Moderne Studienform
341
Das Studium bietet darüber hinaus die Möglichkeit, im Freien Wahlbereich (10%) Module aus dem Angebot der vier Berliner Universitäten zu wählen. Dies ermöglicht einen fächerübergreifenden Austausch, stärkt aber vor allem die Möglichkeiten einer individuellen Profilbildung der StudentInnen. Der Erwerb von Sprach-, Management- und Wirtschafts- und Genderkompetenz wird nahe gelegt. Zum Studium gehört ebenfalls ein mindestens 12-wöchiges Berufspraktikum, so dass auch die AbsolventInnen eines naturwissenschaftlichen Studiums schon erste Erfahrungen in der beruflichen Praxis sammeln können. Die hohe Flexibilität dieses Studiengangs spiegelt sich auch in der Möglichkeit wider, einen individuellen Studienverlaufsplan ab dem ersten Semester zu erstellen. Es wird des weiteren angeboten, eine Bachelorarbeit in einem interdisziplinären Team zu erstellen und ein Thema aus unterschiedlicher fachspezifischer Sichtweise zu untersuchen. Dieser hohe Individualisierungsgrad wird von uns als Chance gesehen, einen höheren Frauenanteil in natur- und ingenieurwissenschaftlichen Fächern zu realisieren und mittelfristig diese Themen insgesamt für eine breitere Zielgruppe interessant zu machen.
3 Schlussfolgerung Die Herangehensweisen von Frauen und Männern an technische Fragestellungen, aber auch ihre Anforderungen an Technik unterscheiden sich mitunter stark. In der wissenschaftlichen Community ist bekannt, dass die Herangehensweisen der Frauen zur Technologie verstanden werden muss, um sie dann in pädagogischen Konzepten entsprechend zu berücksichtigen. [Col01, Sch99] In Universitäten wie die TU Berlin, die sich überwiegend auf Technologie konzentrieren, ist der niedrige Anteil weiblicher Studierender in den Ingenieurwissenschaften besonders wahrnehmbar. Gleichzeitig werden an diesen Universitäten die meisten Naturwissenschaftler, Informatiker und Ingenieure die in Deutschland leben und arbeiten, ausgebildet. Folglich spielen diese Universitäten eine wichtige Rolle in der überwindung des „Gender Gaps“ in den Ingenieur- und Naturwissenschaften. Daher sollten sich gerade die deutschen Technischen Universitäten besonders auf die Aufgabe vorbereiten, technische und techniknahe Studiengänge zu entwickeln, die gleichermaßen attraktiv für beide Geschlechter sind. Um den Anteil der weiblichen Studierenden innerhalb der technologischen Disziplinen zu erhöhen, werden im wesentlichen zwei Ansätze unterschieden: Der direkte Ansatz schlägt vor, dass änderungen durch ein aktives Herangehen an ein „Problem“ und seine Ursachen erreicht werden, während der indirekte Ansatz vorschlägt, zuerst die vorhandene Stärke der Frauen zu verbessern und dann das „Problem“ und seine Ursachen zu lösen. Im Oktober 2004 startete die TU Berlin ein Programm, das auf dem zweiten Ansatz basiert: Genesis [DJK+ 04], gefördert durch den ESF. Im Rahmen von Genesis wurde untersucht, wie z.B. Lernplattformen eingesetzt werden können, um die individuellen Vorstellungen und Wünsche von Frauen besser in der Lehre zu unterstützen. Im Zuge dieser Untersuchungen zeichnete sich ab, dass viele Studentinnen
Abb. 1 Exemplarisch ist hier ein Schwerpunkt auf die Informatik gelegt, es kann im Wahlpflichtbereich aber auch eine beliebige andere Kombination geben
342 Christian Schröder et al.
Moderne Studienform
343
vor allem von dem Image techniknaher Studiengänge abgeschreckt waren, so dass die Idee entstand, neue Studiengänge zu entwerfen. Als ein Ergebnis wurde im November 2006 das Studienreformprojekt Galilea an der TU Berlin ins Leben gerufen, mit dem Auftrag, neue modellhafte Studiengänge zu entwickeln und umzusetzen. Beide Projekte beabsichtigen dabei nicht nur, mehr Frauen für technische und naturwissenschaftliche Studiengänge zu gewinnen, sondern versuchen außerdem, die TU Berlin als Vorreiter in ihren Bemühung zu stärken, die Nachfragen und die Bedürfnisse von Gesellschaft, Wissenschaft und Wirtschaft abzudecken. Das Gesamtziel ist es, das Image von techniknahen Studiengängen zu modernisieren. Die Berücksichtigung der unterschiedlichen Bedürfnisse und Anforderungen von Männern und Frauen sowie die Anwendung moderner Lehr-/Lernformen im Zusammenhang mit IuK-Technologien werden dabei fokussiert.
4 Anhang Der Bachelorstudiengang “Naturwissenschaften in der Informationsgesellschaft” Die Regelstudienzeit des Studiengangs einschließlich der Bachelorarbeit beträgt sechs Semester. Inhalt und Aufbau des Studiums sowie das gesamte Prüfungsverfahren sind dabei so gestaltet, dass die Studierenden innerhalb dieser sechs Semester ihr Studium beenden können. Der Studienumfang beträgt 180 Leistungspunkte (LP) nach dem European Credit Transfer System (ECTS). • Pflichtbereich Mathematik In diesen Modulen bekommen die Studierenden den mathematischen Hintergrund basierend auf Analysis und Linearer Algebra vermittelt, der für alle Naturwissenschaften notwendig ist. Es wird besonders auf physikalische Fragestellungen eingegangen und die Zusammenarbeit in Kleingruppen gestärkt. • Pflichtbereich Informatik In Computerorientierter Mathematik (CoMa) wird die Syntax und Symantik von Programmiersprachen am Beispiel der objektorientierten Sprache Java erlernt. Die Teamarbeit ist dabei von zentraler Bedeutung, im zweiten Semester wird in Gruppenarbeit ein Projekt bearbeitet. In der Einführung in numerische Mathematik erwerben die Studierenden Kenntnisse zur numerischen Lösung von Problemen und Fähigkeiten im Umgang mit entsprechender Software. • Pflichtbereich Naturwissenschaften Inhalte werden in den Bereichen Mechanik, Thermodynamik, Elektrodynamik und Optik theretisch in Vorlesungen vermittelt. Hauptschwerpunkt bildet aber das Projektlabor, in dem jeweils sieben Studierenden in einer Gruppe selbständig Experimente planen, aufbauen, durchführen und auswerten. • Pflichtbereich Informationsmanagement Die Studierenden erwerben Recherche-, Präsentations-, Publikations- und Kommunikationsfähigkeiten auf wissenschaftlichem Niveau. Die Lehr-/Lernplattform Moodle wird intensiv eingesetzt.
344
Christian Schröder et al.
• Wahlpflichtbereich Die Studierenden können aus einer Liste von mehr als 70 Modulen aus Biologie, Chemie, Informatik, Mathematik and Physik wählen. • Freier Wahlbereich Studierende können Module aus dem gesamten Angebot der Berliner Universitäten wählen, auch zusätzliche Module aus der Wahlpflichtliste. • Berufspraktikum Die Studierenden bekommen einen Eindruck von möglichen späteren Arbeitsfeldern. Das Praktikum dauert mindestens 12 Wochen. • Bachelorarbeit Jede/r Studierende muss eine eigene Bachelorarbeit vorlegen. Das Thema der Bachelorarbeit sollte sich auf einen Studienschwerpunkt beziehen. Es können auch mehrere Studierende ein gemeinsamen Thema aus unterschiedlichen Fachrichtungen (z.B. physikalisch, chemisch, biologisch) bearbeiten, wobei aber jede/r eine eigenständige Arbeit anfertigen muss.
Literaturverzeichnis [(2008]
HRK Hochschulrektorenkonferenz (2008). Für eine Reform der Lehre in den Hochschulen. Technical report, Bonn, 2008. [Col01] Sabine Collmer. Wie Gender in die Technik kommt – Computerkompetenz für Frauen, 2001. http:/www.frauenakademie.de/veranst/vortrag/img/collmer.pdf, downloaded 31.10.08. Nina Dahlmann, Sabina Jeschke, Friederike Körner, Lars Oeverdieck, Rue[DJK+ 04] di Seiler, and Erhard Zorn. GENESIS – Gendersensitive Virtual Knowledge Spaces for Mathematics and Natural Sciences, Proposal European Social Fund. Technical report, 2004. [DJTW06] Nina Dahlmann, Sabina Jeschke, Christian Thomsen, and Marc Wilke. Overcoming the Gender Gap: New Concepts of Study in Technological Areas. In Proceedings of the 2006 ASEE Annual Conference, Chicago/USA, June 2006. [Ham03] Thorsten Hampel. Our Experience With Web-Based Computer-Supported Cooperative Learning – Self-Administered Virtual Knowledge Spaces in Higher Education. In Proceedings of the Site 2003 – Society for Information Technology and Teacher Education - International Conference, pages 1443–1450, Charlottesville (Va.), USA, 2003. Association for the Advancement of Computing in Education. [Kom99] Europäische Kommission. Frauen und Wissenschaft – Mobilisierung der Frauen im Interesse der europäischen Forschung. Technical Report KOM(99) 76, 1999. [Kom01] Europäische Kommission. Wissenschaftspolitik in der Europäischen Union, Förderung herausragender wissenschaftlicher Leistungen durch Gender Mainstreaming. Bericht der ETAN-Expertinnengruppe “Frauen undWissenschaft”, Brüssel, 2001. [oERCtAoEUC] Confederation of EU Rectors Conferences and the Association of European Universities (CRE). The Bologna Declaration on the European space for Education: an explanation. Verfügbar unter: http://europa.eu.int/comm/ education/policies/educ/bologna/bologna.pdf [31.10.08]. Der aktuelle Stand zur Umsetzung des Bologna-Prozesses ist unter http://www.ond.vlaanderen. be/hogeronderwijs/bologna/ einzusehen. Der aktuelle Stand zur Umsetzung
Moderne Studienform
[Pan08] [pro07] [Sch99]
[Sch03]
[TBLF06] ¨ [W98]
345
des Bologna-Prozesses ist unter http://www.ond.vlaanderen.be/hogeronderwijs/ bologna/ einzusehen. Franziska Pankow. Die Studienreform zum Erfolg machen! Erwartungen der Wirtschaft an Hochschulabsolventen. Technical report, Berlin, 2008. Projektgruppe Studierbarkeit. Studierbarkeit an der Humboldt-Universität: Wie läuft das Experiment „Studienreform“? Technical report, Berlin, 2007. Heidi Schelhowe. Interaktivität der Technologie als Herausforderung an Bildung. Zur Gender-Frage in der Informationsgesellschaft. In Forschungsinstitut Arbeit, Bildung, Partizipation (FIAB): Jahrbuch Arbeit, Bildung, Kultur, volume 17, pages 49–55. 1999. Barbara Schwarze. Wer ist wirklich drin? Gender in der Informationsgesellschaft, Analyse mehrerer Studien und darauf aufbauende Handlungsempfehlungen. Technical report, 2003. Felicitas Thiel, Irmela Blüthmann, Stefan Lepa, and Markus Ficzko. Ergebnisse der Befragung der Studierenden in den Bachelorstudiengängen an der Freien Universität Berlin. Technical report, Berlin, 2006. Berlin. Christine Wächter. Frauen in der Technik – Pionierinnen in Technopatria. In C. Wächter and et. al., editors, Technik Gestalten, Interdisziplinäre Beiträge zur Technikforschung und Technologiepolitik. Kluwer Academic Publishers, München & Wien, 1998.
Microtraining for Workplace-Related Learning Anne Carina Thelen, Sascha Daniel Herr, Frank Hees, Sabina Jeschke
Abstract Today’s working and business life is characterized by broad economic and social trends. Due to these external effects, companies face an increasing demand for up-to-date knowledge in order to stay competitive. Microtraining is one answer to increasing organizational learning needs. Its methodology is based on the concept of short learning units, supporting workplace-related learning, which is especially important in the context of small and medium-sized enterprises (SMEs). Microtraining is time and cost-saving, highly flexible and can be tailored to organizational learning demands (short, medium and long term). Keywords Work-place related learning · informal training · active learning · tacit knowledge · on-the-job learning
1 Introduction Today’s working and business life is characterized by dynamics of change and increasing uncertainty. Companies have to cope with global economic and social trends like acceleration of technological innovation, shorter production cycles, tertiarisation of jobs, knowledge-based production and sales processes, demographic change, merging of working and learning etc. The described challenges and framework conditions in the socio-economic and technological context have led to a shortened half-life period of knowledge in many cases. The span between the moment certain knowledge is needed and the time this knowledge has become obsolete gets more and more shortened. Against this background, knowledge and learning have become critical success factors for companies, which are forced to update and supplement their internal knowledge continuously. Therefore companies are looking for efficient training and A.C. Thelen (B) ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany e-mail: [email protected]
learning methods. But especially SMEs lack time and money to invest in formal training methods. Therefore informal and workplace-related learning will gain more and more relevance in the context of SMEs. Microtraining is one answer to increasing organizational learning needs and poses a systematic, holistic and structured learning approach to integrate learning to the workplace. The content of the present paper shows up the need for internal training in SMEs and introduces Microtraining as a systematic training approach which supports informal learning and knowledge exchange bound to the workplace. Furthermore the theoretical and methodological background of Microtraining is presented.
2 Training and Learning in The Context of SMEs As Fig. 1 indicates, SMEs use a variety of different measures of knowledge management and are sensitive for a knowledge-based design of their business segment in general. However, based on their heterogeneity (e.g. size, sector, structures, strategies etc.), SMEs do not show a consistent knowledge management and rather implement single components and concentrate on certain focal points [Aea03]. With regard to the topic dealt in this paper it is very important that learning out of (project-) experiences (80%) as well as internal and continuous training and learning (74%) form the basis of knowledge management in SMEs and can be assigned to the phase’s “creation and diffusion of knowledge” of the knowledge management cycle [PRR99]. In contrast to the general need for continuous learning described above, in comparison to large enterprises especially SMEs lack time and money to invest in external training. According to a current empirical study nearly half of the queried SMEs indicated that heavy workloads and time pressure prevent training and learning.
Fig. 1 Measures of knowledge management in SMEs [PGHP06]
Microtraining for Workplace-Related Learning
349
Almost 40 percent of the respondents pointed out that high training costs are responsible for the lack of learning [fB08]. Moreover most SMEs hardly possess an internal training infrastructure in comparison to large-scale enterprises and are characterized by a lack of systematization and conceptualization of training and learning. Therefore work-related, self-organized and informal models of training and learning become more and more important for SMEs [DT08] in contrast to formal training. In times of rapid economical and technological changes, traditional learning concepts become more and more insufficient as they lack the need for just-in-time and just-enough training and are not flexible enough to react prompt to rapidly changing learning needs. Only about twenty percent of relevant on-the-job knowledge is acquired through traditional and formal training, while the remaining eighty percent is gained in informal settings [Cro07], like on-the-job or workplace-related learning. Consequently, organizations look for more efficient and workplace-bounded alternatives to traditional and formal training methods. In practice however, informal learning is considered more effective than formal learning, because it is more individual and the learner becomes accountable [dVL08]. With regard to innovation, informal learning (sharing experiences, trial and error etc.) and adequate transfer and diffusion of knowledge on the work floor are keys towards innovation for SMEs. Microtraining and its informal and workplace-related training approach addresses shortages of skilled workers and improves the quality of the employees’ skills [Vea06]. It helps to maintain, distribute and share incorporated knowledge in a systematic and structured way, so that it does not become incidental. Microtraining offers rooms of face-to-face- reflection which support the workplace-related distribution of knowledge in organizations [Wah96]. Due to its tailored design Microtraining can be adjusted to specifically those problems and questions which are of internal relevance. Thereby it ensures that learning and training is aligned with the needs of individual employees as well as current and future directions of the organization [Far09]. Core elements of Microtraining are the shortness, the nearness to the workplace and the integrated reflection.
3 Methodology 3.1 The Microtraining Design Since the Microtraining approach is highly customizable and adaptable to urgent organizational issues, organizational objectives have to be clarified before setting up the training. The selection of topics will be composed of specific and workplacerelated issues, determined by the knowledge requirements of the employees and the objectives of the organization. Typical examples are the usage of new software (e.g. bookkeeping program), sales strategies (e.g. customer service), exchange of experiences about different topics; new media; commercial training; new regulations/ laws/ requirements etc.
350
Anne Carina Thelen et al.
Microtrainings are typically held with groups of 6–8 trainees in order to ensure diversity of knowledge levels, background, and experiences, as well as to engage trainees in discussions. Larger groups often show that participation is reduced due to less involvement of passive group members. Smaller groups limit the amount of pre-training knowledge held by its members, which consequently reduces the possibility of knowledge flows from one trainee to another. The role of the Microtrainer can be filled by anyone in the organization that is committed to training employees on organizational issues. In contrast to traditional formal training settings, the Microtrainer’s task is mainly to motivate and involve the trainees, facilitate and structure discussions, as well as to monitor the learning progress of the group. Microtraining consists of sessions which follow the same structure, described beneath.
3.2 The Microtraining Session Microtraining begins with an active start, followed by a demo or exercise, continues to a discussion or feedback and ends with a preview or lookout to the next session (see Fig. 2). 1. The active start raises attention and prevents the trainees from leaning back. Due to the shortness of the sessions, the trainees have to be instantly caught and engaged in a cognitive task in order to gain the most of the available time. This can be accomplished by the use of a confronting question that prompts the trainees to think about a topic in a personal, reflective manner. After giving the trainees some time to sort their thoughts, a short discussion is initiated. Hereby personal experiences with respect to the topic and the question are shared within the group. Furthermore, organizational objectives for the session are communicated and trainees are encouraged to formulate personal learning objectives. It is important to underline the relevance of the subtopic for the organization as well as the individuals.
Fig. 2 The Microtraining session [mic10]
Microtraining for Workplace-Related Learning
351
2. What comes next is a demonstration or exercise that illustrates most of the actual learning content of the session. Typical procedures used for this part are brainstorming, sequencing, role-plays, mutual interviews, fill-in questionnaires, and case studies. 3. Feedback to the exercise and discussions about the implications that arise from it are following. Experiences and insights gathered from the exercise are shared among the members and linked to the topic of the session. Trainees reflect on what they learned from the experience. At this point, the Microtrainer evaluates everyone’s understanding of the topic by asking questions on the learning content and gives feedback on the group process. 4. At the end of the session, the trainees are prompted to discuss and agree on how to retain the acquired knowledge beyond the session. Practical opportunities to apply the knowledge and behavioural adaptations that are in line with the learned content are considered. Trainees are invited to evaluate the session briefly. Moreover, a short preview of the following session is presented in order to prepare trainees on what to come and link the present topic with the upcoming themes.
3.3 The Microtraining Cycle A complete Microtraining cycle focuses on an overarching main topic and is divided into a series of 15–30 minute sessions (see Fig. 3). Every cycle starts with an introductory session, which clarifies the training concept, introduces the main topic and provides the trainees with the opportunity to get involved in the planning of the upcoming sessions. At the end of every cycle, a closure session takes place that rounds off the whole cycle, reflects on what learning achievements were accomplished and previews further cycles. In between the introductory and closure sessions, a number of subtopic sessions are employed. These sessions contain the actual learning content. The subtopic sessions break down the main topic into manageable and suitable pieces of information for the length of up to 30 minutes. Depending on the complexity of the main topic, the number of subtopic sessions can vary.
Fig. 3 The Microtraining cycle [mic10]
352
Anne Carina Thelen et al.
3.4 Evaluation of Microtrainings Evaluation of a training method is an important process for every company interested in further development. However, in practice, training evaluation is often done insufficiently. Since the Microtraining method is highly customizable, there is a lot of room for organizations to adapt it towards their needs and fit it within their context. Structured evaluation is the information source for this adaptation and continuous improvement process. Therefore, the Microtraining approach incorporates evaluations by design. There are three real-time evaluation points in one cycle. Firstly, at the end of every session, a brief evaluation of the group process, the content and the trainer is initiated. This is regarded as a snapshot of the present mood and reaction of the group. Secondly, a more extensive evaluation occurs at the closure session at the end of a cycle. Trainees are invited to give their feedback and general opinion on the cycle, the content, the Microtrainer and their personal assessment of what they learned. Thirdly, after each session and after the closure session, the Microtrainer him- or herself evaluates his or her own functioning, the appropriateness of the subtopics, as well as the assessed learning progress of the group. The real-time assessments are a part of a more formal framework for the evaluation of the Microtraining method. It is based on Kirkpatrick’s four levels of evaluation [Kir98]: Level 1: Reaction, which is the immediate reaction or feeling towards the training method. Data for this level is gathered at the real-time assessments directly after the sessions. Level 2: Learning, which describes what explicit knowledge the trainees acquired. Data for this level is gathered mainly at the closure session. Level 3: Transfer, which is the extent to which the acquired knowledge manifests itself as improved or adapted behaviour on the job. This has to be observed after a mid-term period by supervisors or the Microtrainer him- or herself. Level 4: Business results, which are the extent to which the training translated into measurable business results, e.g. increased sales or less work-related injuries. This level has to be assessed after a long-term period.
4 Theoretical Foundation 4.1 Social Constructivism Theory The core of the Microtraining concept is derived from social constructivism theory, which states that knowledge is created in social groups, due to interactions between members of a team and mutual learning from each other [Vyg78]. According to constructivism, each individual establishes his/her own understanding of the world
Microtraining for Workplace-Related Learning
353
by actively constructing knowledge and mental models. In social settings, e.g. work teams, the members of a group receive bits of information from other members in order to enrich their mental model of a given subject. As a consequence of this process, individuals synchronize their ideas, experiences and attitudes, thereby forming a small group culture that exhibits a shared mental model. Recent research suggests that shared mental models are vital for effective collaboration and coordination in work teams, as well as effective team learning [min]. Since SMEs typically exhibit a low level of hierarchy and lack extensive quality control mechanisms, effective team work and a shared understanding of organizational objectives are crucial factors for productivity and quality of work. Microtraining supports this by focusing on group work and knowledge exchange between the trainees, thereby creating an organizational learning culture.
4.2 Tacit Knowledge Since the Microtraining method is intended for the exchange and enrichment of workplace-related knowledge, the notion of tacit knowledge becomes important. Tacit knowledge is knowledge that cannot be easily transferred from one person to another by means of writing it down or verbalizing it. It is mainly acquired by experience and on-the-job learning and represents the main mode of job-relevant knowledge in work settings [NT95]. Organizations have to incorporate ways of transferring this type of knowledge because, according to recent research [Cro07], what people need to know to do their jobs well is essentially composed of tacit knowledge. Compared to larger organizations, SMEs are characterized by a lower degree of formalisation and documentation of activities and more individual control and responsibility. Furthermore, SMEs do not have access to a large applicant pool. Consequently, individuals are less exchangeable due to the tacit knowledge they hold. This poses a threat for SMEs in case of turnover or sick leave. Moreover, in order to benefit the most from this knowledge, other employees have to gain access to it. Transfer of tacit knowledge is accomplished by socialisation, which occurs in informal settings, e.g. conversations during work, and shared experiences [NT95]. Informal learning is personal and authentic, contributing to a pleasant and effective learning experience. Furthermore, socialisation is a natural way of learning, following the notions of social learning [Aro08, Ban77]. The Microtraining method incorporates these considerations by design. It offers an informal and safe environment for personal experience exchange and facilitates discussions about ideas and solutions.
4.3 Active Learning The procedure of Microtraining sessions is characterized by activities that follow the implications of active learning. The trainee is not viewed as passive receiver of
354
Anne Carina Thelen et al.
presented knowledge like in traditional training methods, but as an active participant in the learning process [BE91]. Active learning requires the trainee to be cognitively engaged in a task or discussion and to take responsibility over his/her own learning achievements [May04]. This responsibility shift towards the trainee also fosters the use of self-regulatory learning strategies, essentially improving the personal ability of trainees to acquire knowledge in a sustained and efficient manner [BW95]. Moreover, Microtraining facilitates the trainees’ metacognitive reflection on their learning progress due to the open and explorative nature of training sessions and the immediate and frequent use of feedback loops [Har01]. The Microtraining method even takes these properties of active learning further by involving the trainees in the planning and content selection process of the sessions. Since SMEs operate in a largely non-formalized manner and are dependent on individual employees to take action and responsibility, the mode of working and the mode of learning should match. The Microtraining approach accomplishes this fit by the use of active learning principles.
5 Development of the Microtraining Approach The utilization of short learning units for organizational learning is a very recent development in didactical research and commonly called Microlearning. It is grounded on the emergence of E-Learning offers. Due to the nature of web-based services, learning content became increasingly dynamic and loosely linked [Wei02]. Web 2.0 applications included the social aspect in generating knowledge together with others and found increasing attention with E-Learning companies [Dow05]. Furthermore, mobile learning that utilizes short learning units on mobile phones holds promise to expand with today’s wide availability of powerful mobile hardware [HNM05]. Since the roots of Microlearning are grounded on the E-Learning approach, Microlearning content mostly revolves around electronically delivered factual knowledge that acts as a refresher of complementary training methods. However, for the transfer of tacit knowledge, the learners need to have face-to-face contact. The proposed Microtraining approach was developed to fill this gap. It builds on the work of two EU-funded projects that were aimed at developing a framework for flexible, cost- and time-efficient training units that are able to address the learning needs of SMEs. 31 partners from 7 European states worked together between October 2004 and March 2007 in order to develop short face-to-face learning trainings that were applied in several pilot tests conducted with SMEs [mic07]. Extensive practical experience from about 50 European SMEs shaped the work of a sequel project that ran from November 2007 to February 2010 in the framework of the European Leonardo da Vinci Life Long Learning program [mic10]. Under the leadership of the Technical University of Delft, Netherlands, 5 European states collaborated in the development of supporting tools that allow SMEs to implement Microtrainings effectively. The support tools cover a broad range of information material, e.g. a quickstart guide, Microtraining manual, handbook for Microtrainers,
Microtraining for Workplace-Related Learning
355
case descriptions, evaluation tools, templates, as well as a support website (http:// www.microtraining.eu) that can be used freely by organizations to share their experiences with the Microtraining method.
6 Conclusion The Microtraining method offers an advanced and didactically sound framework for SMEs to tackle today’s need for flexible and time-efficient training. Since the method is highly customizable, SMEs can adjust and tailor their Microtrainings towards their individual requirements and resource availability. Microtraining supports workplace-related learning in SMEs and packages learning so that it is accessible and directly relevant. Due to the informal approach, employees associate the trainings with an enjoyable experience and are more likely to accept the method and participate actively than with traditional, more formal solutions. Microtraining facilitates to share ideas, practices and knowledge among employees and embraces a culture of learning and engagement in the longterm. Since the importance of human capital as a key asset is expected to grow even further in coming decades, the application of a suitable solution for building the workforce’s knowledge-base is crucial for SMEs in order to stay competitive. Understanding that new ways of learning have to be implemented can make the difference between downfall and survival. Appropriate training methods that are tailored specifically for workplace-related learning can form a competitive advantage for SMEs that are proactive and adopt them early. Overall Microtraining increases competitiveness, innovative capability, customer service and retention, output of products and services and the quality of work. At the same time it reduces error rates, wastage of materials, costs and time per task. Furthermore learning related to the workplace helps to address shifting market preferences and ensures better health and safety records as well as an engaged workforce (e.g. improved workplace culture and employees retention rates) [oC09]. Future research on the topic of Microtrainings should deal with a set of considerations. First of all, since Microtraining is a newly developed training approach, the long-term effects of the method have to be evaluated properly. How does it affect productivity? Work satisfaction? Learning capacity? Etc. Second, it would be interesting to see how Microtraining performs when applied to larger organizations and how it would have to be adapted to meet the requirements that come with such a target group. Third, a large scale comparative study that measures how the Microtraining method fares in competition with other training methods on similar areas could reveal insights: In which settings can Microtraining yield the most benefits? Which conditions may render Microtraining not suitable? The success of the method is dependent on its applicability, acceptance and effectiveness. Empirical data originating from the considerations above will contribute to a more complete picture of what role Microtraining can play in the future of organizational learning.
356
Anne Carina Thelen et al.
References [Aea03]
E. Abele and et al. Wissensmanagement in KMU. Entwicklung eines Werkzeugkastens für kleine und mittelständische Unternehmen. Zeitschrift für wirtschaftlichen Fabrikbetrieb, pages 375–379, 2003. [Aro08] E. Aronson. The social animal. Worth/Freeman, New York, 2008. 10th ed. [Ban77] A. Bandura. Social Learning Theory. Prentice Hall, New Jersey, 1977. [BE91] C. Bonwell and J. Eison. Active Learning: Creating Excitement in the Classroom AEHE-ERIC Higher Education Report No.1v. Jossey-Bass, Washington, D.C., 1991. [BW95] D. Butler and P.H Winne. Feedback and self-regulated learning: A theoretical synthesis. Review of Educational Research, 65:245–281, 1995. [Cro07] J. Cross. Informal Learning: Rediscovering the Natural Pathways that Inspire Innovation and Performance. Pfeiffer, San Francisco, 2007. [Dow05] S. Downes. E-learning 2.0. eLearn Magazine. www.elearnmag.org/subpage.cfm? section=articles&article=29-1, downloaded 2010-05-17, 2005. [DT08] O. Döring and S. Turnwald. Personalentwicklung in kleinen und mittleren Unternehmen: Anforderungen, Möglichkeiten, Grenzen und Perspektiven. Retrieved. www. f-bb.de/uploads/tx_fffbb/Fachartikel_PE_in_KMU_Doering.pdf, downloaded 201004-20, 2008. [dVL08] P. de Vries and T. Leege. Final Report WP 1:Bedarfsanalyse. Reload project. DE/07/ LLP-LdV/TOI/147058. Leonardo Project: European Union. Technical report, Leonardo Project: European Union., 2008. [Far09] N. Farvaque. Guide for Training in SMEs. DG Employment, Social Affairs and Equal Opportunities. European Commission. www.ec.europa.eu/social/main.jsp?langId=en &catId=89&newsId=544&furtherNews, downloaded 2010-05-10, 2009. [fB08] Bundesinstitut für Berufsbildung. Weiterbildungsbeteiligung in KMU. www.kibb.de/ cps/uploads/559_Weiterbildung_in_KMU_Antwort1.1219152837800.pdf, downloaded 2010-04-20, 2008. [Har01] H. J. Hartman. Metacognition in Learning and Instruction: Theory, Research and Practice. Kluwer Academic Publishers, Dordrecht, 2001. [HNM05] A. Holzinger, A. Nischelwitzer, and M. Meisenberg. Mobile Phones as a Challenge for m-Learning: Examples for Mobile Interactive Learning Objects (MILOs). In 3rd International Conference on Pervasive Computing and Communication (IEEE), pages 307–311, 2005. [Kir98] D.L. Kirkpatrick. Another look at evaluating training programs. American Society for Training & Development, Alexandria, VA:, 1998. [May04] R. Mayer. Should there be a three-strikes rule against pure discovery learning? The case for guided methods of instruction. American Psychologist, 59:14–19, 2004. [mic07] Microteaching - modular teaching and learning solutions for a needs based education. www.microteaching.org, downloaded 2010-05-10, 2007. [mic10] Microtraining for effective learning. www.microtraining.eu, downloaded 2010-05-10, 2010. [min] Shared mental models, team coordination, and team performance. In Symposium conducted at the annual meeting of the Society for Industrial and Organizational Psychology, Orlando, FL. [NT95] I. Nonaka and H. Takeuchi. The knowledge creating company: how Japanese companies create the dynamics of innovation. Oxford University Press, New York, 1995. [oC09] Conference Board of Canada. Workplace Learning in Small and Medium-sized Enterprises: Effective Practices for Improving Productivity and Competitiveness. Overview Report. 2009. http://www.ccl-cca.ca/pdfs/OtherReports/CBofC-WorkplaceLearningSME-OverviewReport.pdf, downloaded 2010-04-28. [PGHP06] Peter Pawlowsky, Lutz Gerlach, Stefan Hauptmann, and Annett Puggel. Wissen als Wettbewerbsvorteil in kleinen und mittelständischen Unternehmen. Empirische
Microtraining for Workplace-Related Learning
[PRR99] [Vea06] [Vyg78] [Wah96] [Wei02]
357
Typologisierungen auf Grundlage einer bundesweiten Befragung. FOKUS prints 09/ 06. Technical Report 09/06, Technische Universität Chemnitz, Chemnitz, 2006. http:// www.tu-chemnitz.de/wirtschaft/bwl6/prints/fokus_prints_09-06.pdf, downloaded 2010-04-20. G. Probst, S. Raub, and K. Romhardt. Wissen managen. Wie Unternehmen ihre wertvollste Ressource optimal nutzen. Gabler, Wiesbaden, 1999. R. Verburg and et al. Managing Technology and Innovation. An introduction. Routledge, Oxford, 2006. L. Vygotsky. Mind in Society. Harvard University Press, London, 1978. H.K. Wahren. Das lernende Unternehmen: Theorie und Praxis des organizationalen Lernens. Gruyter, Berlin/New York, 1996. D. Weinberger. Small pieces loosely joined: A unified theory of the web. Basic Books, New York, 2002.
Teachers need robotics-training, too Ursula Vollmer, Sabina Jeschke, Barbara Burr, Lars Knipping, Jörg Scheurich, Marc Wilke
Abstract Since robotics holds a special fascination among all the technological fields – due to its cross-disciplinary approach as well as its popularity through movies and literature – this field is attracting even the interest of pupils and students who are normally more reserved and cautious towards technological areas. In particular, practical experiences have shown that out of all the different technological areas robotics has a very high motivation impact on women. Thus, robotics can also be utilized as an instrument to bridge the gender gap in technological areas and fields of natural sciences. However, so far the majority of projects and initiatives focus on the instruction of the pupils and students themselves which naturally leads to restrictions concerning the scope of application. In order to enhance the impact factor, the integration of “multipliers”, in particular the “educators” in elementary schools, middle schools und high schools - and even in the kindergarten - is an important task. Within this paper, we describe a concept for teaching teachers how to implement robotics curriculum into today’s classrooms which is currently under development at the University of Stuttgart and the Technische Universität Berlin. Keywords robotics · teachers · hands-on training
1 Introduction Within a large number of projects all over the world, robotics has proven to be an extraordinary powerful tool to excite pupils and students about science and technology. Taught in a gender-sensitive way, robotics is also attracting women and girls, that usually are not so much interested in natural sciences or technological issues. As there is – or soon will be – a shortage on engineers in Germany, we need U. Vollmer (B) Institute of Information Technology Services, University of Stuttgart, Allmandring 30A, 70569 Stuttgart, Germany e-mail: [email protected]
to get more young people interested in these fields, so that the number of students choosing engineering or natural sciences as their studies increase. Looking at the percentage of women in natural sciences and engineering, there lies much potential in this social group for achieving a higher number of students in the respective fields. At the moment, robotics does not yet have the necessary significance in education that corresponds to the possibilities it offers. In order to improve this situation, the RoboTeach program aims at teaching, informing, and supporting teachers and educators, that would like to introduce robotics courses and projects in their institutions.
2 Concept and Pedagogical Approach The RoboTeach program aims at improving children’s education in robotics by offering robotics trainings, information and material to educators in kindergarten and teachers as well as teacher trainees for all different school levels. As the situation of the target audiences, such as time, place, and other needs, has to be accounted for, the program consists of four different modules that complement one another. The situation of the children, e. g. age, social background, are accounted for in all of those modules.
2.1 RoboTeach-AttendanceCourses The best way to teach robotics includes hands-on training and so has to be done in attendance courses. We plan to offer courses for • teacher trainees • teachers of all different school levels • educators in kindergarten The courses for teacher trainees are integrated into the Robinson program (see section 3.1, [JKVW08]) established at the University of Stuttgart and the TU Berlin as a new module. A specific robotics program is offered for teacher trainees during the semester. This program consists of a seminar, in which – after a few introductory lectures – students are supposed to work out a robotics topic and to prepare a lesson or hands-on training on this topic for to their fellow students. The overall topics are given by the lecturer, but the exact subtopic can be chosen and defined by the students. This gives all students attending this course the possibility to learn more about robotics itself on the one side and to practice teaching robotics on the other side. In combination to this seminar, a hands-on training is offered, where the students are supposed to design, built and program a robot in small teams. For this training, the LEGO Mindstorms NXT robotics kits [FFA07] and the NXT-G [Kel07] graphical programming language are used. This allows an easy access to such a robotics project, as the robotics kits and the graphical programming language are designed for children and are so easy to get acquainted with. Nevertheless, very interesting and challenging projects can be implemented with these tools. The introductory
Teachers need robotics-training, too
361
lessons provide an overview on the robotic field. They also give information on robotics kits, course possibilities for the different ages of the school children, and gender issues. Possible competitions for children and preparation information are also presented in the introductory lessons. As teachers and educators usually do not have much time during the week due to their jobs, different possibilities for courses have to be offered for them. Thus, we will offer weekly courses as well as compact courses in one week during the school holidays. These courses will begin with a few introductory lectures, followed by a hands-on training. The introductory lessons will contain a short introduction in robotics and information on robotics kits, robotics competitions for children, and gender issues. During the hands-on training the teachers and educators have the possibility to become acquainted with the robotics kits and get ideas for possible exercises and projects they can do with the children at school or in kindergarten, respectively. For those teachers and educators, who first want to get an idea on how robotics could be used for educational purposes, compact courses of one or two days with a short introduction in robotics and some hands-on exercises are offered, too. As for the students’ hands-on training, the LEGO Mindstorms NXT robotics kit [FFA07] in combination with the NXT-G graphical programming language [Kel07] is used for these courses, too.
2.2 RoboTeach-DistanceLearning Additionally to the RoboTeach-AttendanceCourses, distance learning possibilities are offered to deepen the course contents (blended learning), to provide means for practice and experience exchange, and to provide courses for people, that are not able to participate in attendance courses. All distance learning means provide information on technical issues of robotics, on specific robotics kits (e. g. LEGO Mindstorms [FFA07], Fischertechnik [fis], Asuro [asu], . . . ), on competitions, on the suitability of robotics kits for specific groups of children, and on gender issues. Means used for distance learning include virtual laboratories, e-learning platforms and forums for experience exchange.
2.3 RoboTeach-OnDemand Teacher trainees, teachers and educators will definitely have specific demands that can not be covered with any of the other course offerings. Those demands could refer to different robotics kits (LEGO WeDo, Fischertechnik, Asuro, c’t bot . . . ), specific target groups of the children (such as disabled children, children with difficult social background, . . . ), combination with specific subjects, . . . The specific demands of an interested group are determined and a special course is then planned and prepared for this group. In case single persons ask for such special demands, a course is planned on the premise, that more participants are found for this special course.
362
U. Vollmer et al.
2.4 RoboTeach-Material In order to prepare good lessons, courses, and activities for the children, teachers and educators need to have access to good teaching material. We develop material for teaching robotics under the following aspects: • • • • • • • •
age of children target group (e. g.: girls, boys, children with difficult social background, . . . ) focus (e. g.: exercises adapted to a specific topic or (school) subject) intended use (support during lesson, learning at home, . . . ) form (book, video, game, building instructions, . . . ) content (robotics basics, specific robotics kit, gender, . . . ) information on different hardware / robotics kits available competitions
3 Embedding The RoboTeach program is embedded in the Robinson program, established at the University of Stuttgart and the TU Berlin and cooperates with the Roberta® project at the IAIS Fraunhofer Institute in St. Augustin via the RobertaRegioZentrum in Stuttgart and Berlin, respectively.
3.1 Embedding in Robinson Program The Robinson program at the University of Stuttgart and the TU Berlin consists of the following modules, all offering robotics lectures, seminars, and hands-on training to students and – in the case of Roberta® - pupils: • Robinson-Ing: Courses in this module aim at students of natural sciences and engineering. They provide the students with an insight into the working life of an engineer and introduce basic concepts of hard- and software engineering. Additionally, soft skills like team work and presentation skills are trained in these courses. • Robinson-Mixed: This module aims at students of humanities and social sciences. The overall goal is, to bridge the gap between natural sciences and engineering on the one hand and humanities and social sciences on the other. The students get an insight in the way of working and thinking of an engineer, as well as an idea about the social relevance of such a field and about the creativity, that is needed in engineering. • Robinson-Med: Besides the target group, i. e. medical engineering students, and the necessary adaptions to this target group, this module is very similar to the Robinson-Ing module • Roberta®: The Roberta® module provides the framework for the RobertaRegioZentrum and the Roberta® courses, respectively.
Teachers need robotics-training, too
363
The courses for teacher trainees, described in section 2.1, are integrated as a new module Robinson-Ped of the Robinson program.
3.2 Cooperation with Roberta® project The Roberta® project [fAiSA06] was initiated by the IAIS Fraunhofer Institute in St. Augustin. The main goal of this project is, to increase the number of women in the natural sciences and engineering field by offering gender-sensitive robotics courses for pupils. The IAIS Fraunhofer Institute in St. Augustin constitutes the basis for many so called RobertaRegioZentrum throughout Germany and some other European countries. Course instructors are prepared for their work in specific courses at the IAIS Fraunhofer Institute in St. Augustin. The RoboTeach project extends the coverage of the RobertaRegioZentrum in Stuttgart and Berlin, respectively by providing professional instructors, such as teachers and educators, with the information necessary for establishing robotics courses for pupils by themselves.
4 Similar Projects Some projects with similar aims are described in the following sections.
4.1 Course for Junior High School Teachers in Japan Takahashi et al. [TKM+ 07] propose a one day training course for Japanese teachers. The course consists of an one-hour lecture, followed by a four-hour hands-on training. The lecture, based on a textbook, introduces robotics and the fundamental issues concerning the master slave robot system used for the hands-on training.
4.2 Material for Hands-on Robotics Training Mataric et al. [MKFS07] provide teachers with materials for hands-on robotics training and STEM education.
4.3 Robotics as Integrating Agent in Primary Education Martin et al. [MBG00] describe a project, in which Irish teachers were introduced to robotics in order to enable them using technology as an integrating agent in their work. In this project, story-telling and robotics were combined in courses at primary schools.
364
U. Vollmer et al.
4.4 Evolution of Robotics Kits Martin et al. [MMR+ 00] describe the technological evolution of robotics kits, beginning with Seymour Papert’s Logo work [SP76]. This summary is followed by an introduction in possible activities with the presented robotics kits.
5 Outlook The RoboTeach program will improve and spread education in robotics in kindergarten and schools, as the instructors are better prepared and know more about the topic. Besides the technical aspects, instructors also know more about gender issues. This knowledge is not only applicable for the robotics courses and projects they plan, but also for their normal classes. Hopefully, this will improve the way mathematics and natural sciences are taught in schools and get more pupils interested in these subjects. If this succeeds, the number of students – especially women – choosing natural sciences or engineering as their studies will increase and we can possibly meet the needs of engineers that will arise shortly.
References [asu] [fAiSA06] [FFA07] [fis] [JKVW08]
[Kel07] [MBG00]
[MKFS07] [MMR+ 00]
[SP76] [TKM+ 07]
AsuroWiki - Home Page. http://www.asurowiki.de/pmwiki/pmwiki.php. St. Augustin Fraunhofer-Institut für Autonome intelligente Systeme AIS. Roberta Grundlagen und Experimente, volume 1. IRB Verlag, 2006. Mario Ferrari, Guilio Ferrari, and David Astolfo. Building Robots with Lego Mindstorms Nxt. Syngress Media, April 2007. fischertechnik - Building blocks for life. http://www.fischertechnik.de/en/index.aspx. Sabina Jeschke, Lars Knipping, Ursula Vollmer, and Marc Wilke. The Robinson program: Robotic Curricula for Interdisciplinary Academic Education. In Proceedings of the SEFI Annual Conference, Aalborg/Denmark, July 2008. James Floyd Kelly. LEGO MINDSTORMS NXT-G Programming Guide. Computer Bookshops, July 2007. Fred G. Martin, Deirdre Butler, and Wanda M. Gleason. Design, story-telling, and robots in Irish primary education. In Proceedings of the IEEE International Conference on Systems, Man, and Cybernetics, 2000, volume 1, pages 730–735, 2000. Maja J Mataric, Nathan Koenig, and David Feil-Seifer. Materials for enabling hands-on robotics and STEM education. In In AAAI Spring Symposium on Robots and Robot Venues: Resources for AI Education, 2007. Fred Martin, Bakhtiar Mikhak, Mitchel Resnick, Brian Silverman, and Robbie Berg. To mindstorms and beyond: Evolution of a construction kit for magical machines. Interactive Technologies, pages 9–33, 2000. Morgan Kaufmann Publishers Inc. San Francisco, CA, USA. Cynthia J. Solomon and Seymour Papert. A case study of a young child doing turtle graphics in LOGO. In Proceedings of the June 7–10, 1976, national computer conference and exposition, pages 1049–1056, New York, USA, 1976. ACM. Y. Takahashi, N. Kanai, M. Miwa, T. Yoshidom, N. Kimura, K. Shigeri, I. Ikari, and Y. Kawarads. Proposal of Robotics Education Training Course for Junior High School Teachers. In Proceedings of the International Workshop on Robotics in Education, 2007, 2007.
RELOAD - A Semantic-based Learning and Knowledge Platform for Employees of the Do-It-Yourself Industry Florian Welter, Olivier Pfeiffer, Anja Richert, Sabina Jeschke
Abstract The focus of the project RELOAD is set on employees of the Do-ItYourself (DIY) industry. Employees and consultants in this sector play a decisive role because they communicate directly with end customers during sales and consulting talks. Nevertheless, it is a fact that many employees and consultants in DIY stores are untrained, low qualified workers, or even workers from other sectors. Moreover, RELOAD considers customers with regard to means of informing them on products. Concerning commercial success, it is important for the DIY sectors that end customers know how to use and apply products which are purchasable in DIY stores. Only if this know-how can be transferred, the end customers will buy the respective products. In other words the maxim for this sector is true that ‘the knowing customer buys more’. RELOAD tries to address employees and costumers at the same time by offering a Knowledge Platform as a multimedia and semanticbased solution. This platform contains eLearning modules which are specified to the individual learning needs of the employees and which should enable them to actualise their knowledge much more efficient and faster. Together with the knowledge platform in DIY stores, eLearning applications in form of short learning modules will support the self-directed learning of the employees. This kind of learning can be integrated into dynamic daily work processes more easily than classical types of learning and at the same time it is also more cost-efficient. As a consequence customers of DIY stores will gain more profit through a more efficient consultancy, too. Keywords Do-It-Yourself industry · eLearning · Microtraining · Semantic-based Knowledge Platform
1 Introduction With an approximate annual business volume of 37 Billions Euro in Europe the DIY branch is an important economic factor in the European Union. For this reason RELOAD focuses on the vocational training of the employees of the Do-It-Yourself (DIY) stores and their consulting skills. This plays a particularly important role in this sector as they communicate directly with the end customers in sale consultations. But as a matter of fact many employees and consultants in the DIY stores are un-trained low qualified or workers from outside the sector. To be commercially more successful in this sector, the employees have to inform the customers about how to use and apply products more efficiently on the one hand. On the other hand the DIY industry has to deal with an augmenting variety of product information which needs to be updated constantly. Concerning this point, preparing the product information in a simple and clear manner so that employees of DIY stores are able to use it for sales, is not a trivial task, because employees have to be trained e.g. in electronic devices or different material properties. Besides this fact there is a growing number of costumers which possess an comprehensive previous knowledge mostly achieved through the internet. The knowledge acquisition behaviour of the customers is a consequence of the so called “EduCommerce” [Lee07] which is a hybrid form of eLearning and eCommerce. Therefore it is essential for the DIY stores to train their employees at least as good as the customers to ensure a good quality of consulting services. In other words for the DIY branch the maxim is true that ‘the knowing customer buys more’.
2 Project Objectives For the challenges described above, RELOAD establishes a Knowledge Platform as a multimedia and semantic-based solution in DIY stores, which contains eLearning modules to enable the employees to learn ‘on the job’ on a self-directed and efficient way [BRH07]. The project primary aims at untrained and less qualified employees in the DIY-sector, who are not used to self-directed and media-based learning. A special didactical approach is required in RELOAD to motivate this target group. To ensure an effective learning process for the target group, a blended learning concept was chosen which supports self-learning processes accompanied by experienced colleagues while it takes place as collaborative learning in the group. The interaction with the Knowledge Platform, the Microtrainings which constitute a didactical concept characterised by short learning units and the support of experienced colleagues or trainers assure the motivation of the learners [BH07a, vdMSHL95]. These Microtrainings can be integrated into daily working processes more easily than common types of learning e.g. the teacher-centred learning [BH07b]. Additionally, the platform allows tests of the formerly learned units which support the control of the learning process for the target group. Furthermore, RELOAD focuses on supporting companies on their way to become a learning organisation and increase the service capabilities of employees.
RELOAD
367
Therefore, the long term survival ability of the participating enterprises will be increased. The application of the Knowledge Platform as well as the participation in and the creation of the Microtrainings implicates that the workplace of the future is differently formed and better adapted to the new challenges of lifelong learning processes. Nevertheless, these processes are necessary to cope with growing product diversity and an augmenting flexibility of services. RELOAD is coordinated by the Center for Learning and Knowledge Management and Institute of Information Management in Mechanical Engineering, RWTH Aachen University. The European DIY-Retail Association (EDRA), producers and DIY stores as well as scientific partners from the education and IT sector in Europe are part of the international consortium.
3 Methodology and Technology Used The service website “www.k-21.net/baumarkt” provides a basic technology for the RELOAD Knowledge Platform. A centralised data pool with course and learning units (e.g. ‘How to use a grinder correctly?’) concerning various DIY products has been created in close cooperation with the academy “Bauen & Wohnen” (build & live), which is a federation of German DIY branch experts. Due to the fact that RELOAD is an innovation transfer project applying existing technologies and concepts it does not aim to ‘reinvent the wheel’. Hence, the latter mentioned basic technology is used as a base and will be developed continually to an integrated learning and knowledge management environment. Besides, during the process of system development mainly open source learning architectures are used. In addition to that, the integration of Microtrainings as multimedia-based short learning units which were conceptualised in another European project will supplement the Learning and Knowledge Platform [VB08]. Especially the innovation of using blended concepts and their transfer to the DIY branch is one of the unique features of RELOAD. Microtraining is distinguished as an innovative concept which helps employees of the DIY-branch to cope with the increasing information of the manufacturers faster and to incorporate this information into the consultation of customers [BRH09]. The Microtraining learning units are integrated into this platform and are characterised by high applicability and by the attempt to reduce the units to a minimum regarding time-consumption. The Knowledge Platform bases on a semantic net to offer flexible and nonlinear ways of learning to the users (cf. Fig. 1). The contents of the platform are logically connected and they are related as instances of classes and objects to each other, which can be described as general functions of a knowledge map [SFM+ 07, SBH07]. Furthermore, the content of the platform can be easily accessed either via an index or via a search function enabling a clear and simple navigation through the system for the target group [RM95, Bro98]. By providing these functions, the semantic-based Knowledge Platform offers information for which employees of the DIY branch searched for as well as information for which users
368
Florian Welter et al.
Fig. 1 Example of a semantic net (www.w3.org)
did not search for, but which is also important, because it is complementary information. In other words the architecture of the system arouses the interest among users to navigate through the platform and thus to receive information that educates their branch specific skills. Additionally, the semantic-based platform contains vital technical information about various products and electronic tools, e.g. a drilling machine or a grinder. This information is visualised via the Microtrainings which are embedded on the platform as animations, 3D models and video- or audio sequences [SBH08] (cf. Fig. 2).
Fig. 2 3D model for using a grinder as an example of a learning unit (www.baumarktwissen.eu)
RELOAD
369
A typical Microtraining unit is characterised by duration of approximately 15 minutes. The relatively short time of 15 minutes is important with regard to the target group consisting of mainly untrained and less qualified employees, from which at least some would lose interest or would be overstrained by longer learning units. Thus, each unit begins with a general multimedia introduction into the topic, e.g. a video about technical details and handling of a grinder. The theoretical introduction is followed by a phase of practice for the users which contains questions to different items and product features. In the final phase of the Microtraining unit, a short summary of the formerly learned content is provided underlining the main learning achievements for the users. After the employees have learned with the help of the Microtraining units (today over 50 different units are available on the Knowledge Platform) they have to pass a web-based test which verifies their knowledge. If the test-phases are finished successfully, a certificate can be obtained in a final demonstration confirming the Microtraining graduation. An important aspect concerning the development of the Knowledge Platform depicts the consideration of user feedback in the entire development process. Besides the fact that feedback is used to improve the system continually, the users shall be enabled to create and edit Microtraining units personally in future. Hence, by supplying possibilities of interaction on the platform, the interest of employees to visit and learn new units is augmented. Furthermore, with regard to an ongoing demographical change towards an older society with e.g. older employees, the need to share the knowledge of experienced employees is increasing, too. Again a solution can be provided by the RELOAD Knowledge Platform, because it enables the integration of the employees’ knowledge into an interactive learning environment.
4 Results And Concluding Remarks A self-directed and media-supported learning is provided to the employees of the DIY branch by the didactics specifically designed for the target group. This kind of semantically supported learning, which also combines education with entertainment (‘Edutainment’), enables the employees to handle the customers’ rising need of information. Due to this reason the learning approach is very promising with regard to increasing overall sales in the DIY Industry. Thus, a better trained employee of a DIY store supports the fact that ‘the knowing customer buys more’. With the accomplishment of a detailed analysis of needs as well as customer strategy workshops for the employees of the DIY-markets, a didactical concept was conceived for the target group. Based on these results, two test phases started in the participating DIY stores (in August 2008 and in September 2009), in which became obvious that the target group handled the provided applications on the Knowledge Platform in a satisfying manner, although general deficits of the employees in the DIY branch became obvious concerning the general use of computers and the familiarity with internet applications. The latter could be underlined by user interviews which have been conducted after the test phases. As a consequence providing
370
Florian Welter et al.
personal trainers (in combination with an individual training schedule) could represent a solution that should be intensified for those employees who are inexperienced in using computer applications. Nevertheless, the platform received a positive feedback with regard to the usability of its Microtraining units. Referring to this, one has to state that the use of the Microtraining units of the Knowledge Platform seems to be more effective for the employees during their working time (‘learning on the job’) in comparison to complete learning units at home. On the one hand the results underlined that this kind of informal learning is a promising approach for the employees of the DIY branch, but on the other hand the test phases also elucidated the need to continually develop an adequate framework in the entire DIY Industry, including more learning terminals for employees at work. Concerning this, the augmenting efficiency and use of Microtrainings in the branch has to be promoted to convey more partners among the DIY Industry. Hence, an important argument for a further promotion can be described with the fact that didactically prepared Microtrainings can be implemented in a cost-saving manner, because already existing materials and technical descriptions could be simply provided by the DIY Industry. In addition to that, a broader international scope could be realised, too, e.g. by expanding the platform with additional modules in several languages for a European context. In doing so, more international DIY branch experts and customers could be reached. Other important steps towards a development of the RELOAD results are thinkable by a closer integration of suppliers of the DIY branch in the existing Knowledge Platform or a more intensive cooperation with agencies, publishing companies or training centres to disseminate and to exploit the gained results. With regard to the development of a holistic ‘value chain of knowledge’ in the DIY branch, a higher degree of standardisation still seems to be necessary, to be able to integrate already existing eLearning solutions of the entire DIY branch into a value chain. Moreover, concerning the task of exploitation a business model has been developed which envisages that the further development of the Knowledge Platform can be continued in future, apart from its initial public funding.
References [BH07a]
[BH07b]
[BRH07] [BRH09]
S. Brall and F. Hees. Effektives Lernen mit Kurzlerneinheiten. Kompetenzentwicklung in realen und virtuellen Arbeitssystemen. GfA, page 209–214, 2007. Dortmund, Germany. S. Brall and F. Hees. Microtraining: Activating knowledge transfer in businesses. In G. Papadourakis and I. Lazaridis, editors, New Horizons in Industry, Business and Education, page 447–451. Heraclion, Greece, 2007. S. Brall, A. Richert, and F. Hees. Self-directed Learning, Knowledge modules for effective learning. ZLW/IMA, Aachen, 2007. S. Brall, A. Richert, and F. Hees. Wissensaktualisierung durch Kurzlerneinheiten. In K. Henning and C. Michulitz, editors, Unternehmenskybernetik 2020: Betriebswirtschaftliche und technische Aspekte von Geschäftsprozessen, page 339–344. Duncker & Humboldt, Berlin, 2009.
RELOAD [Bro98] [Lee07] [RM95] [SBH07]
[SBH08]
[SFM+ 07]
[VB08]
[vdMSHL95]
371 T. Brooks. The semantic distance model of relevance assessment. In Information Access in the Global Information Economy, volume 35, page 33–44, 1998. T. Leege. EduCommerce–Informierte Kunden kaufen mehr. In BHB Info Baumarktkongress, page 24–25, 2007. P. Resnik and C. S. Mellish. Using information content to evaluate semantic similarity in a taxonomy. In IJCAI-95, page 448–453, 1995. S. Sattari, W. Backhaus, and K. Henning. The Web-Knowledge Map in Higher Education: A semantic-based tool to answering the challenge of technology use, succession of generations and changing learning patterns. In Proceedings of the 8th International Conference on Information Technology Based Higher Education and Training-ITHET, Kumamoto, Japan, 2007. S. Schröder, W. Backhaus, and K. Henning. RELOAD: A Knowledge Platform with a Blended Learning Concept for Employees in the Do-It-Yourself (DIY) Industry. In Association for the Advancement of Computing in Education (AACE), editor, Proceedings of the World Conference on Educational Multimedia, Hypermedia & Telecommunications (ED-MEDIA), page 2722–2725, Vienna, Austria, July 2008. E. Sjoer, P. Fabian, E. McQuade, J.C. Nascimento, P. Pimenta, and S. Sattari. Implementation of a web knowledge map from. In Joining forces in Engineering Education towards Excellence Proceedings SEFI-IGIP Joint Annual Conference, Miskolc, Hungary, July 2007. P. De Vries and S. Brall. Microtraining as a Support Mechanism for Informal Learning. www.elearningpapers.eu/index.php?page=doc&doc_id=12788&doclng=6, downloaded 2010-03-16, 2008. H. T. van der Molen, G. N. Smit, M. A. Hommes, and G. Lang. Two Decades of Cumulative Microtraining in The Netherlands: An Overview. Educational Research and Evaluation, 4(1):347–378, 1995.
Pre-Freshmen Students Gearing up with Early Bird Erhard Zorn, Sabina Jeschke, Akiko Kato, Olivier Pfeiffer
Abstract We are offering a freshmen course called “Early Bird” where students have the opportunity to take the mathematics courses of the first semester (Calculus I for Engineers and Linear Algebra for Engineers) before they are enrolled at our university. Participants accomplishing sufficiently many homework assignments are qualified to take the final written examinations even if they are (still) not enrolled. The grades of these examinations may be accepted if the students will be enrolled afterwards. In this 9 weeks course the regular calculus I and linear algebra lectures are taught in the same lecture/tutorials together. Though the workload in this course is very high for students, 99 % are recommending this course to other prospective engineering students. The intention of this course is to provide the first semester students with the mathematics that will be usually used in non-math classes before it can be taught in the math classes. As mathematical knowledge and skills are some of the most important tools for engineers the Early Bird course has proved as very effective to prepare engineering students for their engineering courses. Before winter term 2008/09 we successfully offered this course for the third time. This year we had no additional financial resources to offer very small classes. On the other hand, in the week between this course and the final examinations a summer camp has been organized where recitation lessons were voluntarily offered by teaching assistants. In this article, we are comparing the final examination results of Early Bird students and regular students. We are presenting the results of the Early Bird courses of the last three years. The results will be compared with the data of regular students who took the same written exams.
1 Key Features of Early Bird At the Berlin Institute of Technology we offer an intensive course in mathematics called “Early Bird” during summer holidays. Within this course the content of the regular courses “Calculus I for Engineers” and “Linear Algebra for Engineers” are taught together during nine weeks. Our target group is the students of engineering departments who have to attend these two courses in their first (or first and second) semester at university. They are allowed to attend Early Bird in advance in summer directly before their enrollment in autumn. If they successfully attend the course and pass the final exam, the earned credits can be applied to their engineering studies. We offered this course in the last three summers in 2006–2008 teaching about 300–400 students per course.
2 Motivation In the freshman courses in engineering or natural sciences we observe all too often that engineering students do not have sufficient mathematical skills to understand the subject of the classes. However, sound standing knowledge in mathematics is a basic tool for every engineer. There are at least two possible explanations for this phenomenon. It is possible that the students have poor knowledge of high school mathematics. Surely it is not the task of a university to make up for the failures in high school mathematics lessons. Nevertheless, our school of mathematics and natural sciences, as well as many other universities, offer remedial courses in high school mathematics, so called bridge courses, to prepare high school graduates for their studies. The other reason may be found in the mismatching curricula of the subjects of the studies. The understanding of non-mathematical subjects often requires knowledge in mathematics which has not yet been taught in the mathematics classes. At the Berlin Institute of Technology, an attempt has been made in the past by the lecturers to adjust the curricula of different classes. Although many improvements have been achieved regarding the choices of the relevant topics and their chronological order, there are still problems left which are difficult to resolve. As an example the freshman course “Introduction to Classical Physics for Engineers” starts with basics of classical mechanics, including Newton’s laws as a matter of course. However, to understand, e. g. Newton’s second law (“Force is the first time derivative of momentum”), one has to have basic knowledge of calculus. However, one-dimensional differential calculus will not be discussed until the middle of the first semester, not to mention the differential calculus in higher dimensions. Now this may not be a serious problem because most students have already learned how to differentiate real differentiable functions in high school. But, for instance, the line integral to calculate work is treated in mathematics within the scope of vector calculus in second semester, at the earliest. The same course covers the topics “oscillations and waves” as well, whereas the course “Ordinary and Partial Differential Equations for Engineers” is intended for third semester students (sophomores). In
Pre-Freshmen Students Gearing up with Early Bird
375
practice, lecturers of engineering classes manage this situation by inserting exercises on relevant mathematical topics when needed. This is mostly dissatisfying for both professors and students because of the loss of time, and because, therefore, the mathematical topics cannot be discussed in detail as the matter requires. This is a fundamental problem: Mathematicians wish to teach the mathematics starting with foundations and explaining every detail. Engineering teachers do not want to wait one or two semesters, but they want to teach “their own” subjects to engineering students starting from the beginning to give them hands-on experience as soon as possible. On the one hand, engineering students are happy when they can understand the mathematical concepts thoroughly and do not have to use recipes without comprehension. On the other hand, these students would be very frustrated if they had to deal only with mathematics for one or two semesters. Therefore, this fundamental problem will probably never be resolved. As learning is a “non-linear” process students will profit most by a good mixture of mathematics and engineering classes. Students who participated in Early Bird already have knowledge of the first semester mathematics courses at their disposal when they start their studies. This may help them to get along better with their studies in the first semesters. Another advantage of Early Bird is the fact that the students can focus on learning mathematics during nine weeks without any other classes at the same time. Additionally, students starting their study in winter term have the opportunity to use the time between high school graduation (German Abitur) and their first semester.
3 Implementation At the Berlin Institute of Technology approximately 27.000 students are enrolled. Every semester there are about 1400–2400 (most students are starting in winter term; thus, more first semester students are attending the courses in winter term) participants attending the courses “Calculus I for Engineers” and “Linear Algebra for Engineers” that are compulsory mathematics courses of first semester engineering students. Since it is not possible to teach them all by one professor, we offer a set of parallel lectures with 300–500 participants each. In addition smaller exercise/recitation classes are offered where the students attend lessons by Assistant Professors and student tutors. The idea of Early Bird is to shift two first semester lecture and associated exercise classes into the semester break. Concerning the high school graduates who attend Early Bird in summer, first term mathematics is already available at beginning of their studies in autumn. They get the opportunity to pass the course and final exam before start of studies. In case of failure, there will be no negative consequences for the students because they are not yet formally enrolled. At beginning of Early Bird it is not yet decided who will get admission to the university. Therefore, there is always a certain risk by selecting the Early Bird-participants, because some of them will be readmissioned. By accepting only persons who applied for a university place at Berlin Institute of Technology and
376
Erhard Zorn et al.
Fig. 1 Lecture hall with Early Bird students
who have high chance for admission (high school marks), we effectively can reduce this risk. We want to point out that the Early Bird course is not a remedial course of high school mathematics [Fri07], [Sch01], [SS05]: Many universities are offering a variety of “bridge courses” or preparatory courses. These courses are offered to bridge the gap between high school and university, and many of them are dealing with mathematics or physics. At least at German universities, as far as the authors know, these are additional courses, and they are intended as a repetition of the subjects that (should) have been learned at high school. Additionally, some courses are designed to give the beginning students the opportunity to discover their strength and weakness [Bud95]. The aim of Early Bird is not to offer a remedial course with main emphasis on the repetition of the mathematics from high school, but to give the opportunity to take standard courses of mathematics for first semester students before the first semester starts.
4 Early Bird vs. Regular Courses One semester has the duration of about 15 weeks. The regular semester course “Calculus I for Engineers” includes two lectures and one exercise class per week, “Linear Algebra for Engineers” one lecture and one exercise class per week. The
Pre-Freshmen Students Gearing up with Early Bird
377
Fig. 2 Early Bird Students discussing mathematics after the recitation session
total number of lectures is the same in Early Bird, since every day one lecture is given during nine weeks. Shorter exercise classes take place every day, and the students have to solve (short) homework assignments every day, while the students in regular semester courses get one larger homework assignment per week in both courses. The overall amount of work is therefore the same, but Early Bird lasts only nine weeks and thus has to be considered as a full-time course. Since the participants don’t have any other courses during Early Bird, they can fully concentrate on mathematics and prepare well for both final exams. In the last two years we managed to give the students one free week between the end of the course and the examinations so that they can learn sufficiently. This time, in this free week a summer camp has been organized where for instance recitation lessons were voluntarily offered by teaching assistants. The relaxed atmosphere between the teaching team and the Early Birds has always been conducive to good learning success. For the first and second Early Bird course we had additional funds by the university to have smaller exercise classes (16 students per teaching assistant/student tutor). This time without additional funds we had 24 students per exercise class. Teachers who have been involved in this year’s and the last years’ Early Bird courses have the impression that the negative effects of the larger classes have been compensated by the summer camp.
378
Erhard Zorn et al.
Fig. 3 Early Bird students at the summer camp discussing mathematics during lunch break
5 Results The immediate success is quite evident when we analyze the exams’ results. The Early Birds take part in the same written examinations as the students of the previous summer term so that we easily can compare the results of both groups. The following table shows the percentage of students who passed the exam. Obviously the intensive and exclusive dealing with mathematics by the Early Birds shows its effects on these results. Furthermore, though the workload in this course is very high for students, 99 % (211 out of 213 who answered to this question in a poll at the end of the course) are recommending this course to other prospective engineering students. Table 1 Comparison of the examination results of Non-Early Bird/Early Bird students Year
Subjekt
Exam passed (Non-Early Birds)
Exam passed (Early Birds)
Ratio (Early-Birds/ Non-Early Birds)
2006 2006 2007 2007 2008 2008
Calculus I Linear Algebra Calculus I Linear Algebra Calculus I Linear Algebra
54 % 73 % 47 % 55 % 46 % 71 %
78 % 89 % 63 % 59 % 63 % 73 %
1.44 1.22 1.34 1.07 1.37 1.03
Pre-Freshmen Students Gearing up with Early Bird
379
References [Bud95] [Fri07] [Sch01] [SS05]
Dan Budny. Mathematics Bridge Program. ASEE, 1995. Klaus Fritzsche. Mathematik für Einsteiger. Sepktrum Akademischer Verlag, Heidelberg, 4th edition, 2007. Winfried Scharlau. Schulwissen Mathematik: Ein Überblick. Vieweg, Braunschweig, Braunschweig, 2001. 3rd edition. Winfried Schirotzek and Siegfried Scholz. Starthilfe Mathematik. B.G. Teubner, Wiesbaden, Wiesbaden, 2005. 5th edition.
Part III
Cognitive IT-supported processes for heterogeneous and cooperative systems
Software Architecture, Knowledge Compiler and Ontology Design for Cognitive Technical Systems Suitable for Controlling Assembly Tasks Eckart Hauck, Daniel Ewert, Arno Gramatke, Klaus Henning
Abstract Companies in High-Wage countries face the so called polylemma of production. It consists out of two dilemmas. On the one hand the dilemma of value vs. planning orientation and on the other the dilemma scale vs. scope. To reduce the dilemma of planning vs. value orientation cognitive technical systems are a promising approach. To enact cognitive behavior a technical system has to incorporate an extensive knowledge base. This paper deals with the software architecture of such a system, the concept of a knowledge compiler which is able to transform different knowledge representation formalisms and the design for the ontology which is the foundation of the knowledge base.
1 Introduction In the last years, production in low-wage countries became popular with many companies by reason of low production costs. To slow down the development of shifting production to low-wage countries, new concepts for the production in high-wage countries have to be created. The production industry in high-wage countries is confronted with two dichotomies. On the one hand the dilemma value orientation vs. planning orientation and on the other hand the dilemma scale vs. scope. These two dilemmas span the so called polylemma of production technology [Bea07]. A reduction of this polylemma is the main aim of the cluster of excellence “Integrative Production Technology for High-Wage Countries” of the RWTH Aachen University. It deals with this problem in four core research areas. One of these research areas is “self optimizing production systems”. In this are the implementation of cognitive capabilities as prerequisite for self-optimization of a technical system is evaluated. Self optimization reduces the dilemma between value orientation vs. planning orientation due to a decreased effort in reprogramming the system for new tasks. In order to be able to behave in a cognitive way a E. Hauck (B) ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany e-mail: [email protected]
technical system has to incorporate a Knowledge Base in which the domain knowledge is stored [HGH08]. The paper is organized as follows: Chapter 2 gives an introduction to the software architecture of the system. Chapter 3 explains the design of the Knowledge Base. Chapter 4 then describes the Knowledge Compiler for such a system.
2 Software Architecture 2.1 Overview A cognitive technical system (CTS) suitable for controlling assembly tasks has to meet many requirements. These requirements are reflected in the software architecture. Fig. 1 shows the software architecture of the CTS. It is derived from the three-layer approach commonly used in autonomous robotics [RN02]. The design includes also a Presentation Layer for the human machine interaction and a Logging Module. The Knowledge Module contains the Knowledge Base and the Knowledge Compiler.
Fig. 1 Software architecture of the cognitive technical system
Cognitive Technical Systems Suitable for Controlling Assembly Tasks
385
2.2 Presentation Layer The Presentation Layer incorporates the Human-Machine-Interface (HMI). An operator interacting with the cognitive technical system can give the desired goal state for the assembly task, a description of the finished product, to the cognitive system. The operator can enter this description in a human understandable form, e.g. via a CAD program. To realize the desired machine transparency, the current system state is presented to the operator. If wanted, the operator can access all current and recent data in detail. Also, a knowledge engineer can access and edit the knowledge base through the HMI.
2.3 Planning Layer The Planning Layer contains the core elements that are responsible for decisionfinding. Based on the current world state and a given goal, the Cognitive Processor (CP) computes the best action to execute and returns its decision. An action can either be an abstract command to the Coordination Layer to execute a certain assembly step, or a request for more information. The Kernel component then invokes the action execution. If the CP cannot find a suitable action due to a lack of information about the objects recognized in the current state, and therefore returning a default query action, the kernel sends a query to the Knowledge Module asking for actions which are applicable to the present objects. The Knowledge Module returns rules accordingly. The Kernel then augments the ruleset of the CP with these rules. Should the query not have resulted in new rules, the Kernel sends a query to the operator. The user now has to decide which action to execute. The current approach relies on Soar as the Cognitive Processor. Soar is a cognitive architecture based on the “unified theory of cognition” [New94], which aims to model general intelligence [LNR87]. Soar is a rule based production system. Rules are fired if they match elements of the inner representation of the current world state and can modify this representation. Via input- and output-links Soar can communicate with its environment, e.g. to retrieve sensor information or invoke external commands.
2.4 Coordination Layer The Coordination Layer is the executable layer of the CTS. It provides executable services to the Planning Layer. These services correspond to the actions the Cognitive Processor can invoke. The Coordination Layer also processes the sensor data received from the connected hardware, and provides an abstract representation of the current world state on demand. The provided services are abstract actions, for example “move(blockA, blockB)”, meaning to grab a “blockA” and place it ontop a “blockB”. The Coordination Layer
386
Eckart Hauck et al.
realizes this command by invoking the necessary robot commands via the Reactive Layer to achieve the desired movements. That way, the Planning Layer is exculpated from the details of the robot movements, e.g. the exact coordinates of the blocklocations, etc., which leads, due to a reduced problem space, to faster decisions and more look-ahead.
2.5 Reactive Layer The Reactive Layer and its components handle the low level control of the system. The Controller allows the implementation of quick responding control loops, which monitor certain sensor values for validity and invoke, if necessary, given actions, for example a shutdown in case of an emergency. The Communicator sends the commands of the Coordination Layer to the robotic manipulators, were the actual physical execution of the commands takes place.
2.6 Knowledge Module The Knowledge Module provides the knowledge that is necessary for the assembly task. This knowledge is stored within the Knowledge Base (KB). The component Knowledge Compiler allows the translation of the KB’s formalism into the formalism of the Cognitive Processor. The translation is realized by the Translator component, while the Reasoner component is used to query the Knowledge Base. Fig. 2 gives a more detailed view of these components and their interfaces, together with the components of the Planning Layer.
Fig. 2 Component diagram of Planning Layer and Knowledge Module
Cognitive Technical Systems Suitable for Controlling Assembly Tasks
387
The functionalities of the Knowledge Module’s components are described in detail in the Chapters 3 and 4.
2.7 Logging Module The Logging Module is accessible layer-independently by all other modules. Its purpose is to provide a log of all data that are generated during the program execution. This data is either available online through the Presentation Layer, or, in case of a system crash, using external tools, to allow failure diagnosis.
3 Knowledge Base The Knowledge Base holds the entire domain knowledge within an OWL ontology [Sea04], which is based on description logics [Bra04]. The ontology contains, among other things, the two main classes “Service” and “Object” (See Fig. 3). “Service” holds descriptions of all services which are offered by the Coordination Layer. A service is described with the following elements: • Name: the services’ name, • Precondition: a logical formula, presented in SWRL [HPB+ 04], describing the necessary conditions for the service to allow its execution • Effect: a logical formula, in accordance to the Precondition • Parameter: list of parameters needed to invoke the service execution The service description is loosely orientated around the description of web-services in OWL-S [Mar04], particularly the service model. The class “Object” subsumes all objects that can occur as parts of the world state during the assembly process. These objects are classified into subclasses, where each subclass holds to all those individual objects which can be manipulated by a certain service. Right now, this classification has to be done manually. In future the ontology will be improved in a way that allows this classification to be inferred automatically. This can be achieved by augmenting the ontology with specifications of the robotic devices in such a way, that we can derive the objects on which a robot manipulator would be applicable, from its specifications. E.g., from the fact that a robot gripper could only be spread 10 cm, it can be derived, that it can grip accordant blocks with a maximal edge length of that size. Figure 3 shows a part of the Knowledge Base describing the class “MoveableBlock”. Classes are presented as circles, containing other circles as subclasses or individuals, which are presented as diamonds. All individuals of the class “MoveableBlock” are related to the same individual service “moveService”. We have organized our ontology in such a way, that each object individual in the ontology corresponds to the set of objects in the real world that have the same shape. We have found this being a good practice, since we don’t need to distinguish
388
Eckart Hauck et al.
Fig. 3 Diagram of a part of the Ontology showing the relation between blocks and services
between single objects when deciding about which service is applicable. More information about single objects is not needed within the ontology [NM01].
4 Knowledge Compiler The component Knowledge Compiler provides the means to extract requested information from the knowledge base and to generate accordant rules in the formalism of the CP. Figure 2 shows a more detailed view of this component. The compiler receives a request from the Planning Layer containing objects for which the CP could not find any action to apply to. For those objects the compiler now generates a query asking for services which are applicable to the particular object. These queries are sent to the Reasoner which accesses the Knowledge Base to derive this information. Out of the precondition, name and effect-descriptions of the returned services, the compiler then builds rules in the formalism of the CP and sends these back to the Planning Layer. Figure 4 shows the complete decision finding process including knowledge conversion.
Cognitive Technical Systems Suitable for Controlling Assembly Tasks
389
Fig. 4 Activity diagram of the decision finding process
Remark: Recently it is often advised to use headings instead of skip links since there is a broad support by screen readers to navigate with headings [Tha07]. Even, in the WCAG 2.0 both approaches are included. Here, skip links are preferred because the use of headings for meta content as navigation doesn’t match the ideas of the Semantic Web. Otherwise, heading navigation is not broadly supported by common browsers when they are used by physically challenged people. These people need as well a quick navigation through the content of a document. Thus, this workaround doesn’t match the idea of a technology independent pattern.
Using Web Accessibility Patterns for Web Application Development
585
Access Keys: Working with software applications Hotkeys (Shortcuts) is a great help since they simplify repetitive activities. In the context of accessible Web applications they are called Access keys. This inherently easy to solve problem is confounded by the absence of any general, application and system independent standards. Similarly, the operating modes of browser are not uniform. According to the Web Accessibility Technical Services (WATS) [Web05] all shortcuts with Latin letters are already occupied – in the English speaking world alone. In an international context, it is still more confusing. Only the number buttons remain available. In addition, the occupancy of Access keys usually is not shown. Currently this means that the use of Access keys is partially discouraged. Until today, there is no international standardization approach for an Access key-space in HTML documents like the standards of Apple or Microsoft for their system platforms. For visualization of Access keys different methods are possible: • • • •
Underlining in the text Explicit indication in the text Separate listing Notification at focus or Mouse-over
Bread Crumbs: In hierarchical document trees an indication of the current path supports users in orientation. The user is intuitively supplied with additional information about the current location, status etc. The use of breadcrumbs extends the main navigation with low additional consumption of space in the layout of the document. Instone [Ins03] mentions three different types of breadcrumbs: location breadcrumbs, path breadcrumbs and attribute breadcrumbs. Location breadcrumbs present the current position in a hierarchical document tree, path breadcrumbs show the navigation history whereas attribute breadcrumbs are used with product categories et.al. Further types of breadcrumbs are possible. Main Navigation: Regarding Accessibility main navigation provides the following functionality [Hel05]: overview over the application and its components, guidance through the application and simplified, more direct access to documents, parts of sites or functionalities. The lack of any HTML markup elements for persistent navigation in Web sites complicates development. Appropriate elements and attributes are included in the to-be recommendations of the W3C – Accessible Rich Internet Applications (WAI-ARIA) and XHTML 2. Without these elements the use of lists in HTML is advisable including a clear identification of the current position. Lists can be nested to represent hierarchical navigation structures. Site Map: is a table of contents providing an overview of all existing pages of a Web application. It complements the offer of navigation and improves clarity and guidance. It helps the user to find documents when the structure of the navigation is not obvious to him. Like navigation it must be easily accessible. The requirements are the same as for Main Navigation. Various levels of the hierarchy must be clearly marked. In HTML nested lists are recommended. Utility Navigation: The utility navigation offers supplementary functionality. These include for example a site map, help, contact, legal, search function or glossary. In current Web links to these functionalities are often available from all
586
Helmut Vieritz et al.
pages and besides the main navigation. Hong [Tha07] recommends that the links are included at the top-right corner of each document. Additionally, – to support accessibility – the functionalities can be easy activated with skip links and keyboard. The recommendations for the pattern Main Navigation are valid here as well. Sequence: The correct sequence of elements in a document (Web site etc.) is a content-relevant information and should not be changed by layout. Additionally, the sequence of repetitive elements in sites is always the same. Basic elements such as a link to the main page, navigation structures etc. are always in the same layout and presented in the same place providing a better overview and fast orientation. A recommended sequence [Ihm07] is as follows: 1. skip links, 2. breadcrumbs, 3. content, 4. main navigation and 5. utility navigation. Web Accessibility Patterns concerning Content: These patterns relate to the access to the content of documents, including any interactive components. Since they are not yet included in the presented concept these patterns are only mentioned here (see Table 1). Some of these patterns are discussed in detail in [Ihm07]. Supporting accessibility improves general usability as well. Nevertheless, both terms are not identical. The difference becomes more pronounced when examining their negations. Poor usability causes inconvenience as the user is presented with the wrong information or functionality. Poor accessibility means, however, that the information is not visible or available to the user. Thus, accessibility integrates aspects of usability, but the impact is more severe, as pattern set for specific users are often required for correct use – e.g., skip links and metadata. Currently our work is focused on the integration of overview, orientation and navigation patterns. The second category is related to content which is often generated by authors at runtime. This is part of future work.
4 Extending UWE for Accessibility UWE’s capabilities – the consistent support of the established UML standard and especially the use of UML activity diagrams – designate the UWE approach as a starting point to facilitate the development of accessible Web applications. Activity diagrams are well-suited to integrate a user-centered development approach based on the analysis and modeling of workflow processes. In addition, UWE is even compliant to the Model-Driven Architecture (MDA). The ATAG note that an authoring tool for accessible content should be accessible itself. This requirement is also outlined in the presented approach. The modeling notations and the UI of the tools have to be accessible. UML as a graphical notation language is difficult to access for the visually impaired requiring an alternative form of representation. A one-to-one mapping of graphical models into tactile illustrations is not possible because haptic sense can capture neither the details nor structural relationships. For this purpose, the Object Management Group (OMG) has developed the Human-Usable Textual Notation (HUTN) [Obj04]. HUTN was
Using Web Accessibility Patterns for Web Application Development
587
designed to present UML diagrams in a short text form. The already existing XML Metadata Interchange (XMI) was developed to support the exchange of machine data based on MOF. XMI is not well suited to be read by human. HUTN fills this gap and is compliant to XMI. Thus, it is possible to integrate HUTN seamlessly into the UWE design process. With the use of HUTN the models themselves can be presented in an accessible mode. Details of integration are part of further research. The concept is presented with examples for a lecture management tool. UWE modeling starts by gathering requirements. Use case diagrams describe functionalities. Additionally, hypertext-related use cases are modeled and signed with UWE-specific stereotypes. Icons are available in all models to identify the stereotypes. In Fig. 4 a small rectangle icon is used to mark hypertext-specific use cases. The lecturer can create, edit, and delete lectures whereas the student can search and attend lectures. Since the lecturer inherits the use cases of the student he/she can search and attend lectures (e.g. for testing purposes). In UWE, process models - notated as activity diagrams - can complete the requirements modeling for non-trivial use cases with functionality more then navigation. A process structure model allows the modeling of data beyond the navigation and conceptual model. A class diagram serves for notation. If the process model is detailed sufficiently, it may be transformed directly and executed. Thus, a quick evaluation of modeling can be supported. The concept presented here extends the UWE process with an extended task model. Task modeling is the appropriate starting point to integrate the requirements of all users from the beginning (see sec. 1). Pre-modeling activities like Systematic Layout Planning (SLP) serve to identify the use cases and details of workflow and to analyze the mental models of the to-be users. Tasks are independent from
Fig. 4 Use case diagram for the lecture management tool
588
Helmut Vieritz et al.
Fig. 5 Elementary actions for task modeling
personal, contextual or technical restrictions (see fig. 1). They are separated into activities and actions, which can be considered as being atomic. On the other side, they are still independent from physical operations to avoid dependency on UI technology like mouse, keyboard, or screen reader etc. These actions can be separated into five basic action classes - input, edit, select, execute, and inform (see fig. 5). Some sub-classifications are possible to facilitate further modeling (not yet used): start/go, stop/exit, create/delete, duplicate, toggle, monitor etc. The use of workflow patterns [Wor07a] facilitates the modeling of activities. The task model includes actions, their relations and context (resources). Figure 6 facilitates the modeling of activities. The task model includes actions, their relations, and context (resources). Figure 6 shows an example of a task model by grouping a use case as a set of actions. The actions are signed according to the corresponding stereotype for the five basic classes. The stereotype system serves to signify application actions. The relations between the actions are presented as edges (associations). Context is modeled with objects, constraints etc. A detailed task model serves to support modeling of data, navigation, and abstract presentation. Part of current research is the evaluation of Business Process Modeling Notation (BPMN) as modeling language for the task model. BPMN as a notation language focuses on business process analysis and planning and finds wider use in this field than UML. BPMN is more intuitive and its use would open the modeling process to other developer groups. UML activity diagrams and BPMN are both equally powerful. The OMG is responsible for the development and standardization of both notations. The activity chart shown in fig. 6 would look similar in BPMN. The integration of BPMN would simplify the
Using Web Accessibility Patterns for Web Application Development
589
Fig. 6 Activity diagram for use case create Lecture
notation of task models. The UWE conceptual model describes the data, which the application manages (see fig. 7). The specification of content and users is modeled with a class diagram. Content and user related data are separated. This supports the integration of adaptability into the design process. A UML class diagram is used to model the navigation structure in UWE (see fig. 8). Based on this conceptual model the development of the navigation model is supported by a semi-automatic transformation. The nodes of the model represent the available information and the edges possibilities to change nodes. Since nodes do not always represent pages edges are not always links to be activated by the user. The navigation model includes process nodes, which serve as entry points to supported workflow processes. The UWE process also supports the abstract modeling of the UI (see fig. 9). Only the structure of UI components is mapped and details such as colors, fonts, positions etc. are omitted. A class diagram is used for notation. UWE supports the concept Page for notation of a single Web site. Presentation classes can be grouped hierarchically in a page.
590
Fig. 7 Conceptual model for the lecture management tool
Fig. 8 Navigation model for lecture management tool
Helmut Vieritz et al.
Using Web Accessibility Patterns for Web Application Development
591
Fig. 9 Abstract UI model: Example for Home
4.1 Integration of Web Accessibility Patterns We now discuss the use of the presented Web Accessibility Patterns to model accessible UI components. As already mentioned in Sect. 3, the patterns serve for navigation, guidance, and overview. The UWE constructs used are described in the UWE meta-model and Profile [KK08]. The recommended information about documents in the pattern Metadata is anchored in the Presentation Model of the UWE process. Therefore, the Page class of the presentation package is extended with the attributes title, description, keywords, company, author, copyright, and robots. Additional metadata is conceivable, corresponding to the proposals of the Dublin Core Metadata Initiative (DCMI). The integration of default values and generalized parameters facilitates modeling bigger applications. The UWE Navigation Model includes information to realize the pattern Document Relations. These are attributes of nodes of the navigation package. Nodes with set isLandmark or isHome attribute are added as link elements in the head area of the HTML documents. In addition, attributes from the Presentation Model must be known for title attributes in the link element. The pattern Main Navigation will be implemented in HTML as a nested list of li elements. Therefore, the same data as for Document Relations is used. The integration into layout is specified in the UWE Presentation Model. The Pattern Utility Navigation includes links to auxiliary functionalities, which are accessible from all pages. Such nodes have the attribute isLandmark in the UWE Navigation Model. Additionally, the attribute includesUtilityFunction is introduced for presentation classes in the Presentation Model, which must be set manually.
592
Helmut Vieritz et al.
This attribute differentiates between elements of the main and those of the utility navigation. The modeling of the presentation of the utility navigation is done through the Presentation Model. The pattern Skip Links requires information from the UWE Presentation Model. The presentation classes, which are direct sub-elements of the Page class are compiled into HTML with anchors and directly linked from the beginning of the HTML page. In addition, a hasSkipLink attribute is added for presentation classes in order to handle Skip Links in complex layouts. This attribute can be set in presentation model to enforce a direct link to the related page component. The information for the pattern Site Map can be gained from the Navigation and Presentation Model. Beginning with the isHome node navigation routes between nodes associated with Pages are weighted. An edge to an isLandmark node gets a higher weight. The navigation path with the highest weight - the shortest outgoing from the main page - is included in the hierarchical Site map structure. The Site map is notated in the presentation model as an independent navigation node and implemented in HTML as a nested list. Three basic types of Breadcrumb pattern have been categorized (see sec. 1). Without adaptation, only the Location Breadcrumb can be modeled. It follows the path in the Site map and can be generated automatically from it. The other breadcrumbs can be generated when the modeling of adaptation in UWE can be used. That is part of further work. For the pattern Access keys an additional attribute is required in the Navigation and Presentation Model. Nodes or presentation classes are extended with an additional hasKey attribute. The corresponding HTML elements are marked up with the accesskey attribute. As discussed, the challenges for user agents must be solved too finding a standard for HTML documents. The pattern Sequence is realized by integrating the different pattern generated in the previous steps according the sequence (see sec. 3). Tab.2 shows the additional attributes require to support our approach. All patterns of sec. 1 are included assuring a broad support of accessibility in navigation, guidance and overview.
Table 2 Additional attributes in UWE metamodel Attribute name Class name Type author company copyright description keywords robots title includesUtilityFunction hasSkipLink hasKey hasKey
Using Web Accessibility Patterns for Web Application Development
593
5 Conclusion In summary, the presented Web Accessibility Patterns can be included in the modeling process. Thus, combined with a user-centered approach and task modeling, model-driven Web development has the potential to facilitate the development of accessible Web applications. The extra effort is small and does not demand additional, specialized knowledge about the requirements of accessibility from the developer. The advantage of the approach is in the close interaction with the UWE modeling process. What can be modeled and generated with UWE can be modeled and generated accessibly in the fields of navigation, guidance, and overview using our approach. The presented concept facilitates the development of accessible applications since a model developer does not need a detailed experience of accessibility. The approach combines the advantages of model-driven development with the support of accessible UIs. The design process will be also accessible for all participants and allows the early evaluation of the declared models. The project’s ongoing work is the integration of pattern for accessible Web content. Not all mentioned Web Accessibility Patterns are easy to support. However, the basic idea can be extended to other fields. Part of future work is the implementation and evaluation of the concept and further integration of applications dealing with the presentation of complex information such as knowledge. Also, patterns for dynamically generated content are under consideration. With Web 2.0, new forms of content generation are getting more and more importance. Dynamic generated content causes new accessibility problems which are focused by the Accessible Rich Internet Applications Recommendation of the WAI (WAI-ARIA) [Hon06].
References [Bre05] [CCT+ 03]
[De 98]
[GHSY] [Hel05] [Hon06] [Ihm07] [Ins03]
G. Bretthauer. Automatisierungstechnik – Quo vadis? Neun Thesen zur zukünftigen Entwicklung. at – Automatisierungstechnik, 53(4-5):155–157, 2005. G. Calvary, J. Coutaz, D. Thevenin, Q. Limbourg, L. Bouillon, and J. Vanderdonckt. A Unifying Reference Framework for Multi-Target User Interfaces. Interacting with Computers, 15(3):289–308, June 2003. De Troyer, O.M.F. and Leune, C.J. WSDM: a User-Centered Design Method for Web Sites. In Computer Networks and ISDN Systems, volume 30, No. 1-7, pages 85–93, 1998. C. Goble, S. Harper, R. Stevens, and Y. Yesilada. Dante – Mobility Support for Visually Impaired Web Travellers. http://dante.man.ac.uk, Accessed: 05/22/2008. Jan Eric Hellbusch. Barrierefreies Webdesign – Praxishandbuch für Webgestaltung und grafische Programmoberflächen. dpunkt.verlag, Heidelberg, 2005. David Hong. Utility Navigation. http://jimthatcher.com/skipnav.htm, Accessed:10/26 /2008, 2006. Simon Ihmig. Web-Accessibility Patterns. Thesis, Universität Hamburg, Department Informatik, 2007. Keith Instone. Breadcrumbs. http://instone.org/files/KEI-3Breadcrumbs.pdf, Accessed: 08/22/2008, 2003.
594 [JPV08]
Helmut Vieritz et al.
S. Jeschke, O. Pfeiffer, and H. Vieritz. Developing Accessible Applications with User-Centered Architecture. 2008 International Conference on Computer and Information Science (IEEE), Portland/Oregon, May 2008. [JV06] S. Jeschke and H. Vieritz. Accessibility and Model-Based Web Application Development for eLearning-Environments. CISSE 2006 – Second International Joint Conferences on Computer, Information, and Systems Sciences, and Engineering (OnlineConference), Springer, December 2006. [JV07a] S. Jeschke and H. Vieritz. BeLearning: Using Mental Models to Develop Accessible eLearning Applications (ICTA/UTIC 2007). Conference Proceedings of First International Conference on Information and Communication Technology & Accessibility, Hammamet, Tunesia, April 2007. [JV07b] S. Jeschke and H. Vieritz. BeLearning: Using Mental Models to Design Accessible eLearning Applications. FIE 2007 – The 2007 Frontiers in Education Conference (IEEE), Milwaukee/Wisconsin, October 2007. [KK02] Nora Koch and A. Kraus. The Expressive Power of UML-based Web Engineering. http://www.pst.informatik.uni-muenchen.de/personen/kochn/IWWOST02koch-kraus.PDF, Accessed: 08/22/2008, June 2002. [KK08] Christian Kroiß and Nora Koch. UWE Metamodel and Profile. User Guide and Reference. Technical report, Institute for Informatics, Ludwig-Maximilians-Universität, München, 2008. [KKWZ07] Alexander Knapp, Nora Koch, Martin Wirsing, and Gefei Zhang. UWE – Ein Ansatz zur modellgetriebenen Entwicklung von Webanwendungen. i-com, 3:5–12, 2007. [Obj04] Object Management Group (OMG). Human-Usable Textual Notation Specification (HUTN) V 1.0. http://www.omg.org/technology/documents/formal/hutn.htm, Accessed: 08/22/2008, 2004. [Pat00] F. Paternò. Model-Based Design and Evaluation of Interactive Applications. Springer, Berlin, 2000. [PCY+ 05] P. Plessers, S. Casteleyn, Y. Yesilada, O. De Troyer, R. Stevens, S. Harper, and C. Goble. Accessibility: A Web Engineering Approach, May 2005. [Tha07] J. Thatcher. Skip Navigation Links. http://groups.ischool.berkeley.edu/ui_ designpatterns/webpatterns2/webpa tterns/pattern.php?id=25, Accessed:04/29/2008, 2007. [Web05] Web Accessibility Technical Services (WATS). Accesskeys and Reserved Keystroke Combinations. http://www.wats.ca/show.php?contentid=43, Accessed: 08/22/2008, 2005. [Whi04] Stephen A. White. Process Modeling Notations and Workflow Patterns. http://www. bpmn.org/Documents/Notations, Accessed: 08/22/2008, 2004. [Wor00] World Wide Web Consortium (W3C). Authoring Tool Accessibility Guidelines 1.0. http://www.w3.org/TR/ATAG10/, Accessed: 08/22/2008, February 2000. [Wor02] World Wide Web Consortium (W3C). User Agent Accessibility Guidelines 1.0. http://www.w3.org/TR/UAAG10/, Accessed: 08/22/2008, December 2002. [Wor06] World Wide Web Consortium (W3C). Authoring Tool Accessibility Guidelines 2.0 (Working Draft). http://www.w3.org/TR/ATAG20/,Accessed:08/22/2008,December 2006. [Wor07a] Workflow Patterns Initiative. Workflow Patterns. http://www.workflowpatterns.com, Accessed: 04/28/2008, 2007. [Wor07b] World Wide Web Consortium (W3C). Accessible Rich Internet Applications Suite (WAI-ARIA). http://www.w3.org/WAI/intro/aria, Accessed: 08/22/2008, November 2007. [Wor08a] World Wide Web Consortium (W3C). Web Accessibility Initiave. http://www.w3. org/WAI/, Accessed: 08/22/2008, 2008. [Wor08b] World Wide Web Consortium (W3C). Web Content Accessibility Guidelines 2.0. http://www.w3.org/WAI/intro/wcag20/, Accessed: 08/22/2008, May 2008.
Part V
Semantic networks and ontologies for complex value chains and virtual environments
Crystalline Ge1−x Snx Heterostructures in Lateral High-Speed Devices Sabina Jeschke, Olivier Pfeiffer, Joerg Schulze, Marc Wilke
Abstract This paper describes an approach to manufacture high-speed Germanium MOSFETS with strained channels made from Ge1−x Snx -alloys while embedding the needed technology process flow into a virtual knowledge management environment based on a virtual nano electrical lab. Keywords MOSFET · CVD · Epitaxy · virtual knowledge spaces · MBE
1 Introduction In the last two decades, the materials system Si1−y Ge y (0 < y ≤ 1) was subject of intense successful research for the fabrication of high-speed devices and their integration into standard CMOS (Complementary Metal-Oxide-Semiconductor) and bipolar technology [WHTG94, HNE+ 02, KLC+ 06] and into novel device concepts [Sch05, BSE05, DHJ+ 85, MSM+ 01]. For future nanoelectronics, one step further along the group-IV-elements of the periodic table – from Silicon/Germanium to Tin – opens the complete new field of materials research on Ge1−x Snx with (indicated by theoretical calculations [ABB+ 08]) a large potential for new applications ranging from high-speed transistors to optical infrared detectors and light sources compatible to mainstream Silicon-technology. Main reasons for this are: 1) all three elements, Silicon, Germanium and Tin, crystallize in a diamond crystal lattice with different lattice constants leading to the possibility of strain- and band gap-induced band engineering and 2) the incorporation of Tin into a Germanium crystal matrix transforms the indirect semiconductor Germanium into the direct semiconductor alloy Ge1−x Snx (x ≥ 0.1) [ABB+ 08, Sor06, KC07]. S. Jeschke (B) IMA/ZLW & IfU - RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany e-mail: [email protected]
At the moment, worldwide only three groups are working on manufacturing Ge1−x Snx -alloys by using different methods like MBE (Molecular Beam Epitaxy) [TSY+ 07], MBE-based SPE (Solid Phase Epitaxy) [THAP96] and CVD (Chemical Vapor Deposition ) [KC07, TCF+ 06]. First results shown in [KC07, TSY+ 07, THAP96, TCF+ 06] verify that the epitaxial growth Ge1−x Snx -crystals is a very challenging topic if the whole range 0 < x ≤ 1 is addressed, due to the very small solubility of Tin in the Germanium-matrix. Furthermore, a doping of such alloys seems to be challenging, too. In conclusion the reported results reveal that the best – and momentarily maybe only approach – to fabricate strained, crystalline and doped Ge1−x Snx -alloys with a large range of Tin concentration x is given by the use of a low-temperature MBE. A dedicated research and development work introducing those alloys in nano electronic devices was not reported so far. This paper is organized as follows. Following this introduction, chapter 2 describes the technical concept of the system manufacturing of Ge1−x Snx -alloys with high electric quality for the immediate use in electronic devices. Chapter 3 outlines the major challenges for reaching the capability to produce the proposed device structures by linking the existing technology lines of the partners and describes the process flow. Chapter 4 describes the embedding of the presented process in the NetLabs framework to strengthen research cooperations. As far as available we describe some related work in chapter 5. Prior to the conclusion, chapter 6 outlines the roadmap of the.
2 Expected Outcome Based on the situation described in chapter I, the central idea of this project to focus on the manufacturing of Ge1−x Snx -alloys with high electric quality for the immediate use in electronic devices. As a first test device, a Germanium-MOSFET (MetalOxide-Semiconductor Field-Effect Transistor) with a transistor channel made from Ge1−x Snx is targeted. The proposed structure of the device is shown in Fig. 1. Next to the technology process that has to be established to produce these MOSFET-structures and the experimental results obtained from this nanoelectronics
Fig. 1 Schematic structure of planned MOSFET-structure
Crystalline Ge1−x Snx Heterostructures in Lateral High-Speed Devices
599
research project answers to the following questions with both, scientific as well as industrial relevance have to be given: • Is it possible to grow crystalline Ge1−x Snx -alloys on virtual Si1−y Ge y - or Germanium-substrates in a wide x-range keeping strain and relaxation processes under control? • Is the electrical quality of these crystals high enough to propose Ge1−x Snx -alloys as new promising materials for advanced future electronics in the nanometer regime? If so, for what device concepts should it be used? This is combined with the sub-questions: 1. Is there a significant mobility enhancement by introducing Tin into the Germanium-matrix with industrial relevance e.g., for CMOS-technology? 2. Is the theoretical expectation of a direct Ge1−x Snx -semiconductor for x > 0.1 true? To reduce the obvious difficulties caused by the spatial distance between the team and the manufacturing labs, a cooperative virtual infrastructure meeting the demands of a future virtual nanoelectronics lab, allowing the partners to act as one nanoelectronics lab with fast internal communication, joint technology flows and a common data and result base. First goals are the establishment of a virtual joint protocol system/ log book for sample tracking, data and result storage and data mining, a remote control platform for resource sharing of selected equipment as electrical characterization tools, and the integration of video and audio conferencing capabilities for a direct communication [JSR08, JPT08]. Therefore, the overall objective is twofold: • A low-temperature Ge1−x Snx -technology for MOSFET manufacturing has to be developed and • Technology flow has to be embedded into a virtual knowledge environment.
3 Description on the Ongoing Activities The major challenge for reaching the capability to produce the proposed device structures is the establishment of a joint technology line, where the existing technology lines of the partners are linked together. To produce the devices, the following process flow must be established: 1. Low-temperature MBE growth of the semiconductor layer stack (n(p)-type doped Ge1−x Snx /p(n)-type doped Germanium) on Germanium-on Insulatorsubstrate (GOI-substrate) or virtual Germanium-substrate 2. Patterning of the p(n)-type doped Germanium-top layer to form Source and Drain 3. Atomic layer deposition (ALD) of Si3 N4 to form the Gate-oxide 4. Deposition of p(n)-type doped poly-Si1−y Ge y by CVD
600
S. Jeschke et al.
5. Patterning of the poly- Si1−y Ge y -film to form a Gate-electrode 6. CVD of Si3 N4 to protect the layers 7. Contact hole opening and metallization All technology steps are accompanied by several analytical measurements as secondary ion mass spectrometry (SIMS), ellipsometry, µRAMAN spectrometry, profilometry, scanning electron microscopy (SEM), etc. to control and to characterize the individual technology steps. After manufacturing, the devices are characterized electrically by using ICV-measurements, Hall-measurements, and special high-frequency (HF) measurements. In the last two decades, one major research field of the IHT of University of Stuttgart epitaxial growth of Si1−y Ge y -heterostructures by means of low-temperature MBE [Kas00, ESK04, WOK+ 08, KOL08] and CVD [YSM06, NSM07, SSM06, SSM08], respectively, and their implementation into vertical optic-electrical devices as MOSFET, MODFET (Modulation-Doped FET), (H)BT ((Hetero)Bipolar Transistor), and pin-photodetectors. The group is an expert in epitaxial growth of group-IV-semiconductors with high worldwide reputation and in manufacturing and characterization of devices made from these materials. Running such a process in spatially distributed labs forms the needs for additional structures to reduce risks of shipping, increase communication between partners, unify protocol systems, bundle measurement tasks etc. Therefore, the idea of a virtual nanoelectronics lab has been developed. The major activities for realization are to create a virtual joint protocol system/ log book for tracking the status of the samples and to collect all technology data/ protocols and analytical results. It will be accessible by all team members of the partners at any time. The data will be stored in a way that special data mining algorithms can be started by the individual operators to find causalities between individual technology steps and the electrical performance data of the devices. In detail: The aim of supporting the proposed nanotechnology project by means of virtual knowledge spaces is the construction of a “second generation” repository addressing the technical challenges of experimental natural sciences and engineering [JSR08, JPT08]. While classical repositories host static and long-living data, e.g., publications, the aim of this concept is an extension of already existing repositories in order to include • volatile data, e.g., primary data from current experiments, • “dynamic data”, i.e. informal documents and documentations (protocols, notes) which are produced along the scientific workflow in the laboratory, • the life cycle of all date, • the experiments and analysis tools themselves, thus broadening access to heterogeneous experimental resources for research and education purposes, • enhancing scientific communities by promoting international cooperation and collaboration in high-technology. Moreover, semantic retrieval mechanisms come into place, enabling intelligent searchability of primary data as well as automated and semi-automated plausibility checking of results. Finally, to determine and distinguish relevant from irrelevant
Crystalline Ge1−x Snx Heterostructures in Lateral High-Speed Devices
601
influences, learning systems using AI technologies are needed. An interactive virtual platform linking diverse remote controllable analysis tools for material production in nanotechnology features sustainable long-term scientific advantages [JBH+ 09c]: A more efficient and systematic development of new materials is promoted: compiled knowledge is indexed and secured on a long-term basis. Once developed syntheses can – by a successful standardization – be made accessible to researches whose primary interests are e.g., physical measurements on that special material, rather than its synthesis. Constant material availability independent of the synthesizing chemist permits a continuous reliable research on this material. This point is also of particular importance since syntheses are usually developed by graduate students and/or postdocs, who stay only for a limited period at their respective research institute and knowledge transfer does not always take the prominent role that it should. Since only well functioning syntheses are standardizable, some kind of quality control for the synthesis directives is automatically integrated. Thus, standardization of the synthesis will also be an important section of challenge within a synthetic Ph. D. or master’s thesis. The interactive platform also facilitates the access to research methods that are not available locally because of missing giant equipment, special devices, or particularly developed research equipment. Similarly, extended ways of contact are possible on the scientific level that have come about on another way, thus promoting the development of national and in particular international contacts. An independently generated electronic laboratory journal has the advantage that all results together with the necessary parameters and all steps: synthesis ⇒ characterization ⇒ application are retained uniformly and clearly arranged. In addition, research projects that have been classified as not successful and therefore remained unevaluated are archived. At second glance, an expert might possibly come to a surprising discovery (increase the ratio of innovative random discoveries). Established synthesis approaches can also be used also as exemplary sample experiments in teachings. Starting in May 2009, the project BW-eLabs [JOH+ ] is funded by the ministry of sciences, research and arts Baden-Württemberg (MWK) The aim of the BWeLabs architecture (networked virtual and remote labs in Baden-Württemberg) is the expansion of the access to heterogeneous experimental resources (remote and virtual [JRST07b]) including data access for research purposes [JBH+ 09c]. The BWeLabs architecture is not restricted to nanotechnology but used this field as a pilot discipline because of its exceptionally big influence on the current technological development and to which the role of a central key technology of the 21st century is ascribed. In this cost intensive area, access to experimental equipment is an important prerequisite for ensuring scientific progress for all scientific communities involved. Existing infrastructure, e.g., digital libraries, decentralized tools, and repositories, are embedded into the 3D-Plattform BW-eLabs. The scientific content management system eSciDoc, developed by FIZ Karlsruhe within a BMBF-funded project, plays an important role since it already facilitates users-inquiries for static data to the corresponding repositories, thus realizing the essential semantic infrastructure for indexing the raw data. Metadata sets suitable for the purpose of describing
602
S. Jeschke et al.
experiments and raw data of the technical disciplines), particularly safeguarding a complete documentation and repeatability of the experimental setups, are being developed within the project. BW-eLabs - as well as its CMS component eSciDoc stand under OpenAccess-Policy and sees itself as an open network for scientific data and experimental set-ups. The project BW-eLabs is complemented by Lila (Library of Labs, [JBR09]), funded by the E as part of the eContentPlus program. Whereas BW-eLabs builds a virtual knowledge space for cooperative experiments from the viewpoint of research, LiLa addresses the same goal but from the viewpoint of academic education [JBH+ 09a]. Today, the insufficient availability of experimental capacities has a great impact on the efficiency of the academic education in the areas of engineering and natural sciences since experiments form a central part of the research methodology within these fields. Their realization is frequently subject to numerous restrictions, particularly financial constraints, areal capacities, and support potential. The employment of new media offers new concepts to overcome these challenges by using remote and virtual laboratories. In both projects, a common and open 3Dbased platform is build: A “virtual world”, based on SUN Wonderland, provides the basis for the integration of the hereby available laboratories and for the creation of a communication platform for promoting communication between distributed working research groups. Here, experiments, communication tools, and available digital libraries are embedded in a simulated three-dimensional world. Through this, not only the geographical distance between researchers can be bridged by means of manifold communication tools; by the electronic connection to the experiment at the same time archiving of measurement data and processes (see above) is achieved. Integration of familiar tools and established infrastructure, intuitive access to the system and individualization help raising acceptance and usability of the portal.
4 Embedding Into NetLabs and Strengthening Research Cooperation Strong interdisciplinary factors led to active exchange and overlapping of chemistry, physics, material sciences, biology with nano technology. The multitude of scientific methods and approaches necessary for the realization of research, stimulates and demands the joint collaboration of diverse research groups, should they choose to experiment, evaluate and publish any subject related works. This led to the construction of larger scientific centers in countless universities worldwide. In particular, the acquisition and common usage of larger, more complex devices can be realized only in some of these centers. On account of the interdisciplinary and interlinking necessities, nanotechnology is an extremely fitting approach to be used within N ET L ABS – in the form of interlinking virtual and remote laboratories to set up new standards in the knowledge management field [JPT06, JPT08]. The Berlin Institute of Technology, and in particular the Thomsen study group [rem], disposes of extensive experience in terms of nano structure research on international level: research done on carbon nano tubes and related systems: nano
Crystalline Ge1−x Snx Heterostructures in Lateral High-Speed Devices
603
ribbons, fullerene, graphene as well as research on Si-nano wires and CdSe-nano rods. These experimental and theoretical preparations deliver the ideal conditions for the implementation of remote-nanotechnology labs. The devices available used nowadays in nanostructure research, are made accessible to the scientific community within the N ET L ABS scheme: • A Remote-Raman System with physical characterizations of carbon nano tubes is provided (e.g. the diameter regulations and grouping effects were examined). Raman spectroscopy permits a contact free analysis of physical properties and is by means of purely optical technology also indestructible. The provided remoteRaman-spectrometer has a stimulating wavelength of 532nm and covers a spectral range from -1400cm−1 to 3300cm−1 . The world-wide first remote-spectrometer provided already today permits the spectroscopy of nano-materials. The remote farm is used by more than 1000 visitors per month from all parts of the world. • A second device that can also be used ubiquitous by many study groups is the scanning electron microscope. This device allows for image production, as well as other characterizations of nanostructures. After minor technical extensions both experimental set-ups are integrated in the portal within the NetLabs scheme. The scanning electron microscope has an acceleration voltage of 5kV and a continuously adjustable spatial resolution of up to 30nm. It is therefore ideal for the characterization of nano structures. Both set-ups complementing the “small and large analysis “already commenced in the BW- E L ABS [JBH+ 09b, JBH+ 09a], require suitable scheduling, a test dispatch and on-site support. Nevertheless, the main focus points are: integration of systems in the 3D Wonderland engine [pro], development of more suitable interfaces between data delivered by the devices and data infrastructures originating from other channels, development of specialist nanotechnology ontologies, and the development of an intelligent access control system. The presented project allows the partners to step into a new, challenging and progressive materials topic for future nanoelectronics and gives the chance to be at the leading edge in research and development of the proposed topic. Furthermore, the partners are be a perfect prototype for a new way of cooperation – the cooperation through virtual and remote labs – increasing efficiency and success of the contributing partners by using synergies, data, mind and personnel sharing, and decreasing costs by technology and equipment sharing. From the experiences gained from the work on Si1−y Ge y -alloys in the past one can state, that if Ge1−x Snx -alloys reveal as the new generation materials a long term research is waiting to be performed.
5 Related Work To the knowledge of the proposing authors there are not comparable joint activities on the proposed topics “Ge1−x Snx -alloys for nanoelectronics” and “virtual nanoelectronics lab” at the moment worldwide.
604
S. Jeschke et al.
A project similar to N ET L ABS is accomplished by the MIT in the USA: the project [HdAL+ 08, mit] supported also by Microsoft aims at making experiments remote-controlled, and thus having them accessible by web-services for other members over the internet. Very similar to I L ABS , N ET L ABS is not fixed to a certain topic, but addresses all engineering and natural scientific fields. Similar to I L ABS , we aim at a “single sign on” process to gain access to our resources. This process will be integrated into the Wonderland architecture. The Blekinge Institute of Technology in Sweden started in 2007 the VISIR project [GZH+ 07] Quite similar to N ET L ABS and iLabs, VISIR tries to increase laboratory utilization by sharing remote-controlled equipment across universities. Another similarity is that as many other projects [BI06, JRST07a] deploy the LabVIEW [nil] software by National Instruments to wire experiments to the internet. I L ABS
6 Roadmap Based on the already mentioned process flow the overall plan for MOSFETmanufacturing is defined as follows: After focusing on the epitaxial growth of doped Ge1−x Snx -films on GOIsubstrates or virtual Germanium-substrates on standard Silicon wafers in the 1st year – as a starting the Tin-concentration in the Germanium-matrix will be varied in the range 0 < x < 0.05. For n- or p-type doping Antimony or Boron will be used, respectively – a final doping concentration 1 · 1018 cm −3 should be obtained. The major task is deriving a growth recipe for the growth of n(p)-type doped epitaxial Ge1−x Snx -films at growth temperatures ≤ 600◦ C to avoid Tin-precipitation. This defines the first major milestone. Partners start producing and characterizing reference MOSFETs made from pure Germanium with a similar structure as depicted in Fig. 1, fabricated by the joint process flow mentioned above. The second major milestone of the 1st year is an established technology flow combining the labs fully operational for MOSFET manufacturing (the fabricated reference MOSFETs give the proof of concept). Within the 2nd year the first MOSFETs with Ge1−x Snx -channels with x < 0.05 are manufactured and characterized. Working transistors with Ge1−x Snx -channels with x < 0.05 are marking the first major milestone of the 2nd year. In parallel the epitaxial growth task is addressed to the Tin-concentration range 0.05 ≤ x < 0.1. Here one dominant topic the examination of increasing problem of the critical layer thickness for the mechanical relaxation of the strained, doped Ge1−x Snx -films as function of x will be. The second major milestone of the 2nd year is given by a recipe for the low-thermal epitaxial growth of strained Ge1−x Snx -films 0.05 ≤ x < 0.1 at growth temperatures ≤ 500◦ C with thicknesses smaller than the critical thicknesses. In the 3rd year the new recipe derived in the 2nd year will be used for the manufacturing of MOSFETs with Ge1−x Snx -channels with x < 0.1. Working transistors will again mark the first major milestone of the 3rd year. For epitaxy the last step is to try to increase the Tin-concentration in the Germanium-matrix up to x = 0.2. In case of success those films will be used for MOSFET-fabrication, too.
Crystalline Ge1−x Snx Heterostructures in Lateral High-Speed Devices
605
The overall plan for the cooperative project on virtual nanoelectronics lab creation is defined as follows: In the 1st year of the project, the 3D-infrastructure from the BW- E L ABS project is adapted to the projects need, investigating the integration of existing remote analysis tools and laboratories and determining the fundamental connection to the content management system realized by the scientific eSciDoc infrastructure. The project-specific components as experimental tools, rights management and life-cycle of data and documents are defined and integrated. A first prototype of one remotely-controllable tool is implemented. Existing metadata profiles are expanded, and new concepts for the semantic retrieval and searchability on the digital content, in particular the primary data, are investigated. Within the 2nd project year, additional remote analysis experiments and tools for data analysis are embedded. The entire process of data generation, data filing and access control are mapped and evaluated. Systems for distributed authentication (single sign on, shibboleth) are included. The first prototype integrating metadata profiles and a preliminary retrieval system for semantic searchability of primary data is tested exemplary. Networking with already existing information resources (digital library) is established. Within the 3rd year, the concepts of searchability and semantic retrieval of primary data are elaborated. Particular attention is given to a new model for data plausibility checking which covers the comparison between different but otherwise comparable experiment series, between different arbitrary experiments in order to find hidden similarities, and between theoretical or otherwise expected prognoses and the measured data. Additionally, the 3rd year is dedicated to consolidating infrastructure and services such that they can be integrated sustainably in production systems. Further laboratories are integrated, and community-building actions are strengthened (workshops, conferences, etc.). Technical support documents are provided, simplifying the integration of further experiments. Activities to widen the user-base that started at the beginning of the project are strengthened in the 3rd year.
7 Conclusion Even though this is an ambitious project for constructing an infrastructure for (virtual) experiments, we want to stress that it is not our aim to substitute the traditional experiment as its value lies beyond the gained scientific insight, namely in training the social communication skills with colleagues and fellow students. Even though we try to mimic these structures as far as possible in the virtual world, their replication remains necessarily incomplete. Instead, we hope to make the best out of the financial problems universities have to face – and establishes a strong federation of universities which, as a group, are able to support their scientists better than a single isolated institution could. • A more efficient and systematic development of new materials is promoted. • Constant obtainable material quality, that would be independent of the synthesizing chemist, permits a continuous reliable research on this material. This item is of
606
S. Jeschke et al.
particular importance since syntheses are usually developed by graduate students and/or Postdocs, who stay only for a limited period at their respective research institute and knowledge transfer does not always take the prominent role that it should be. • Established synthesis approaches are used also as exemplary sample experiments in teachings. are the long-term scientific advantages offered by an interactive platform sustainably linking diverse remote and virtual laboratories features.
References [ABB+ 08] K. Alberi, J. Blacksberg, L. D. Bell, S. Nikzad, K. M. Yu, O. D. Dubon, and W. Walukiewicz. Band anticrossing in highly mismatched SnxGe1–x semiconducting alloys. Phys. Rev., B 77, 2008. 073202. [BI06] H.A. Basher and S.A. Isa. On-campus and Online Virtual Laboratory Experiments with LabVIEW. In Proceedings of the IEEE SoutheastCon, pages 325–330. Springer, 2006. [BSE05] K. K. Bhuwalka, J. Schulze, and I. Eisele. “Scaling the Vertical Tunnel FET with Tunnel Bandgap Modulation and Gate Workfunction Engineering”. IEEE Transactions on Electron Devices, 52(5), 2005. (909). [DHJ+ 85] H. Daembkes, H.-J. Herzog, H. Jorke, H. Kibble, and E. Kasper. Technical Digest IEDM-1985, volume 28.7. 1985. [ESK04] I. Eisele, J. Schulze, and E. Kasper. Films by Molecular Beam Epitaxy, in Silicon – Evaluation and Future of a Technology. Springer Verlag, Berlin, 2004. [GZH+ 07] I. Gustavsson, J. Zackrisson, L. Håkansson, L. Claesson, and T. Lagö. The VISIR project – an Open Source Software Initiative for Distributed Online Laboratories. In Proceedings of the Annual Int. Conf. on Remote Engineering and Virtual Instrumentation, 2007. [HdAL+ 08] V.J. Harward, J.A. del Alamo, S.R. Lerman, P.H. Bailey, J. Carpenter, K. DeLong, C. Felknor, J. Hardison, B. Harrison, I. Jabbour, P.D. Long, M. Tingting, L. Naamani, J. Northridge, M. Schulz, D. Talavera, C. Varadharajan, W. Shaomin, K. Yehia, R. Zbib, and D. Zych. The iLab Shared Architecture: A Web Services Infrastructure to Build Communities of Internet Accessible Laboratories. Proceedings of IEEE, 96(6):931–950, June 2008. [HNE+ 02] J.L. Hoyt, H.M. Nayfeh, S. Eguchi, I. Aberg, G. Xia, T. Drake, E.A. Fitzgerald, and D.A. Antoniadis. Strained Silicon MOSFET Technology. Technical Digest IEDM, 2.1.1:23, 2002. [JBH+ 09a] S. Jeschke, B. Burr, J.-U. Hahn, L. Helmes, W. Kriha, M. Krüger, A.W. Liehr, W. Osten, O. Pfeiffer, Th. Richter, G. Schneider, W. Stephan, and K.-H. Weber. BWeLabs - Knowledge Management in Virtual and Remote Labs. In Proceedings of the INGRID - 4th Int. Workshop on Distributed Cooperative Laboratories: Instrumenting the Grid, Alghero, Italy, September 2009. Springer. [JBH+ 09b] S. Jeschke, B. Burr, J-U. Hahn, L. Helmes, W. Kriha, M. Krüger, A.W. Liehr, W. Osten, O. Pfeiffer, Th. Richter, G. Schneider, W. Stephan, and K.-H. Weber. Networking Resources for Research and Scientific Education. In Proceedings of the 3rd IEEE International Workshop on e-Activity, Daegu, Korea, May 2009. [JBH+ 09c] S. Jeschke, B. Burr, J.-U. Hahn, L. Helmes, W. Kriha, M. Krüger, A.W. Liehr, W. Osten, O. Pfeiffer, Th. Richter, G. Schneider, W. Stephan, and K.-H. Weber. Networking Resources for Research and Scientific Education in BW-eLabs. 3rd Int. Workshop on
Crystalline Ge1−x Snx Heterostructures in Lateral High-Speed Devices
607
advances in theory, systems, and applications for all electronic activities, May 2009. Daegu, Korea. [JBR09] S. Jeschke, D. Böringer, and Th. Richter. Library of Labs - Dissemination of Remote and Virtual Laboratories for Natural Sciences and Engineering, Proposal for eContentPlus call. April 2009. funded by the EC, 9 European partners. S. Jeschke, W. Osten, L. Helmes, G. Schneider, M. Krüger, A.W. Liehr, J-U. Hahn, [JOH+ ] and W. Kriha. BW-eLabs proposal, Fabrication and Properties of N-Channel SiGe/Si Modulation Doped Field-Effect Transistors Grown by MBE: Knowledge Management in Virtual and remote Laboratories. funded by the MWK BW (June 2009 – November 2011) , partner: Univ. Stuttgart, FIZ Karlsruhe, Univ. of Freiburg, Stuttgart Media Univ. [JPT06] S. Jeschke, O. Pfeiffer, and C. Thomsen. Vernetzung experimenteller Ressourcen in Forschung und Ausbildung für Nanotechnologien und Nanowissenschaft. pages 85– 89, GI Jahrestagung (1), 2006. [JPT08] S. Jeschke, O. Pfeiffer, and C. Thomsen. Networking Resources for Research and Scientific Education in Nanoscience and Nanotechnologies. In ICONN 2008 Int. Conference On Nanoscience and Nanotechnology (IEEE), Melbourne/Australia, February 2008. [JRST07a] S. Jeschke, T. Richter, H. Scheel, and C. Thomsen. On Remote and Virtual Experiments in eLearning in Statistical Mechanics and Thermodynamics. In Innovations in E-learning, Instruction Technology, Assessment, and Engineering Education, pages 329–333. 2007. [JRST07b] S. Jeschke, Th. Richter, H. Scheel, and C. Thomsen. On Remote and Virtual Experiments in eLearning in Statistical Mechanics and Thermodynamics. In Proceedings of the PerEL2007 - 3rd IEEE International Workshop on PervasivE Learning, New York/USA, March 2007. [JSR08] S. Jeschke, U. Sinha, and Th. Richter. Embedding Virtual and Remote Experiments into a Cooperative Knowledge Space. In The 2008 Frontiers in Education Conference (FIE 2008, IEEE), Saratoga/NY/USA, October 2008. [Kas00] E. Kasper. Properties of Silicon Germanium and SiGe:Carbon. EMIS Datareviews Series, Vol. 24, 2000. London INSPEC (IEE). [KC07] J. Kouvetakis and A.V.G. Chizmeshya. New classes of Si-based photonic materials and device architectures via designer molecular routes. J. Mater. Chem., 17:1649– 1655, 2007. [KLC+ 06] R. Krithivasan, Y. Lu, J.D. Cressler, J.S. Rieh, M.H. Khater, D. Ahlgren, and G. Freeman. Half-Terahertz Operation of SiGe HBTs. IEEE Elec. Dev. Lett., 27 (7):567–569, 2006. [KOL08] E. Kasper, M. Oehme, and J. Lupaca-Schomber. High Ge content SiGe alloys: Doping and contact formation. ECS Transactions, 16(10):893–904, 2008. [mit] iLabs: Internet access to real labs – anywhere, anytime. http://icampus.mit.edu/ iLabs/, last accessed 12/06/09. [MSM+ 01] S. Monfray, T. Skotnicki, Y. Morand, S. Descombes, M. Paoli, P. Ribot, A. Talbot, D. Dutarte, F. Leverd, Y. Le Friec, R. Pantel, M. Haond, D. Renand, M.-E. Nier, C. Vizioz, D. Louis, and N. Buffet. First 80 nm SON (Silicon-On-Nothing) MOSFETs with Perfect Morphology and High Electrical Performance. Technical Digest IEDM-2001, 29(7):645ff, 2001. [nil] NI LabVIEW - The Software That Powers Virtual Instrumentation - National Instruments. http://www.ni.com/labview/, last accessed 12/06/09. [NSM07] H. Nitta, M. Sakuraba, and J. Murota. Carbon Doping Effect on Strain Relaxation during Si1-xGexCy Epitaxial Growth on Si(100) at 500C. Semicond. Sci. Technol., 22, 2007. [pro] Project Wonderland. https://lg3d-wonderland.dev.java.net/, last access: 12/06/09.
608 [rem]
S. Jeschke et al.
Remote Farm. http://remote.physik.tu-berlin.de/farm/index.php?id=1&L=1, last access: 12/06/09. [Sch05] J. Schulze. Konzepte Silizium-basierter MOS-Bauelemente. Springer-Verlag, 2005. [Sor06] R. Soref. Present, and Future of Silicon Photonics. The Past, IEEE J. Selected Topics Quantum Electron, 12 (6):1678–1687, 2006. [SSM06] K. Sugawara, M. Sakuraba, and J. Murota. Thermal Effect on Strain Relaxation in Ge Films Epitaxially Grown on Si(100) Using ECR Plasma CVD. Thin Solid Films, 508:143, 2006. [SSM08] T. Seo, M. Sakuraba, and J. Murota. Impact of Ge Fraction Modulation upon Electrical Characteristics of Hole Resonant Tunneling Diodes with Si/Strained Si1xGex/Si(100) Heterostructure. Thin Solid Films, 517:110, 2008. [TCF+ 06] J. Tolle, A.V.G. Chizmeshya, Y.Y. Fang, J. Kouvetakis, V.R. D’Costa, C.W. Hu, and I.S.T. Tsong. Low temperature chemical vapor deposition of Si-based compounds via SiH3SiH2SiH3: Metastable SiSn/GeSn/Si(100) heteroepitaxial structures. Appl. Phys. Lett., 89, 2006. [THAP96] M.E. Taylor, G. He, H.A. Atwater, and A. Polman. Solid phase epitaxy of diamond cubic SnXGe1-X alloys. J. Appl. Phys., 80(8):4384–4388, 1996. [TSY+ 07] S. Takeuchi, A. Sakai, K. Yamamoto, O. Nakatsuka, M. Ogawa, and S. Zaima. Growth and structure evaluation of strain-relaxed Ge1–xSnx buffer layers grown on various types of substrates. Semicond. Sci. Technol., 22:231–235, 2007. [WHTG94] J. Welser, J.L. Hoyt, S. Takagi, and J.F. Gibbons. Strain Dependence of the Performance Enhancement in Strained-Si n-MOSFETs. Technical Digest IEDM, 15.2:373, 1994. [WOK+ 08] J. Werner, M. Oehme, O. Kirfel, K. Lyutovich, and E. Kasper. MBE growth of lowdefect Si layers highly doped with Sb. Thin Solid Films, 517:227–228, 2008. [YSM06] A. Yamada, M. Sakuraba, and J. Murota. Photo Detection Characteristics of Si/Si1xGex/Si p-i-n Diodes Integrated with Optical Waveguides. Thin Solid Films, 508, 2006.
Digitale Produktion via Enterprise Application Integration Tobias Meisen, Philipp Meisen, Daniel Schilberg, Sabina Jeschke
Zusammenfassung Produktionsprozesse in der heutigen Zeit werden zunehmend komplexer. Daher ist es notwendig solche Prozesse vor der eigentlichen Implementierung zu simulieren und zu testen. In der Vergangenheit wurde eine Vielzahl unterschiedlicher Simulationswerkzeuge, zur Simulation unterschiedlicher Prozesse und Verfahren, entwickelt. Die Simulation gesamter Produktionsprozesse ist jedoch bis heute nicht möglich. Der Grund hierfür liegt in der Inkompatibilität der unterschiedlichen Simulationswerkzeuge, die sich durch heterogene Formate und Datenmodelle auszeichnen. In dieser Arbeit wird die Architektur eines Frameworks vorgestellt, das es ermöglicht heterogene Simulationswerkzeuge miteinander zu koppeln. Die Funktionsweise des Frameworks wird anhand eines Beispiels demonstriert, in dem der Fertigungsprozess einer Linepipe simuliert wird. Dabei werden sechs unterschiedliche, heterogene Simulationswerkzeuge verwendet. Schlüsselwörter Enterprise Application Integration · Enterprise Service Bus · Informationsintegration · Framework
unternehmensweite Ressourcenplanung in den meisten Unternehmen mittels ERP Systemen [Gro], die EAI Konzepte zur Daten- und Applikationsintegration nutzen, realisiert. Auch der Einsatz von Business Intelligence (BI) Systemen zur prozessübergreifenden und wertschöpfungsorientierten Informationsgewinnung in Produktionsbetrieben [Pan] wird erst durch den Einsatz von Integrationsverfahren ermöglicht. Die einheitliche Betrachtung anderer Unternehmensprozesse als die der Geschäftsfunktionen, beispielsweise von Simulationsprozessen in der Forschung und Entwicklung, findet allerdings selten statt [SP09]. Dabei sind Fertigungsverfahren wie Schweiß-, Umform- oder Wärmebehandlungsprozesse, die bei der Produktion von Gütern durchlaufen werden, größtenteils durch Simulationen beschreibbar. Diese sind jedoch auf einzelne Prozesse spezialisiert und verfügen weder über standardisierte Schnittstellen noch standardisierte Datenformate. Es handelt sich um sogenannte Insellösungen. Zur durchgängigen Kopplung zu einem Simulationsprozess müssen die einzelnen Simulationsergebnisse daher derzeit manuell überprüft und für nachfolgende Simulationen angepasst werden. Dies erfordert einen hohen Zeitaufwand und birgt ein enormes Fehlerpotential. Außerdem ist die Nutzung gängiger EAI Lösungen für solche Integrationsvorhaben wegen unzureichender Unterstützung nicht möglich. So werden beispielsweise keine großen Datenmengen unterstützt, die aber in Simulationsprozessen charakteristisch sind. Weiterhin fehlen Standards, die Integrationsbemühungen im Simulationsumfeld vorantreiben. In diesem Beitrag wird ein Framework beschrieben, das die Simulation eines Fertigungsprozesses unter Verwendung bestehender Insellösungen ermöglicht. Im Fokus steht dabei nicht die technische Kopplung der Applikationen, die mittels moderner Middleware Techniken [Mye02][Ser02] gelöst werden kann, sondern die Integration der jeweiligen durch die Applikation generierten Daten. Das Framework wird im Rahmen des Exzellenzclusters „Integrative Produktionstechnik für Hochlohnländern“ im Teilprojekt „Integrierte Plattform für verteilte numerische Simulation“ entwickelt. Die Validierung des Frameworks erfolgt anhand der Simulation des Fertigungsprozesses einer Linepipe, die im Rahmen eines weiteren Teilprojektes des Exzellenzclusters durchgeführt wird. In Kapitel wird zunächst der Stand der Technik dargestellt, um in Kapitel den Anwendungsfall für das in diesem Beitrag vorgestellte Framework zu beschreiben. In Kapitel folgt die Beschreibung der Frameworkarchitektur, in Kapitel die Vorstellung der für den Anwendungsfall notwendigen Erweiterungen am Framework. Kapitel schließt diesen Beitrag mit einem Fazit ab.
2 Stand der Technik Datenintegration und damit auch EAI ist bereits seit den Achtzigern, spätestens aber seit den neunziger Jahren eines der wichtigsten Themen, wenn es um die Beantwortung applikationsübergreifender Fragestellungen geht [HAO06]. Grundsätzlich lassen sich heute eine Vielzahl verschiedener Datenintegrationsprodukte finden,
Digitale Produktion via Enterprise Application Integration
611
Abb. 1 Hauptbereiche der Datenintegration [Whi05]
welche in unterschiedlichen Anwendungsgebieten nutzen finden. Im Allgemeinen lassen sich diese in drei Hauptbereiche [Whi05] unterteilen (vgl. Abbildung 1): • Datenverdichtung (Data Consolidation) • Datenvereinigung (Data Federation) • Datenverbreitung (Data Propagation) Im operationalen Bereich wird der Bereich der Datenverbreitung zur applikationsübergreifenden Nutzung von Daten angewendet. Diese wird häufig durch EAI realisiert. Wie in [Whi05] dargestellt, fokussiert sich EAI im Wesentlichen auf Nachrichten und Business Transaktionen, also kleinere Datenmengen, die zwischen verschiedenen Applikation ausgetauscht werden. Ein modernes Architekturkonzept, dass im Rahmen dieses Beitrags aufgegriffen wird und zur Realisierung von EAI verwendet wird, ist der Enterprise Service Bus (ESB). Es wurde im Rahmen des Aufkommens von servicebasierten Ansätzen entwickelt [Cha04]. Grundgedanke von ESB ist die Bereitstellung von Diensten innerhalb eines Systems, ähnlich der Verwendung von Integration Brookern [Sch04]. Abbildung 2 zeigt den schematischen Aufbau eines ESB. Ein Dienst stellt eine fachliche oder technische Funk-
612
Tobias Meisen et al.
Abb. 2 Schema eines Enterprise Service Bus
tionalität zur Verfügung, mit der die Geschäftsprozesse unterstützt werden. Alle Dienste sind über den Integration Bus miteinander verbunden. Transformationsdienste stellen allgemeine Funktionen zur Verfügung, um Daten eines Formats und Modells in ein anderes Format und Modell zu überführen. Routingdienste hingegen dienen zur übertragung von Daten an andere Dienste. Adapter letztendlich nutzen Transformations- und Routingdienste, um über den Service Bus bereitgestellte Daten in das Format und Modell einer Anwendung zu überführen. Transformationsdienste fördern somit die Wiederverwendung von implementierten Datentransformationen. Der Vorteil einer ESB Lösung liegt in der losen Kopplung der einzelnen Dienste. Nachteilig ist jedoch die fehlende physische Datenkopplung [RD08]: Sollen erfasste Daten nachträglich ausgewertet oder analysiert (z.B. mittels Datenexplorationstechniken wie OLAP oder Data Mining) werden, so müssen diese erneut von einer Datenquelle ausgelesen und transformiert werden, wodurch eine historische oder Langzeit Datenauswertung nicht möglich ist. Zur Realisierung einer solchen einheitlichen Betrachtung über alle Daten hinweg müssen andere Bereiche der Datenintegration herangezogen werden (vgl. Abbildung 1). Eine Lösung für die Schaffung einer einheitlichen Sicht auf Daten ist die Datenvereinigung, die in dem Umfeld der Enterprise Information Integration (EII) untersucht wird. Hierbei ist es das Ziel, Daten aus unterschiedlichen Datenquellen unter einer uniformen Schnittstelle zu einen [Whi05][BH08]. Diese Schnittstelle kann dann genutzt werden, um Anfragen an diese Datenquellen zu formulieren und auszuwerten. Da die Anfrage auf den unterschiedlichen Datenquellen ausgeführt wird und keine Konsolidierung innerhalb einer Datenquelle erfolgt, handelt es sich bei diesem Ansatz um die Implementierung einer virtuellen Sicht auf die Daten. Hierzu ist es jedoch erforderlich, dass die Daten in verteilten Datenspeichern zur
Digitale Produktion via Enterprise Application Integration
613
Verfügung stehen und auf diese zugegriffen werden kann. Trifft dies nicht zu müssen Techniken der Datenverdichtung angewendet werden. Unter Datenverdichtung wird das Zusammenführen unterschiedlicher Daten in eine gemeinsame, einheitliche Datenstruktur verstanden. Mittels Extract Transform Load (ETL) wird ein gängiger Prozess für die Integration von Daten beschrieben [VSS02]. ETL umfasst die Extraktion der Daten aus einer oder mehreren, meistens operativen, Datenquelle, die Transformation des Datenformats und -modells in das Zielschema und das Laden in die Zieldatenbank.
3 Anwendungsfall Bei der Fertigung einer Linepipe kommen unterschiedliche Fertigungsverfahren zum Einsatz. Im Rahmen des Anwendungsfalls werden die Fertigungsverfahren durch für diese Verfahren spezialisierte Werkzeuge simuliert. Den zugrundeliegenden Fertigungsprozess stellt Abbildung 3 dar. Zunächst wird das Glühen, Warmwalzen und Abkühlen des Bauteils durch die von Access entwickelte Applikation CASTS simuliert. Anschließend wird das Spanen und Umformen mit Abaqus (SIMULIA) dargestellt. Das Schweißen und Expandieren der Linepipe wird in SimWeld, einer Entwicklung des ISF der RWTH Aachen und der Software SysWeld, eine Entwicklung der ESI-Group, simuliert [RMD07]. Außerdem findet eine Simulation der änderungen im Gefüge des Bauteils mit den Simulationen Micress [ANS98] und Homat [Las02], beide Entwicklungen von Access, statt. Insgesamt umfasst der Anwendungsfall somit sechs unterschiedliche Simulationen, denen jeweils unterschiedliche Formate und Modelle zugrunde liegen. Neben diesem Anwendungsfall werden im Rahmen des Projektes „Integrierte Plattform für verteilte numerische Simulation“ vier weitere, hier nicht näher
Abb. 3 Fertigungsprozess Linepipe
614
Tobias Meisen et al.
erläuterte, Anwendungsfälle untersucht, welche die Basis für die Anforderungen an das in diesem Beitrag vorgestellte Framework bilden. Diese Anforderungen wurden bereits in [SMH09] untersucht und beschrieben und in vervollständigt. Zentrale Anforderung ist dabei die Möglichkeit zur Datenverbreitung sowie die Notwendigkeit nach einer prozessorientierten Datenverdichtung (vgl. Abbildung 1) zur späteren Visualisierung und Analyse der im Prozess gesammelten Daten. Ein weiteres wichtiges Ziel ist es, neue Simulationsprozesse einfach abzubilden und neue Simulationswerkzeuge schnell einzubinden, ohne größere Anpassungen an der Anwendung vornehmen zu müssen.
4 Architektur des Frameworks 4.1 Systemarchitektur Basierend auf den in Kapitel geschilderten Anforderungen wurde die Architektur des Frameworks entwickelt. Sie ist angelehnt an das ESB Architekturkonzept, wobei die Möglichkeit zur Datenverdichtung durch Einführung eines zentralen Datenspeichers realisiert wurde [SGH08]. Die Systemarchitektur ist in Abbildung 4 dargestellt. Die Kommunikation mit den Applikationen erfolgt über eine Middleware. Im Rahmen des Anwendungsfalls wurde die anwendungsorientierte Middleware Condor [TTL05][CBKB08] verwendet. Routingdienste, wie in einem ESB üblich,
Abb. 4 Systemarchitektur des Frameworks
Digitale Produktion via Enterprise Application Integration
615
sind kein Bestandteil des Frameworks, da diese Funktionalität durch die Middleware gekapselt wird. Das Framework wird zur Realisierung der Integrationsebene verwendet. Dienstanbieter stellen unterschiedliche Dienste zur Integration, Extraktion und Transformation von Daten bereit und sind direkt mit dem Integration Bus verbunden. Die Anbindung erfolgt über die Java RMI Technologie [Gro01] und ist damit unabhängig vom verwendeten Betriebssystem. Einen zentralen Unterschied zum ESB Architekturkonzept stellt die Verwendung eines Integration Servers und eines Datenbank Servers dar. Der Integration Server empfängt alle eingehenden Daten von der Middleware, analysiert sie und stellt sie anschließend über den Integration Bus den einzelnen Dienstanbietern zur Verfügung. Diese greifen die benötigten Daten ab und verarbeiten sie weiter. Nach jedem abgeschlossenen Verarbeitungsschritt prüft der Integration Server die Konsistenz der Daten und ermittelt, welcher Bearbeitungsschritt für die gegebenen Daten als nächstes durchgeführt werden muss. Der Integration Server ist somit die zentrale Kontrollinstanz des Datenintegrations- und Datenextraktionsprozesses. Als zentraler Datenspeicher wird ein Datenbank Server verwendet, der ebenfalls mit dem Integration Bus verbunden ist. Sowohl die Dienstanbieter als auch der Integration Server greifen direkt auf den Datenbank Server zu. Hierdurch wird eine Datenverdichtung und somit eine nachgelagerte Analyse der über den Prozess gesammelten Daten ermöglicht.
4.2 Softwarearchitektur Das Framework umfasst die drei Hauptkomponenten: • Integration Server • Service Provider • Client die im Folgenden im Detail beschrieben werden. Zur Kommunikation zwischen den Komponenten existiert außerdem eine Codebase Komponente, in der komponentenübergreifende Funktionalität gekapselt ist. Die Hauptkomponenten sind in Abbildung 5 dargestellt und werden im Folgenden erläutert. Client
Abb. 5 Hauptkomponenten des Frameworks
616
Tobias Meisen et al.
Die Client Komponente dient zur Realisation der Kommunikation zwischen Middleware und Integration Server. Sie enthält Adapter, die eine Weiterleitung der Anfragen an den Integration Server über unterschiedliche Quellen, bspw. TCP/IP, RMI oder SOAP [KBM08], ermöglichen. Für einen konkreten Anwendungsfall, der nicht durch die bereits integrierten Technologien abgedeckt ist, bietet die Client Komponente Grundgerüste an, die zur Implementierung eines eigenen Adapters verwendet werden können (vgl. Kapitel 5). Integration Server Die Integration Server Komponente enthält die Implementierung eines Integration Servers. Ein Integration Server fungiert als Dienstanbieter für die einzelnen Clients und stellt Integrations-, Extraktions- und Konvertierungsdienste bereit. Für jeden dieser Dienste ist ein Dienstprozess im Integration Server hinterlegt. Stellt ein Client eine Anfrage, so wird der zugrundeliegende Dienstprozess gestartet und abgearbeitet. Ein Dienstprozess beschreibt in Form eines gerichteten Ablaufes, welche Dienste abgearbeitet werden müssen, um die vom Client angeforderte Funktionalität bereitzustellen. Die im Framework realisierten Dienstprozesse sind in Abbildung 6 dargestellt. Der Konvertierungsprozess wird durch den Integrations- und Extraktionsprozess, die beide auf einem erweiterten ETL Prozess basieren, definiert. Während dem Integrationsprozess eine Nachbearbeitung der integrierten Daten nachgelagert ist, wird bei dem Extraktionsprozess vor dem eigentlichen ETL Prozess eine Datenanreicherung durchgeführt. Die in Abbildung 6 grün markierten Ablaufschritte sind Funktionalitäten, die vom Dienstanbieter innerhalb der Integrationsebene bereitgestellt werden. Der Integration Server dient dabei als Vermittler [GHJ95]
Abb. 6 Dienstprozesse im Framework
Digitale Produktion via Enterprise Application Integration
617
Abb. 7 Datenschema zur prozessbezogenen Datenintegration
und vermittelt die entsprechenden Daten an denjenigen Dienstanbieter, der die geforderte Funktionalität und Kapazität bereitstellt. Somit sind die für die Datenintegration relevanten Algorithmen, die abhängig vom Anwendungsfall sind, in den konkreten Dienstanbietern gekapselt. Außerdem realisiert der Integration Server eine prozessbezogene Datenintegration. Der Integration Server überwacht die Zuordnung der Daten zu dem Prozessschritt und vermittelt den Prozesskontext an die Dienstanbieter. Das zugrundeliegende Datenschema ist in Abbildung 7 dargestellt. Ein Prozess wird darin durch einen Namen (Attribut Identifier) gekennzeichnet. Ebenso verfügt jeder Prozessschritt über einen frei wählbaren Namen (Attribut Identifier). Der Prozessablauf wird abgebildet, indem die Nachfolger eines Prozessschrittes gespeichert werden. Dabei kann ein Prozessschritt beliebig viele Nachfolger aufweisen, um komplexe Prozesse, wie beispielsweise die Bereitstellung von Simulationsdaten auf Makround Mikroebene, zu berücksichtigen. Service Provider Die Service Provider Komponente stellt die grundlegende Funktionalität für Dienstanbieter im Framework bereit. Die Implementierung eines konkreten Dienstanbieters hängt dabei vom Anwendungsfall ab. So basiert beispielsweise die Integration von FEM-Daten auf einem anderen Datenschema als die von Molekülstruktur-Daten, obwohl es sich um denselben physikalischen Gegenstand und vergleichbare physikalische Größem handelt. Das Framework bietet Schnittstellen zu gängigen ETL-Werkzeugen, wie beispielsweise dem Pentaho Data Integrator (PDI) [etl06]. Somit kann die Integration und Extraktion von Daten auf Basis dieser, im Umfeld von ETL bereits etablierten, Werkzeuge erstellt werden. Umgekehrt ist es jedoch auch möglich, andere Werkzeuge oder Frameworks zur Integration und Extraktion zu nutzen, falls dies im konkreten Anwendungsfall (vgl. Kapitel 5) sinnvoll und nötig ist. Wie in Abbildung 6 dargestellt unterstützt das Framework neben Diensten zur Bereitstellung eines ETL-Prozesses auch Dienste zur Nachbearbeitung und Anreicherung von Daten. Der Nachbearbeitungsdienst ermöglicht dabei beispielsweise die Implementierung von Plausibilitätskriterien, die von den integrierten Daten unabhängig von ihrer Ursprungsquelle erfüllt werden müssen. Diese können in einem Nachbearbeitungsdienst implementiert werden. Innerhalb der Anreicherung werden Datentransformationen durchgeführt, um die im zentralen Datenspeicher hinterlegten Daten so aufzubereiten, dass die an sie gestellten Bedingungen zur
618
Tobias Meisen et al.
Extraktion erfüllt sind. Innerhalb des Frameworks ist eine adaptive, auf Ontologien basierende, Datenanreicherung umgesetzt [SMH09]. Hierdurch kann basierend auf den gegebenen (Ist-Zustand) und den benötigten Daten (Soll-Zustand) eine Transformationskette bestimmt werden, die den Ist-Zustand in den Soll-Zustand überführt.
5 Anwendung des Frameworks Im Rahmen des in Kapitel beschriebenen Anwendungsfalls und den Anforderungen aus vier weiteren betrachteten Anwendungsfällen wurde eine auf dem Framework basierende Applikation implementiert. Dazu wurde zunächst ein Datenschema zur Speicherung der von den Simulationen gelieferten Daten definiert. Die im Anwendungsfall betrachteten Simulationen basieren auf der Finite-ElementeMethode [Ste10], so dass das Datenschema für diese Art von Daten ausgelegt wurde. Abbildung 8 zeigt einen Auszug aus dem zugrundeliegenden Datenschema in UML Notation. Dieses Datenschema wurde als Objektmodell in der Anwendung, wie auch in der relationalen Datenbank abgebildet. Die zentrale Entität in diesen Simulationen ist die Geometrie des zu simulierenden Bauteils, die aus Knoten, Zellen und Attributen besteht. Attribute verfügen über Attributwerte, die abhängig von der Attributklasse der gesamten Geometrie einzelnen Zellen oder Knoten zugeordnet sind. Die im Rahmen des Anwendungsfalls spezifizierten Integrationsdienste lesen die von der Simulation gegebenen Geometriedaten ein, transformieren sie in das zentrale Datenschema und laden das Ergebnis in die Datenbank. Die Extraktionsdienste
Abb. 8 Datenschema für FEM-Simulationen
Digitale Produktion via Enterprise Application Integration
619
hingegen reichern die Daten zunächst derart an, dass die Daten in einer für das Zielformat gültigen Form vollständig vorliegen. Anschließend lesen sie die Geometriedaten aus der zentralen Datenbank aus, transformieren sie in das geforderte Format und laden die Daten in die Zieldatei oder -datenbank. Dieses Vorgehen soll exemplarisch am Beispiel der Konvertierung von Simulationsergebnissen der Simulation CASTS in eine Eingabedatei der Simulation Abaqus erläutert werden. Die Simulation CASTS speichert ihre Simulationsergebnisse im Format des Visualization Toolkits (VTK) [SML04]. Daher wurde ein Integrationsdienst für VTK, basierend auf der vom VTK Entwickler bereitgestellten Programmierschnittstelle, in der Anwendung implementiert. Die Bereitstellung der Dateien und der Datenbankzugriff sind bereits im Framework realisiert. Außerdem wurde ein Extraktionsdienst für das Abaqus Eingabeformat entwickelt, wobei hier aufgrund einer fehlenden Programmierschnittstelle auf das erwähnte ETL Werkzeug PDI zurückgegriffen wurde. Der implementierte PDI Job ist in Abbildung 9 dargestellt. Zur Realisierung der adaptiven Datenanreicherung wurden verschiedene Datentransformationen für FEM-Daten in die Anwendung implementiert. Beispiele für solche Transformationen sind die Umwandlung von Attributeinheiten, die Ableitung von Attributen aus bestehenden Attributen, das Verschieben der Geometrie im Raum, die Veränderung von Zelltypen einer Geometrie (bspw. Hexaeder zu Tetraeder) oder die Neunummerierung von Knoten und Zellen. In Abhängigkeit von den vorliegenden Daten kann die Anwendung die vorhandenen Daten mittels der adaptiven Datenanreicherung so transformieren, dass eine Datenextraktion ermöglicht wird.
Abb. 9 Abaqus Extraktion als PDI Transformation
620
Tobias Meisen et al.
Abb. 10 Beispiel einer Datenanreicherung
Im Falle des oben genannten Beispiels wird so eine Neunummerierung der Knoten und Zellen veranlasst. Dies ist notwendig, da die Zell- und Knotennummerierung in VTK bei 0, in Abaqus hingegen bei 1 beginnt. Ebenso werden in CASTS Vektoren in ihre einzelnen Komponenten zerlegt und als Attributwerte von Knoten abgespeichert, wohingegen Abaqus die Angabe der vollständigen Vektoren erwartet. Die Anwendung ist aufgrund der Datenanreicherung in der Lage, diese Anforderung selbstständig zu erkennen und dementsprechend die Daten als vollständigen Vektoren und nicht als Zerlegung in seine einzelnen Komponenten bereitzustellen. Die Konvertierung von CASTS Daten in Abaqus Daten für den Anwendungsfall ist in Abbildung 10 vereinfacht dargestellt.
6 Fazit Die Entwicklung des in diesem Beitrag vorgestellten Frameworks stellt einen wichtigen Schritt in der Etablierung der virtuellen Produktion dar, da es ermöglicht wird, Fertigungsprozesse Schritt für Schritt unter Verwendung spezialisierter Werkzeuge zu simulieren. Diese ganzheitliche auf spezialisierten Werkzeugen basierende Simulation findet dabei ohne Datenverluste oder zeitintensive manuelle Datenübernahmen von einem Werkzeug ins nächste statt. Ermöglicht wird dies durch das vorgestellte Framework, welches einen Weg eröffnet, die bisher unabhängig voneinander entwickelten, für einzelne Fertigungsverfahren oder Methoden spezialisierten Simulationswerkzeuge miteinander zu koppeln und die während der Simulationen generierten Daten einheitlich und prozessorientiert zu integrieren. Neben der Integration weiterer Simulationswerkzeuge in die bestehende Anwendung ist es in einem nächsten Schritt erforderlich, die Domäne der betrachteten Simulationen um Maschinen- und Fabriksimulationen zu erweitern, um eine ganzheitliche Simulation von Fertigungsprozessen zu ermöglichen. Die größte Herausforderung liegt dabei in der Schaffung eines zentralen Datenschemas, welches es
Digitale Produktion via Enterprise Application Integration
621
ermöglicht, die Daten einheitlich und unter Berücksichtigung ihrer Bedeutung im Gesamtkontext der Fabrik-, Maschinen und Materialebene abzubilden. Aufgrund der vorgestellten Methodik ist es dabei nicht erforderlich die Anwendungen auf dieses Datenschema anzupassen. Dies realisiert die Integrationsanwendung, die, basierend auf dem Framework, zu entwickeln ist. Durch die einheitliche Datensicht und die detaillierte Datenaufzeichnung auf Prozessebene ermöglicht das Framework in Zukunft, die Ergebnisse unterschiedlicher Simulationsprozesse und Simulationswerkzeuge miteinander zu vergleichen. Weiterhin können nun Rückschlüsse auf Fehlerquellen gezogen werden, was in der Vergangenheit nur unter hohen Zeit- und Kostenaufwand realisierbar war. Um dies zu realisieren, werden als nächstes die interessanten Merkmale (Performance Indicators) identifiziert und anschließend innerhalb der Anwendung bereitgestellt werden. Die Herausforderungen liegen hier vor allem bei der Entwicklung der benötigten Datenexplorationsverfahren sowie den Visualisierungstechniken.
7 Danksagung Die vorgestellten Arbeiten wurden von der Deutschen Forschungsgemeinschaft DFG im Rahmen des Exzellenzclusters „Integrative Produktionstechnik für Hochlohnländer“ gefördert.
Literaturverzeichnis [ANS98]
[BH08] [CBKB08]
[Cha04] [CHKT05]
[etl06] [GHJ95] [Gro] [Gro01] [HAO06]
G. Antos, J. Neises, and L. Steinbach. Micro- Macrosimulation of casting processes. In 4ième école d’été de “Modélisation numérique en thermique”, C8 1-42, Porquerolles, 1998. Philip A. Bernstein and Laura M. Haas. Information integration in the enterprise. Communications of the ACM archive, 51(9):72–79, September 2008. Philippe Cerfontaine, Thomas Beer, Torsten Kuhlen, and Christian Bischof. Towards a Flexible and Distributed Simulation Platform. In Lecture Notes in Computer Science (Springer), Computational Science and Ist Applications - ICCSA 2008, Proceedings Part 1, pages 867–882. Springer Berlin / Heidelberg, 2008. David A. Chappell. Enterprise Service Bus: Theory in Practice. O’Reilly Media, 2004. Stefan Conrad, Wilhelm Hasselbring, Arne Koschel, and Roland Tritsch. Enterprise Application Integration: Grundlagen - Konzepte - Entwurfsmuster - Praxisbeispiele. Spektrum Akademischer Verlag, 1st edition, 2005. ETL White Paper - Advanced ETL With Pentaho Data Integration, Breadboard BI. Technical report, 2006. Erich Gamma, Richard Helm, and Ralph E. Johnson. Design Patterns. Elements of Reusable Object-Oriented Software. Addison-Wesley Longman, Amsterdam, 1995. N. Gronau. Enterprise Resource Planning. Oldenbourg, 2nd edition. William Grosso. Java RMI. O’Reilly Media, 1st edition, 2001. Alon Halevy, Rajaraman Anand, and Joann Ordille. Data integration: the teenage years. In VLDB’2006: Proceedings of the 32nd international conference on Very large data bases, pages 9–16, 2006.
622 [KBM08] [Las02]
[Mye02] [Pan]
[RD08] [RMD07]
[Sch04] [Ser02] [SGH08]
[SMH09]
[SML04] [SP09] [Ste10] [TTL05]
[VSS02]
[Whi05]
Tobias Meisen et al. Vipul Kashyap, Christoph Bussler, and Matthew Moran. The Semantic Web: Semantics for Data and Services on the Web. Springer, Berlin, 1st edition, 2008. Gottfried Laschet. Homogenization of the thermal properties of transpiration cooled multi-layer plates. Computer Methods in Applied Mechanics and Engineering, 191(41-42):4535–4554, 2002. Judith M. Myerson. The Complete Book of Middleware. Auerbach Publications, 1st edition, 2002. Z. Panian. Supply chain intelligence in e-business environment. In Proceedings of the 9th WSEAS International Conference on Computers, Athens, Greece. Article No.: 85. Tijs Rademakers and Jos Dirksen. Open-Source ESBs in Action: Example Implementations in Mule and ServiceMix. Manning Publications, 2008. E. Rossiter, O. Mokrov, and U. Dilthey. Integration des Simulationspaketes SimWeld in FEM-Analyseprogramme zur Modellierung von Schweißprozessen. In Sysweld Forum, Weimar, Deutschland, 2007. Roy W. Schulte. Predicts 2003: Enterprise Service Buses Emerge. Technical report, 2004. Gartner. Daniel Serain. Middleware and Enterprise Application Integration: The Architecture of E-Business Solutions. Springer, 2nd edition, 2002. D. Schilberg, A. Gramatke, and K. Henning. Semantic interconnection of distributed numerical simulations via SOA. In Proc. of the World Congress on Engineering and Computer Science, pages 894–897, Hong Kong, 2008. Newswood Limited. D. Schilberg, T. Meisen, and K. Henning. Ontology Based Semantic Interconnection of Distributed Numerical Simulations for Virtual Production. In Proceedings of the IEEE 16th International Conference on Industrial Engineering and Engineering Management (IE&EM), Beijing (PR China)., 2009. Will Schroeder, Ken Martin, and B. Lorensen. The Visualization Toolkit. Kitware Inc., 3rd edition, 2004. G.J. Schmitz and U. Prahl. Toward a Virtual Platform for Materials Processing. JOM Journal of the Minerals, Metals and Materials Society, 61(5):19–23, 2009. Peter Steinke. Finite-Elemente-Methode: Rechnergestützte Einführung. SpringerVerlag Gmbh, 3rd edition, 2010. Douglas Thain, Todd Tannenbaum, and Miron Livny. Distributed Computing in Practice: The Condor Experience. Concurrency and Computation: Practice and Experience, 17:2–4, April 2005. Panos Vassiliadis, Alkis Simitsis, and Spiros Skiadopoulos. Conceptual Modeling for ETL Processes. In Proceedings of the 5th ACM international workshop on Data Warehousing and OLAP, pages 14 – 21, 2002. C. White. Data Integration: Using ETL, EAI, and EII Tools to Create an Integrated Enterprise. Technical report, 2005.
Dynamische Gruppenarbeit Dezentrale Feinplanungsunterstützung durch einen virtuellen Marktplatz Thilo Münstermann, Jens Völzke, Paul Flachskampf
für Gruppen mit wechselnder Zusammensetzung. Für diese Form der Arbeitsaufteilung fehlen daher Informationssysteme, mit denen Gruppen bei der operativen und gruppenübergreifenden Ressourcensteuerung und Auftragsfeinplanung unterstützt werden und dabei die leistungsgerechte Entlohnung berücksichtigen. Ein virtueller Marktplatz soll hier die bestehenden ERP/PPS-Systeme unterstützen und ergänzen. Mit der zusätzlichen Möglichkeit von dezentraler Feinplanung können bestehende Systeme effizienter und zeitsparender genutzt werden. In Praxisbeispielen konnte die Auftragsdurchlaufzeit dadurch um bis zu 32% reduziert werden [LM05]. Eine weitere mögliche Motivation für die Bildung von autonomen Arbeitsgruppen in der Fertigung, liegt in der zu beobachtenden Steigerung der Komplexität und Dynamik der Unternehmensumwelt [Mal04]. Die Unternehmen erkennen, dass Sie häufig durch die Einführung von starren Planungssystemen eher Flexibilität in der Fertigung verloren haben. Dieses Ungleichgewicht zwischen Innen- und Außenkomplexität schwächt die Unternehmen [Ash56]. Um zu überleben muss in den Unternehmen dieser Entwicklung entgegengesteuert werden. Oft führt Gruppenarbeit in Betrieben aber nicht zum gewünschten Erfolg, da sie nach kurzer Zeit wieder „einschläft“ oder Gruppenzuteilungen oder Arbeitgruppen an sich ganz ignoriert werden [FU03]. Grund dafür ist häufig die Vernachlässigung der aufwändigen und komplizierten Absprachen und Planungen. Ein Arbeitstool was hier ansetzt kann nicht nur die Effizienz der Planung und Gruppenarbeit erhöhen sondern auch dazu beitragen sie anwenderfreundlicher zu gestalten und Kommunikation und Planung aufrecht zu erhalten, um so den eingeführten Gruppenarbeitsprozess zu erhalten [AMF08]. Um die Möglichkeiten der dezentralen Feinplanung durch Gruppenarbeit stärker ausschöpfen zu können, sind eventuelle besondere Rahmenbedingungen, insbesondere die begrenzten personellen und finanziellen Ressourcen der Unternehmen zu beachten. Die Reaktion auf variierende Auftragslagen und Abrufschwankungen erfordert beispielsweise dynamische Gruppen, deren Größe und Qualifikation der Mitglieder sich schnell verändern. Dazu gehören einfache und kostengünstige Möglichkeiten zur Messung und Kommunikation der Gruppenleistung sowie Konzepte zur Leistungsbewertung und Personaleinsatzplanung in den Gruppen bei häufigen Gruppenumbesetzungen. Letztere werden erforderlich, wenn z. B. selten eingesetztes Spezialwissen bei einzelnen Mitarbeitern vorhanden ist und in unterschiedlichen Gruppen benötigt wird. Aus dem Bedarf nach Zugehörigkeit einzelner Mitarbeiter zu verschieden Gruppen entsteht eine weitere Anforderung. Die Leistung einzelner Facharbeiter in den Gruppen muss erfassbar sein und mit den Entgeltsystemen gekoppelt werden können. Anforderungen an zeitgemäße Entlohnungssysteme sind Transparenz und Leistungsbezogenheit, Motivierung, Gruppenarbeitsförderung, der wirtschaftliche Einsatz von Ressourcen und die Produktionsverbesserung [HG96]. Im Rahmen eines von der Arbeitsgemeinschaft industrieller Forschungsvereinigungen (AiF) aus Mitteln des BMWi geförderten Forschungsvorhabens entstand in Kooperation des Instituts für Unternehmenskybernetik e.V. und des Instituts für Integrierte Produktion Hannover gGmbH ein Konzept sowie eine beispielhafte Umsetzung eines solchen Informationssystems, im folgenden Arbeitsgruppenassistent (AGA) genannt. Der Fokus des vorliegenden Artikels liegt auf der Beschreibung
Dynamische Gruppenarbeit
625
eines virtuellen Marktplatzes zur Ressourcenverhandlung zwischen Gruppen. Um diese Idee in das Gesamtkonzept einordnen zu können, wird im Folgenden zunächst die grundlegende Funktionsweise kurz vorgestellt. Eine detaillierte Beschreibung des Feinplanungskonzeptes und der zugrundeliegenden Methoden findet sich bei [HFNM09].
2 Vorgehensweise und erreichte Ergebnisse Zu Beginn wurde eine IST-Analyse zum Stand der Gruppenarbeit, der Anforderungen an die Flexibilität und den bestehenden Software-Systemen in kleinen und mittelständischen Unternehmen der Metallverarbeitung durchgeführt und durch eine Technikfolgenabschätzung ergänzt (Flachskampf et al. 2008). Anschließend wurden die benötigten Methoden zur Messung von Zielgrößen, zur Erfassung von Restriktionen und zum Kapazitätsabgleich zwischen Gruppen theoretisch entwickelt [HFNM09], S.20–34. Die Implementierung und Realisierung des AGA erfolgte in einem Prozess der agilen Softwareentwicklung [Wes00]. Dabei wurden beide Pilotanwender von Anfang an intensiv in die Entwicklung einbezogen. Die agile Softwareentwicklung ist durch mehrere Iterationszyklen gekennzeichnet. Jeder Zyklus besteht aus Befragung, Anforderungsanalyse, Umsetzung und Testen. Demzufolge wurde der AGA im Projekt iterativ und inkrementell entwickelt. Die Entwicklung des Softwaretools wurde in mehrere gleichartige Zwischenschritte aufgeteilt, die jeweils eine lauffähige Version mit gestiegener Gesamtfunktionalität zum Ergebnis hatten. Die Anwendbarkeit der entwickelten Methoden wurde durch zwei Pilotanwender des Projektbegleitenden Ausschusses verifiziert. Um auch Unternehmensdaten verwenden zu können, wurde eine Verknüpfung zwischen dem AGA und vorhandenen Planungssoftwaresystemen implementiert. So konnten Daten, z. B. die Terminierung und die Grobplanung, als Grundlage für die Feinplanung genutzt werden. Der AGA wurde abschließend in den Pilotunternehmen eingeführt. Da eine agile Vorgehensweise bei der Entwicklung des Demonstrators gewählt wurde und die Beteiligten von Anfang an stark involviert waren, war keine intensive Schulung der Administratoren und Mitarbeiter oder aufwendige Einführungsmaßnahmen notwendig. Die Erfahrungsergebnisse der Pilotanwender wurden für die Verbesserung der Methoden und des AGA genutzt.
3 Dynamische Gruppenwechsel durch einen virtuellen Marktplatz 3.1 Überblick Der AGA ist als Erweiterung zentraler Produktionsplanungssysteme konzipiert und entweder als eigenständige Software über Schnittstellen mit diesen koppelbar oder als Modul in bestehende Systeme zu integrieren. Zur gruppenbasierten Feinplanung
626
Thilo Münstermann et al.
bietet der AGA dezentrale Planungstabellen für einzelne Arbeitsgruppen. Diese werden zentral gespeichert und sind über eine Konfliktbehandlung miteinander vernetzt, so dass gruppenübergreifende Auftragsplanung möglich bleibt, obwohl jede Gruppe zunächst mit ihrer eigenen Feinplanungstabelle arbeitet. Die Konzeption der Software unterscheidet Mitarbeiter, welche eine Liste von Qualifikationen besitzen können, und Maschinen, welche durch Funktionen charakterisiert sind. Einzelne Arbeitsschritte sind über die zur Bearbeitung erforderlichen Qualifikationen und Funktionen definiert. Im Laufe des Planungsprozesses werden ihnen Zeiträume und konkrete Mitarbeiter und Maschinen zugewiesen. So werden auftrags-, mitarbeiter- und maschinenzentrierte Planungsvisualisierungen ermöglicht [HFNM09], S. 71 f.. Ein zentrales Konzept der Feinplanungstabellen bilden dynamische Pools, welche für einen Auftragsschritt in einem Zeitraum nur solche Ressourcen aus der Gruppe anzeigen, welche die benötigte Qualifikation und Funktion aufweisen, sowie freie und verbuchte Ressourcen farblich unterscheiden.
3.2 Idee des virtuellen Marktplatzes Eine Grundidee des AGA besteht in der Dezentralisierung und damit der Dynamisierung von Feinplanungsprozessen. Gruppen sollen eigenverantwortlich innerhalb bestimmter Grenzen ihre Planung optimieren können und aufgrund der höheren übersichtlichkeit einer Gruppenplanung schnell und flexibel auf änderungen reagieren können. Wie jede Dezentralisierung bringt dies aber auch den Nachteil mit sich, dass betriebsweite Potenziale der Zusammenarbeit zunächst nicht genutzt werden, da keine zentrale Planungsstelle die innerhalb der Feinplanung benötigten oder frei werdenden Ressourcen gruppenübergreifend zuweist. Eine solche zentrale Zuweisung wäre auf dem Detaillierungsgrad einer Feinplanung mit enormem Arbeitsaufwand verbunden und würde gleichzeitig dem Prinzip der Gruppenautonomie widersprechen. Um dennoch Möglichkeiten zur Planungsoptimierung (z. B. Verkürzung der Durchlaufzeit oder Einhalten von Lieferterminen) nutzen zu können wird im Folgenden das Konzept des virtuellen Marktplatzes eingeführt, auf dem Gruppen benötigte und frei gewordene Ressourcen sowie Aufträge tauschen, verschieben und handeln können. Das Konzept des virtuellen Marktplatzes beruht auf der Idee, dass aus der Grobplanung eine vorläufige Zuweisung der Aufträge oder Arbeitsschritte auf die einzelnen Gruppen in der Fertigung vorgenommen wurde und diese Gruppen in ihrer Zusammenstellung und Personalstärke für die durchschnittliche Arbeitsbelastung optimiert sind. Aufgrund von Krankheiten, Maschinenausfällen, unvorhergesehenen Problemen oder zwischengeschobenen Aufträgen hoher Priorität kann es jedoch immer wieder vorkommen, dass von dieser Planung abgewichen werden muss. Nach der Grundidee des AGA werden diese Abweichungen in der dezentralen Feinplanung der Gruppen behandelt. So können bei den einzelnen Gruppen Kapazitätsengpässe und -überschüsse auftreten, die weiteres Optimierungspotenzial bieten. Um dieses Potenzial zu nutzen
Dynamische Gruppenarbeit
627
Abb. 1 Verhandlungsebenen zwischen Gruppen [HFNM09]
soll den Gruppen die Möglichkeit geboten werden, zusätzliche Mitarbeiter oder Maschinen für bestimmte Zeiträume anzufragen oder Arbeitsschritte zur Bearbeitung an andere Gruppen anbieten und abgeben zu können. Abbildung 1 stellt das Prinzip dar und deutet die jeweilige Verhandlungsrichtung aus Sicht der Gruppe A an. Diese Verhandlungen können je nach Bedarf mehr oder weniger vollständig über den virtuellen Marktplatz vollzogen werden. In einer relativ kleinen Produktion mit wenigen Gruppen und Verantwortlichen wird die Nachfrage, ob man einen Mitarbeiter aus einer Gruppe für ein paar Stunden ausgeliehen haben kann, meist persönlich erfolgen. Bei einer größeren Produktion, besonders wenn diese über mehrere Hallen verteilt ist, kann die Anfrage nach einer benötigen Qualifikation für einen Zeitraum über den virtuellen Marktplatz gestellt, von anderen Gruppen beantwortet und schließlich beidseitig bestätigt werden. Die Abläufe zur Anfrage von Qualifikationen und Funktionen sowie dem Anbieten von Arbeitsschritten werden in den folgenden drei Kapiteln ausführlich beschrieben. Im daran anschließenden Kapitel wird eine weitere Ergänzungsmöglichkeit der Idee des virtuellen Marktplatzes, die Versteigerung von Aufträgen und Ressourcen, beschrieben.
3.3 Mitarbeiter anfragen Gründe für den Bedarf einer Gruppe nach einem temporären Mitarbeiter können z. B. Krankheit oder Verzögerung in vorgelagerten Arbeitsschritten sein. Es stehen zwar die benötigten Arbeitsplätze und das Material zur Verfügung und alle Vorgängerarbeitschritte sind erledigt, aber es ist keiner oder zu wenige Mitarbeiter mit der benötigten Qualifikation anwesend. Dies kann sowohl kurzfristig und damit akut, aber auch mehrere Tage im Voraus ersichtlich auftreten. Oft wird nun innerhalb einer Gruppe vielleicht umgeplant, andere Tätigkeiten werden vorgezogen oder es wird ein Mitarbeiter eingesetzt, der diese Tätigkeit nur rudimentär beherrscht. über den virtuellen Marktplatz kann die Gruppe hingegen anfragen, ob eine andere Gruppe für einen Zeitraum t einen Mitarbeiter mit der Qualifikation Q entbehren kann. Es wird also eine Qualifikation für einen Zeitraum angefragt und nicht ein
628
Thilo Münstermann et al.
Mitarbeiter direkt. Prinzipiell kann eine Anfrage speziell an eine Gruppe gerichtet werden oder an alle Gruppen. Im letzteren Fall erreicht die Anfrage alle Gruppen, die einen Mitarbeiter mit dieser Qualifikation haben. Im ersteren Fall unterstützt der AGA falls gewünscht bei der Gruppenauswahl. Sind Qualifikation und Zeitraum ausgewählt, werden die Feinplanungen der anderen Gruppen überprüft, ob eine Gruppe tatsächlich einen Mitarbeiter mit dieser Qualifikation für den Zeitraum t noch nicht verplant hat. Werden mehrere oder keine Gruppen gefunden, die diese Qualifikation nach ihrer derzeitigen Feinplanung sofort zur Verfügung stellen könnten, wird nach der Methode zum Kapazitätsabgleich zwischen Gruppen die Flexibilität der Gruppen verglichen und entsprechend die Gruppe mit der höchsten Flexibilität ausgewählt [HFNM09], S. 32 ff.. Alternativ zu einem festen Zeitraum, kann auch eine Stundenzahl innerhalb eines Zeitraumes angefragt werden. Bei der zweiten Variante ist die Chance auf eine Zusage höher, da einer ausleihenden Gruppe B mehr Flexibilität verbleibt. Auf eine Anfrage antwortet eine andere Gruppe immer entweder mit einer Ablehnung, oder mit einem konkreten Mitarbeiter und einem festen Zeitraum. Der Zeitraum kann sich dabei vom angefragten Zeitraum unterscheiden und somit ein Gegenangebot bilden. Wurde kein fester Zeitraum angefragt oder der Zeitraum von Gruppe B verändert zurückgeschickt, ist die Verhandlung erst durch eine erneute Bestätigung durch Gruppe A abgeschlossen. Auch wenn die Anfrage an mehrere Gruppen gesendet wurde, ist eine abschließende Bestätigung durch Gruppe A erforderlich. Andernfalls kann die Verhandlung bereits nach dem zweiten Schritt von Gruppe B abgeschlossen werden, wenn die Anfrage unverändert besteht. Nach einer erfolgreichen Verhandlung einer Mitarbeiterausleihe werden die Rahmendaten in eine Ausleihtabelle eingetragen. Diese Tabelle enthält immer Zeiträume, die minutengenau abgegrenzt sind, einen konkreten Mitarbeiter, eine verleihende und eine ausleihende Gruppe: von dd.mm.yyyy hh.mm | bis dd.mm.yyyy hh.mm | [mitarbeiter] | von [gruppe] | an [gruppe] Die Filter der dynamischen Pools und die Konfliktbehandlung der Feinplanung berücksichtigen diese Tabelle. Ausgeliehene Mitarbeiter werden in den Pools der Gruppe A für diesen Zeitraum mit eingeplant und können Arbeitsschritten, die innerhalb dieses Zeitraums liegen, zugewiesen werden. Die Konfliktbehandlung überprüft, ob Arbeitsschritte mit einem ausgeliehenen Mitarbeiter für Zeitraum t außerhalb dieses Zeitraumes verschoben werden und zeigt ggf. einen Konflikt an. Der Konflikt wird auch angezeigt, wenn der Mitarbeiter theoretisch nach der Feinplanung von Gruppe B Zeit hätte, um anzuzeigen, dass die im virtuellen Marktplatz getroffene Vereinbarung verletzt worden ist. In der mitarbeiterzentrierten Planung erscheint für den Ausleihzeitraum bei Gruppe A ein neuer Mitarbeiter, bei Gruppe B wird der Zeitraum für den Mitarbeiter grau hinterlegt. Zur Kennzeichnung ausgeliehener Mitarbeiter kann eine zusätzliche farbliche Markierung konfiguriert werden, oder ein beliebiges Symbol als Grafikdatei im AGA eingebunden werden.
Dynamische Gruppenarbeit
629
3.4 Maschinen anfragen Gründe für den Bedarf einer Gruppe nach zusätzlichen Maschinen können z. B. Ausfälle oder ungeplante Wartungsarbeiten sein. Wenn genügend Arbeitsschritte zur Bearbeitung freigegeben sind und beispielsweise genug personelle Ressourcen zur Verfügung stehen, um mehrere Arbeitsschritte parallel erledigen zu können, macht es Sinn bei anderen Gruppen nach Maschinen bzw. Arbeitsplätzen zu fragen. In diesem Fall kann eine Gruppe A über den virtuellen Marktplatz eine Funktion F für einen Zeitraum t bei einer Gruppe B anfragen. Auch bei dieser Anfrage kann ein fester Zeitraum oder eine Stundenzahl innerhalb eines Zeitraums für die Anfrage verwendet werden. Die Gruppenauswahl wird analog zur Anfrage von Mitarbeitern durch den AGA unterstützt, lediglich die Flexibilitätsbewertung von Gruppen entfällt, da diese im AGA lediglich für Qualifikationen stattfindet. Nach erfolgter Anfrage kann Gruppe B die Anfrage ablehnen, unverändert bestätigen oder ggf. den Zeitraum anpassen und als Gegenangebot zurück senden. Nach einer erfolgreichen Verhandlung wird die Ausleihe einer Maschine in eine Ausleihtabelle wie folgt eingetragen: von dd.mm.yyyy hh.mm | bis dd.mm.yyyy hh.mm | [maschine] | von [gruppe] | an [gruppe] Auch diese Ausleihtabelle wird von der Feinplanung berücksichtigt. Ausgeliehene Maschinen werden im Maschinenpool der Gruppe A für diesen Zeitraum angezeigt und können Arbeitsschritten, die innerhalb dieses Zeitraums liegen, zugewiesen werden. Die Konfliktbehandlung überprüft, ob Arbeitsschritte mit einer ausgeliehenen Maschine für Zeitraum t außerhalb dieses Zeitraumes verschoben werden und zeigt ggf. einen Konflikt an. Auch dieser Konflikt wird unabhängig davon angezeigt, ob die Maschine tatsächlich für den Zeitraum verplant ist, sondern zeigt lediglich an, dass der Zeitraum der Ausleihe überschritten wurde. In der maschinenzentrierten Planung erscheint für den Ausleihzeitraum bei Gruppe A eine neue Maschine, bei Gruppe B wird der Zeitraum grau hinter legt. Zur Kennzeichnung ausgeliehener Maschinen kann eine zusätzliche farbliche Markierung konfiguriert werden, oder ein beliebiges Symbol als Grafikdatei in den Arbeitsgruppenassistenten eingebunden werden.
3.5 Arbeitsschritte anbieten Als Alternative zum Anfragen von Mitarbeitern und Maschinen können Gruppen bei Kapazitätsengpässen auch Arbeitsschritte an andere Gruppen abgeben. Im Gegensatz zu den beiden zuvor beschriebenen Verhandlungsszenarien, bittet die Gruppe A dabei nicht um zusätzliche Ressourcen, sondern möchte Arbeitsschritte abgeben. Während bei den zuvor beschriebenen Ausleihen stets eine abstrakte Qualifikation und Funktion angefragt wurde, wird in diesem Fall ein konkreter Arbeitsschritt angeboten. Es wird also kein Arbeitsschritttyp angeboten, sondern ein konkreter
630
Thilo Münstermann et al.
Arbeitsschritt eines Auftrags. Dies bietet in diesem Fall den Vorteil, dass spezifische Kommentare zu dem Arbeitsschritt hinterlegt sein können und bei Bedarf auch bereits eine Maschine oder ein Mitarbeiter zugewiesen sein können. Als Zeitbegrenzung werden ein frühester möglicher Starttermin und ein spätester möglicher Endtermin angegeben. Die entstandene Anfrage wird an alle Gruppen gesendet, die theoretisch die im Arbeitsschritt geforderten Qualifikationen und Funktionen besitzen. Die Gruppe, welche als erstes die Anfrage bestätigt, übernimmt den Arbeitsschritt. Auch bei dieser Variante der Verhandlung kann ein Gegenangebot abgegeben werden, indem Start und/oder Endtermin angepasst werden. Dies ist z. B. sinnvoll, wenn Gruppe B den Auftrag zwar übernehmen könnte, aber nur dann, wenn sie schon ein paar Stunden früher als in der Anfrage angegeben, damit starten könnten. Gruppe A kann dann ihre Planung überprüfen, ob sie es schaffen alle Vorgängerschritte rechtzeitig fertig zu stellen. Die Verhandlung ist in diesem Fall erst abgeschlossen, wenn Gruppe A die änderung des Zeitfensters akzeptiert und bestätigt. Nach einer erfolgreichen Verhandlung wird die übernahme eines Arbeitsschrittes in eine Ausleihtabelle eingetragen. Ausgeliehene Arbeitsschritte erscheinen anschließend in der Feinplanungstabelle der Zielgruppe und verschwinden aus der Feinplanung der ursprünglichen Gruppe. Die Konfliktbehandlung erkennt bei Verschieben des Arbeitsschrittes sowohl, ob der Auftragsschritt mit seinen Vorgängern oder Nachfolgern zeitlich überschneidet, als auch, ob der verhandelte Zeitrahmen verlassen wird. Bei der maschinen- und mitarbeiterzentrierten Planung erscheint der Arbeitsschritt wie alle anderen in der Planungstabelle. Zur Kennzeichnung ausgeliehener Arbeitsschritte kann eine zusätzliche farbliche Markierung konfiguriert werden, oder ein beliebiges Symbol als Grafikdatei in den Arbeitsgruppenassistenten eingebunden werden.
3.6 Versteigerungen über die oben beschriebenen Ausleihen hinaus, kann der virtuelle Marktplatz für Versteigerungen genutzt werden. Die Idee dahinter ist, dass zentral von der Unternehmensleitung oder der Produktionsplanung Aufträge den Gruppen angeboten werden. Dies ist als Ergänzung zu der festen Zuweisung von Aufträgen zu Arbeitsgruppen gedacht. Anstatt die Aufträge auf die Gruppen zu verteilen, kann auch ein Wettbewerb der Gruppen um die Aufträge gefördert werden. Im virtuellen Marktplatz werden dazu von zentraler Stelle neue Aufträge eingestellt. Die Aufträge enthalten bereits alle notwendigen Angaben, damit die Gruppenleiter sie in eine fiktive Planung integrieren können. Diese Kalkulation kann im AGA bereits vor der Zuteilung gemacht werden um die Abschätzungen realistischer zu machen. Es werden zwei Arten von Versteigerungen unterschieden. Versteigerung nach Abgabetermin und Versteigerung nach niedrigstem Kostenvoranschlag. Die zweite Variante ist damit an ein monetäres Bonussystem gekoppelt, bei der die Gruppen für jeden übernommenen Auftrag einen bestimmten Bonusbetrag erhalten. Beide Arten der Versteigerung laufen so ab, dass die Gruppen bis zu einem vorher festgelegten Termin Angebote abgeben können. Nach Ablauf des Termins kann der AGA
Dynamische Gruppenarbeit
631
entweder automatisch den Auftrag an das beste Angebot (frühester Liefertermin oder niedrigster Geldbetrag) vergeben oder eine manuelle Zuteilung findet durch einen Entscheider statt. Je nachdem wie die Gruppen in einem Unternehmen eingeteilt sind, macht die Versteigerung von ganzen Aufträgen an Gruppen keinen Sinn, z. B. wenn jeder Auftrag in der Regel von mehreren Gruppen bearbeitet werden muss. In diesem Fall können genau wie ganze Aufträge auch einzelne Auftragsschritte versteigert werden. Der koordinatorische Aufwand bei hoher Abhängigkeit der Schritte untereinander steigt jedoch, da sichergestellt werden muss, dass die zugesagten Termine zusammen passen und der Auftrag insgesamt termingerecht geliefert werden kann. Wird in einem Unternehmen über Geldbeträge verhandelt verfügen die Gruppen über virtuelle Konten. Diese Konten können ebenfalls genutzt werden, um die oben beschriebenen Ausleihfunktionen für die Gruppen interessanter zu machen. Für einen Mitarbeiter oder eine Maschine kann ein Geldbetrag geboten werden, ebenso für das übernehmen eines Arbeitsschrittes.
3.7 Zusammenfassung Der virtuelle Marktplatz bietet verschiedene Verhandlungsmöglichkeiten. Diese können, neben der individuellen dezentralen Feinplanung der einzelnen Gruppen, für eine unternehmensweite Optimierung der Produktion durch Austausch zwischen den Gruppen genutzt werden. Das Prinzip der virtuellen Verhandlungen stellt sicher, dass Auslastungen gleichmäßig über die Gruppen verteilt und Ressourcen möglichst gut genutzt werden. Ressourcen werden auf diese Weise nur dann abgegeben, wenn die Gruppe sie selber nicht braucht, oder es schafft ihre Planung entsprechend anzupassen. Die Flexibilität in der Fertigung kann durch den Marktplatz erhöht werden. Damit schafft sich das Unternehmen Freiräume, um angemessen auf änderungen der Lage in der Fertigung reagieren zu können. Prinzipiell ist davon auszugehen, dass die Nutzung des virtuellen Marktplatzes über finanzielle Anreize unterstützt werden muss [FMM08]. Es muss sowohl für die Gruppe als auch für den Einzelnen mit Vorteilen verbunden sein, einen Mitarbeiter in eine andere Gruppe auszuleihen oder zusätzliche Auftragsschritte zu übernehmen. über das im folgenden Kapitel kurz beschriebene Entgeltmodul können entsprechende Anreize geschaffen werden.
4 Methode zur Entgeltfindung in Gruppen mit wechselnder Zusammensetzung 4.1 Grundidee der Methode Ziel des Entgeltsystems ist die Ermittlung eines Bonus-Punktwertes für jeden Mitarbeiter. Der in diesem Konzept vorgesehene Punktwert kann sowohl zur Verteilung eines ERA basierten Leistungsbonus, zur Verteilung eines tarifunabhängigen
632
Thilo Münstermann et al.
unternehmensweiten Bonusbetrags oder zur Ermittlung flexibler, individueller Zusatzzahlungen verwendet werden. Die Entscheidung darüber findet also bewusst nicht innerhalb dieser Methode, sondern bleibt der Personalabteilung oder sonstigen Entscheidern überlassen. Die Methode basiert auf einer individuellen Entlohnung für einen Mitarbeiter, die er sich selbst erarbeiten kann, und einer Entlohnung für eine Gruppe. Dieser zweidimensionale Ansatz wird gewählt, um zum Einen der Einzelleistung eines jeden Mitarbeiters gerecht zu werden und zum Anderen den Leistungen innerhalb einer Gruppe Rechnung zu tragen [Boh91]. Die im Folgenden dargestellten Faktoren zur Entgeltberechnung stellen dabei eine Basismenge dar, aus der für jedes Unternehmen eine passende Auswahl getroffen werden kann. Abhängig von der Firmenpolitik und den strategischen Zielen kann so entschieden werden, was belohnt werden soll. Ist z. B. die Etablierung eines stärkeren Gruppengedankens notwendig, kann auf individuelle Leistungsentlohnung verzichtet werden. Gibt es hingegen eine sehr unterschiedliche Leistungsbereitschaft unter den Mitarbeitern, kann eine stärkere Verlagerung auf individuelle Faktoren den notwendigen Anreiz zur persönlichen Einsatzsteigerung liefern. Zur Berechnung vieler Faktoren ist die Betrachtung eines Zeitraums (t) notwendig, z. B. eines Monats oder Quartals.
4.2 Entlohnungsfaktoren Im AGA sind elf Bausteine für den individuellen Entlohnungsteil umgesetzt worden. Jeder Baustein kann als ein Faktor zur Berechnung des Gesamtentgelts gesehen werden. Zahlreiche weitere Faktoren als die hier vorgestellten sind denkbar. Es wird in individuelle- und gruppenspezifische- Faktoren unterscheiden. Zu den individuellen Faktoren gehören: Mitarbeiterqualifikation Eingesetzte Mitarbeiterqualifikation Häufigkeit der Springertätigkeit Dauer der Springertätigkeit und Geleistete Stunden außerhalb der normalen Arbeitszeit (Überstunden). Die gruppenspezifischen Entlohnungsfaktoren bilden: Mitarbeiterausleihstunden Mitarbeiterqualifikationsausleihe Maschinenausleihstunden Maschinenfunktionsausleihe Übernommene Auftragszeit und Pünktliche Auftragsfertigstellung. Die Bedeutung der Faktoren und ihre Berechnungsgrundlage auf Basis von Planungsdaten wird bei Henning et al [HFNM09], S. 35 f. detailliert beschrieben.
Dynamische Gruppenarbeit
633
Wichtig ist festzuhalten, dass eine gruppenübergreifende Abstimmung und Optimierung über einen virtuellen Marktplatz durch entsprechende (finanzielle) Anreize unterstützt werden muss. Ein entsprechendes Entgeltmodul muss stark parametrisierbar sein, um der Unternehmensführung zu ermöglichen, bestimmte Verhaltensweisen entsprechend den Unternehmenszielen und der –kultur zu fördern.
5 Zusammenfassung und Ausblick Das Konzept des Arbeitsgruppenassistenten bietet Methoden und Softwarefunktionen an, um dezentrale Feinplanungsprozesse mit gruppenübergreifender Zusammenarbeit und Optimierung zu unterstützen und diese mit Entgeltbonussystemen zu koppeln. Dabei wurden die einzelnen Möglichkeiten des AGA, zu denen Mitarbeiter anfragen, Maschinen anfragen, Arbeitsschritte anbieten und die Versteigerung zählen, sowie elf sinnvolle Faktoren der Entgeltberechnung beschrieben. Das Bonusentgeltsystem wurde dabei als umsetzungsnahste Variante beschrieben. über eine frei zugängliche Demonstratorsoftware wurde die Umsetzbarkeit des Konzeptes gezeigt. Diese ist als Open Source Software unter http://sourceforge.net/projects/aga/ kostenlos zum Download bereit gestellt. Weitere Möglichkeiten liegen zum einen in der Entwicklung automatischer Optimierungsverfahren, welche Potenziale der gruppenübergreifenden Zusammenarbeit erkennen und dabei die Planungsautonomie der Gruppen berücksichtigen. Henning et al. haben gezeigt, dass über eine Kombination aus dezentraler und zentraler Optimierung eine Reduktion der Komplexität von Produktionsplanungsoptimierung möglich ist [HFNM09], S. 122 ff.. Zum anderen muss das Konzept aufgegriffen und in bestehende ERP/PPS-Systeme integriert werden.
Literaturverzeichnis [AGA] [AHB03]
http://sourceforge.net/projects/aga., downloaded 23.09.2009. C. Antoni, K. Hofmann, and W. Bungard. Gruppen- und Teamarbeit: Neue Organisationsformen in Unternehmen. Ein Handbuch für das moderne Management. Springer Berlin / Heidelberg, 2nd edition, 2003. [AMF08] A. Askri, Thilo Münstermann, and Paul Flachskampf. Produktionsplanung durch dezentrale Feinplanungsschritte optimieren. phi - Produktionstechnik Hannover informiert, 11:10–11, 2008. [Ash56] William Ross Ashby. Introduction to Cybernetics. Chapman and Hall, London, 1956. [Boh91] A. Bohnhoff. Ein prospektiv bewertetes Identifizierungssystem für schnell bewegte Güter in kombinierten Verkehr. VDI Verlag, Düsseldorf, 1991. [FMM08] Paul Flachskampf, Thilo Münstermann, and Christiane Michulitz. Einführung eines Softwaretools zur Unterstützung der Gruppenarbeit. Teil 5. In Jörn-Axel Meyer, editor, Management-Kompetenz in kleinen und mittleren Unternehmen - Jahrbuch der KMU Forschung und -Praxis, pages 319–329. EUL Verlag, Lohmar - Köln, 2008. [FU03] Markus Fecht and Mark Unbehend. Gruppenarbeit in Produktionsbetrieben. Tectum Verlag, Marburg, 2003. [HFNM09] K. Henning, P. Flachskampf, R. Nickel, and T. Münstermann. Konzept für ein Informationssystem zur Unterstützung der Gruppenarbeit in mittelständischen Industriebetrieben der Metallverarbeitung. Books on Demand Verlag, Norderstedt, 2009.
634 [HG96] [LM05]
[Mal04]
[Sau00]
[Wes00]
[Wie05]
Thilo Münstermann et al. K. Heinz and R. Gosmann. Entgeltgestaltung bei Gruppenarbeit - Erkenntnisse und Erfahrungen aus einem Arbeitskreis. REFA-Nachrichten, 49(4):26–29, 1996. G. Lay and S. Maloca. Aufgabenintegration - Abkehr vom Taylorismus? Stand der Nutzung integrierter Modernisierungskonzepte zur Ausweitung des Tätigkeitsspektrums auf Werkerebene. PI-Mitteilungen, 36, 2005. Fraunhofer Institut für Systemtechnik und Innovationsforschung. Fredmund Malik. Systemisches Management, Evolution, Selbstorganisation: Grundprobleme, Funktionsmechanismen und Lösungsansätze für komplexe Systeme. Paul Haupt Verlag, Bern, Stuttgart, Wien, 4th edition, 2004. R. Sauerwein. Zur Diffusion von Gruppenarbeit im Maschinenbau - Gestaltung und Dynamik. In Ulrich Widmaier, editor, Der deutsche Maschinenbau in den neunziger Jahren: Kontinuität und Wandel einer Branche, pages 147–176. Campus Verlag, Frankfurt, 2000. A. Westerwick. Ein objekt- und nutzerorientierter Software-Entwicklungsprozeß am Beispiel eines arbeitsplatznahen Planungssystems in der Fertigung. VDI Verlag, Düsseldorf, 2000. Hans-Peter Wiendahl. Betriebsorganisation für Ingenieure. Carl Hanser Verlag, München, 2005.
Knowledge Base Concepts in the KEA System Combined with Social Networking Techniques Nicole Natho, Sabina Jeschke, Marc Wilke, Olivier Pfeiffer
Abstract We present a knowledge management system for mathematics suggesting a combination of an information retrieval system with social networking techniques to overcome information flood in mathematical and natural scientific texts, and problems of merging databases within the system to structure our data efficiently. With regard to the increasing demand of knowledge management systems in all fields, especially mathematical knowledge management systems remain a major challenge. Particularly systems that automatically extract information using natural language processing methods require a very different semantic analysis of texts than other field specific languages. In this regard, a numerous number of information fragments has to be extracted based on the specifics of the underlying text structures. New concepts are needed to control this information flood. One possible approach is the use of social network techniques. Keywords Knowledge based system · knowledge management · natural language processing · social networking
1 Introduction The mathematical and natural scientific knowledge contained in digital lectures and articles, used in numerous eLearning platforms like Moodle [moo] and in other learning relevant applications, or just in texts in the World Wide Web, is increasing the demand for sophisticated knowledge retrieval and knowledge management systems. These numerous digital natural language texts are easy to find with search engines such as Google, but they emerge in diverse complexity factors. On top of everything, it is not easy to identify which prerequisites are necessary – for experts and nonprofessionals alike - in order to read these texts efficiently.
N. Natho (B) MuLF, TU Berlin, Straße des 17. Juni 136, 10623 Berlin, Germany e-mail: [email protected]
To overcome the resulting information flood, we suggest using a knowledge management system. Such a system has three main purposes: creation, transfer, and execution of knowledge. The following question has to be answered: “What is knowledge in this context, and how does it differ from the concepts of information and raw data? Natural scientific language and especially the mathematical language are the formalization of thoughts embedded in logical conclusions. A simple analysis of raw material such as written text is not enough to develop a real comprehension of the content of these texts. Only reader her-/himself can induce comprehension by actively dealing with the context. Following Leidner’s idea [AL01], we characterize such knowledge as personalized knowledge or personalized information respectively, which is related to facts, concepts, ideas, etc. of a single person. For successful knowledge management, distinguishing between the different kinds of personified knowledge is important; cognitive psychology (cf. Anderson [And01]) classifies knowledge in declarative and procedural while philosophy (cf. Polanyi [Pol74]) groups explicit and implicit knowledge. In written texts or spoken language we find explicit (declarative) knowledge whereas implicit (procedural) knowledge is the ability to use verbalized knowledge to logically conclude and reason, e.g. in mathematical proofs. In [Non94, NT95, NK98] Nonaka et al. use these two kinds of knowledge to create a theoretical foundations of knowledge management systems. Four distinct processes are the base of this approach: socialization, combination, internalization and externalization. The KEA-System [PRR06] is based on this approach. It supports the idea of dynamic developing the user’s knowledge through these four basic processes that are of importance for the comprehension of mathematical texts. Besides the theoretical foundation, the system is build up on the operative, strategic and normative building blocks of Probst at al. [PRR06, PR97]: knowledge identification, knowledge acquisition and knowledge development, knowledge distribution and usage, and knowledge perpetuation. The KEA system is designed to process data from mathematical and natural scientific texts to construct a system for users who have to deal with personalized information in mathematics and natural sciences such as students, teaching- or academic staff, engineers, etc. via an internet platform (knowledge identification). We integrate explicit knowledge in the KEA system in terms of knowledge acquisition. The system tries to formalize explicit knowledge in an automatic way so that no human user is required to collect the data for integrating them into knowledge bases. Different approaches to this challenge exist; one basic approach, which we follow, is the automatic generation of semantic annotations, based on ontologies describing the fields and structures of natural language texts themselves. The KEA system is based on natural language processing techniques, taking advantage of characteristic linguistic structures defined by the language use in natural scientific and mathematical texts (see Figure 1 for an overview of the system). It creates knowledge bases based on these special language constructs, extracted from text fragments in textbooks or webpages. The system organizes and manages numerous single knowledge bases of different authors, texts and fields, and therefore, it
Knowledge Base Concepts in the KEA System
637
Fig. 1 Overview of the KEA system
develops new knowledge based on automatic extraction. In addition, there is little influence which knowledge has to be integrated. The analysis of relevant knowledge content is relocated to the information retrieval of the KEA system. At the backend, KEA has a web-based information retrieval system for different applications areas such as encyclopedias, library systems, and e-learning systems. In this part of the system, system operators may determine knowledge usage and distribution. The system is designed such that it facilitates serving different kinds of applications for different kinds of users. All usable applications have in common that extracted information (explicit knowledge) is adapted to the needs of different backend users. The critical point of such a system is the sheer amount of data and merging processes of different knowledge bases. In recent years, a new interesting field gained in importance in conjunction with the Web 2.0: social bookmarking. By using social bookmarking methods, users
638
Nicole Natho et al.
can organize and manage tags in webpages in a very easy way. These tags can be shared with other users. The advantage of this concept is obvious: human readers who know content and relevance of internet resources classify and rank them in such an efficient way that no current automated system could do. By using social bookmarking techniques, the KEA system integrates the idea of Nonaka et al. [Non94, NT95, NK98]. In this stage, implicit knowledge is rudimentarily integrated into the systems knowledge bases in the sense of the four basic processes: externalization, internalization, socialization, and combination. The users’ interaction with the system induces the generation of personalized knowledge bases representing users’ standard of knowledge, and the improvement of the existing automated generated knowledge bases. However, the disadvantages are lacking standardization and the unstructuredness of tags apart from manipulative ranking methods. However, social bookmarking is widely used and has proven to be a very successful concept, offering efficient ways of bypassing the problem of manual metadata input for search engines by the authors. We present a solution using social networking techniques with the KEA system to overcome the information flood and the problem of merging databases to structure our data efficiently. One approach uses the contained axiomatic structure of mathematical texts to establish different levels in knowledge bases for arranging text fragments from a coarse up to fine-grained structure level. For example, the designator of a mathematical definition lives in a higher structure level in a database than the sentences defining this definition. Another approach is the appropriation of the user data to track the navigational structure and use of tags. Based on user data user profiles are constructed improving the knowledge base structure. In order to adapt to the user standard of knowledge and her/his navigational preferences the knowledge base is virtualized. In the context of social bookmarking, users aid to integrate a missing ranking mechanism into the KEA system, which it cannot automatically extract of text fragments. This tagging technique not only deals with text fragments; beyond that equation fragments can also be tagged. By social networking techniques, we integrated human behavior and knowledge into the system to provide concepts for structuring our knowledge bases and merging mechanisms in an efficient way. After the literature review in the following section we will briefly outline of the complexity of KEA system in section 3. We explain the structure of the knowledge bases and the possible merging concepts in section 4. To overcome the challenges of structuring data within the knowledge bases we represent possible solutions in section 5.
2 Literature Review To date we have not learned of a knowledge management system similar to KEA regarded as a turnkey solution. This fact is probably induced because the KEA system is not a classical knowledge management system but rather a system for processing
Knowledge Base Concepts in the KEA System
639
natural scientific and mathematical natural language information. Primarily it serves rather for processing of data (explicit knowledge in texts) as for the spreading of knowledge in the sense of implicit knowledge (drawing conclusions in mathematics and natural sciences). Implicit knowledge in mathematics and natural sciences can be only imparted through intensive studies of the subject by the user her/himself [Nat05, And01]. Therefore, the KEA System can only guide users to using the appropriate explicit knowledge. Therefore, we use social network techniques to overcome these challenges. However, from the theoretical standpoint of view, the KEA system is based on the fundamental ideas of Nonaka [Non94, NT95, NK98] and Probst [PRR06, PR97]. Furthermore, KEA consists of different components, which can be compared to other projects: basic results on mathematical ontologies were accomplished by Gruber and Olsen [GO94]. MBASE [KF01] is an example of a manually written mathematical ontology. Investigation on English mathematical texts was accomplished by Baur [Bau99]. An example of a manually assembled mathematical encyclopedia is MathWord [wol]. GermaNet [ger] and WordNet [Fel98, Fel] are intelligent thesauri providing a number of suitable synonyms and definitions for given queried expressions similar to the semantic analysis of mArachna core project. Helbig [Hel05] investigated concepts for automated semantic analysis of the German language. Mizar [Urb06, Urb05] is used for describing mathematical proofs through a formal human and machine-readable language as input for an automated reasoning system. DIALOG [PSBK] processes natural language for mathematical validation in an automated reasoning system. HELM [APSS01] uses metadata provided by authors for semantic annotation of mathematical texts used in digital libraries. MoWGLi [AZ04] in combination with HELM is an information retrieval interface based on pattern matching within mathematical equations and logical expressions.
3 The KEA – System The KEA system is a semi-automated knowledge management system based on natural language processing techniques taking advantage of characteristic linguistic structures defined by the used languages in natural sciences and mathematics. One of its core components is the mArachna system, the NLP subsystem, consisting of several separate components: a pre-processed analysis for different text formats, the syntactical and semantic analysis. The mArachna system implemented the TRALE system [Mül, Mül05] based on head-driven phrase structure grammars. Symbols and mathematical equations have to be processed separately in the mArachna System [JNPW07, JNPW08] because of its special structures. The whole analysis based on a linguistic approach [Nat05] taking the special structure of mathematical languages into account to describe content in an axiomatic way. Thus, entities are principal carries of information in such
640
Nicole Natho et al.
texts. New syntactic and semantic rules are integrated through XML documents implemented by a Java Script Interpreter [Fou]. This guarantees an easy access for all users to write new rules for the mArachna system, and improve our system by external users. As a result of the semantic analysis the system creates complex triples structures (information fragments) annotated with addition information about the text corpus. Another core component is the knowledge base system. All information fragments are integrated into the knowledge bases integrated by dint of the Java-based Jena Semantic Web Framework [Jen]. As a result the knowledge bases are easy manageable OWL documents [cWa, cWc, rdf]. Jena provides retrieval mechanisms such as SPARQL [cWb] and an interference engine to extract information from such OWL documents. The third component is the information retrieval provided through a web interface. We have two interface types: administrator interface and user interface. The administrator interface is a fixed construction to provide direct access to all relevant internal structures to control the whole system such as graphical and textual in and outputs of different stages of the analysis, graphical representations of parts of the knowledge bases, etc. Moreover, we try to implement the administrator interface into Eclipse [ecl] as an extension. The user interface follows the ideas of the desktop in a browser corresponding with the concept of Web 2.0 applications. So, the user has the possibility to display only preferred working tools (desktop agents) such as special searching and graphical agents. In addition, for each different user application (encyclopedia, libraries, e-learning tools, etc.) new and reusable desktop agents (different searching agents, graphical representations, history log files, tagging tools, and notepads) can be easily provided. The graphical realization based on the graphic tool package Graphviz [Res]. In this way Web 2.0 and semantic web technologies [Doo06, sem] are combined.
4 KEA’s Knowledge Bases System Each triple (cf. sec. 2) consists of nodes representing an expression, phrase or term, and the relation between nodes describes their associations. Each relation exemplifies different types of linguistic phrases or key words. To support complex dependencies triples themselves are often used as nodes of other triples (clustering). The resulting structures are very complex: for each element, it has to be decided whether it represents OWL classes or individuals of classes. Above all, we found ways of using OWL DL instead of OWL FULL which was a crucial factor in the beginning. This process creates elaborated and intricate knowledge bases with convoluted and very fine-grained structures. To guarantee consistency of the knowledge bases we pursue two strategies: At first, each text generates a single independent knowledge base. Second, we use a semi-automated approach for new information fragments to encounter contradictions during the integration process.
Knowledge Base Concepts in the KEA System
641
4.1 Knowledge Base Concept Every single knowledge base is a complex and huge semantic network. Due to the fine-grained structure of texts in mathematics filler words are used as seldom as redundant information fragments. The language is precise. Each single sentence supports important information about the content. Thus an information flood is accumulated which has to be managed. A knowledge base is a multidimensional object consisting of several levels based on linguistic structures where each level forms an ontology of the text. At the lowest level we find single words and symbols used within a text in a specific way. In the next level there is a collection of collocations or information fragments within the sentences which are used correlative. A single word or phrase is connected specifically with other phrases or words. In the next level we build up entities. Entities are objects such as definitions, theorems or prepositions. Each of these entities has a specific internal structure which is stored in this level. At the bottom most level we map the axiomatic structure of the text itself. The technical implementation is based on the RDF-based Web Ontology Language (OWL) [cWa, cWc, rdf] backed by the JENA Framework [Jen]. Thus, we integrate semantic web standards, and build up semantic Metadata automatically. Additionally mind maps of text ontologies could be provided using common software tools, e.g. Protégé [Fel98]. The fragmentation into linguistic structure levels is the key component to support the merging processes, to include social network techniques, and to implement the information retrieval mechanisms.
4.2 Merging Concepts Due to the strategy that each text generates a new knowledge base, the system has to manage a lot of these databases. For retrieval mechanisms it is complicated to get access to the information, especially if same information fragments are spread above different knowledge bases. The solution to this problem is unifying knowledge bases of identical content. For example: knowledge bases of the same mathematical field from different authors have to be unified. In addition, tools and strategies for portioning unified knowledge bases into smaller ones have to be developed. Such smaller knowledge bases can be adapted to the preferences and requirements of specialized target groups (e.g. students of engineering, physics or mathematics) to harmonize teaching material within courses. Trying to do this we encounter some problems: even in mathematics and natural sciences, different authors use different terms and axiomatic structures to describe similar content. The differences are so subtle that even a human reader needs profound knowledge to be able to manage them. Therefore, another strategy is to use a semi-automated approach in the KEA –system but this has not proved satisfactory.
642
Nicole Natho et al.
Intervention by a human administrator through by means of an administrative webbased interface is not our primary intention. However, first tests show that this intervention is not the rule.
5 Social networking in KEA Social networking is an outstanding method for integrating human knowledge into knowledge bases and the usage of tagging mechanisms is already widely accepted amongst users. However, the disadvantages are lacking standardization and unstructuredness of tags apart from manipulative ranking methods. Nevertheless, our database structures can benefit of the support of social networking.
5.1 Database Concept Generally, there are two types of databases. First, there exits knowledge bases which are containing semantic information fragments (the semantic network) of the analyzed texts. In this regard, knowledge bases are split into two components: One component (textual component) manages pure text portions. The other component manages equations and symbols only. Although from the outside it seems that there is a single knowledge base for all text-relevant issues, on the inside the system has to treat these two components as separated parts because of their different syntactical structures (cf. Figures 2 and 3).
Fig. 2 Structure of a knowledge base
Knowledge Base Concepts in the KEA System
643
Fig. 3 Internal structures of the knowledge base
The second database manages user profiles containing collected data about the users’ interaction with the system and information about the user itself. KEA tracks user navigation through the system, and builds up user profiles using volunteer questionnaires or self-assessment tests to estimate standard of knowledge of a single user. In this database, we also include social networking components to store all bookmarks (tags) of the users. As a result, we get information fragments about the behavior of the user within our system. Figure 4 illustrates the database arrangement.
5.2 Technical Concept Standardized methods and techniques based upon open source projects are used to realize this project. First, RDF1 -based languages to support semantic web technologies and store metadata for further processing is used. Furthermore, the knowledge bases and the some parts of the user profiles databases are based on the Web Ontology Language (OWL), an RDF-based language storing relationships between information fragments. Additionally, FOAF (Friend of a Friend) [fri] is used to store information about the user and the tags adapted with slight modifications for our purposes, not restricted to personal data, but also containing everything of particular importance in the learning process like standard of knowledge, attended lectures
1
resource description framework
644
Nicole Natho et al.
Fig. 4 KEA database system
and exercise courses. This standard is widely used in different kinds of social networks tools [peo, irc] to easily exchange, merge and reuse data describing people and relations between them, and for the process of creating and sharing tags for different formats (pictures, weblogs, etc). FOAF is also an RDF-based language. Consequently, we have the same basis of languages in such a way, that we are able to process all components together, or to query them with one automated search engine, which is able to analyze semantic web resource descriptions. Hence, the databases can be easily managed based on RDF. Input and output of the databases are controlled using the Java-based Framework JENA. JENA has several advantages: it supports OWL documents, comes with a query language SPARQL and an interference engine to get additional information from our databases. The knowledge bases can be analyzed graphically by the graphical tool kit Graphviz [Res]. The graphical representation of the FOAF parts of the user profiles database are provided by the java-based tool TouchGraph [tou]. For an illustration of the technical concept, see Figure 5.
5.3 Merging Concept with Social Networking Merging databases appears to be the adequate means of overcoming the challenges information flood and semi-automated approach. In the following, the algorithm of merging those databases is described. First virtual, i.e. databases not containing content but rather links to the content, databases belonging to each specific user are created (cf. Figure 7 for an illustration). Due to the high amount of users, many of
Knowledge Base Concepts in the KEA System
645
Fig. 5 Technical concept of KEA
these virtual databases have to be created. Of course, similar virtual databases can be merged by checking out similarities. In the knowledge bases, we will find information fragments. The user profile database has several layers: 1. Tracking Layer: tracking ways through the knowledge bases by the users him-/ herself. a. Used information fragments in the knowledge base b. Stop time of the a single information fragments c. Number of visits on a single information fragment 2. Social Network Layer: Bookmarking a. Tags of information fragments b. Time of tagging (history) c. Exchange information to different user groups 3. Questionnaire Layer a. b. c. d.
Personal data Knowledge standard Fields of interest Self assessment
The tracking layer is build up with OWL whereas the social network and questionnaire layer is build up in FOAF.
646
Nicole Natho et al.
Analysis of the databases is accomplished in a different way and analyzing knowledge bases is more differentiated than for user profile databases. As each information fragment of a text is important, every information unit is incorporated in the knowledge base. In a completely automated procedure, the distinction between necessary, less necessary, or even wrong entries is very extremely time-consuming or even impossible. In the semiautomatic approach, a human expert can make the appropriate error corrections, for which the system inquires as soon as an inconsistency is noticed. First observations show however that there are only very few ambiguities. In order to automate this semi-automated procedure, the entries must be weighted and/or evaluated. The same problem arises during the merging process: if two homogenous units of information, however with different peculiarities, are to be brought together with one another, it makes sense to use both respective entries, even though with different weights, corresponding to the page ranking methods: the more often an information fragment is used the higher its ranking gets. When the knowledge bases are united, the linguistic layers are dissolved. At first, the axiomatic structure is merged, followed by the internal structure up to the word level. In this process, the collocation-layer plays a decisive role as the concept formation taking place in this layer can be mapped the most simply. The same process is carried out with formulae and the symbol knowledge base (cf. Figure 6). If ambiguities or inconsistencies arise in the individual levels, both variants are integrated and ranked. The more texts respectively knowledge bases are merged the better the obtained rankings are. During the linguistic analysis information supplementary to the actual semantic net in the knowledge bases, for example target group (mathematicians, chemists, etc.) is stored, that can now be applied. On the other side, we have the user-profile databases (UPDB), which have been produced by the human users and their navigation through the knowledge base.
Fig. 6 Merging process for text knowledge bases
Knowledge Base Concepts in the KEA System
647
Fig. 7 Virtual databases
Therefore, these structures can be filed to the UPDB, based on the user’s standard of knowledge. Comparison with the existing knowledge bases then generates the virtual database. These virtual databases reflecting the human interaction with the system help structuring the databases and only information relevant for the specific user is selected for his/her navigation through the database. By examining the length of stay and frequency of the visits a decision about the importance of an information fragment can be concluded, which determines the ranking of these fragments. Based on this ranking the fine-structured portion of the database is dissolved and only the most significant components are mapped to the virtual database, yielding a customtailored database for each user. For a general presentation of information social bookmarking, neglecting the individual’s previous knowledge, by many persons is appropriate.
6 Conclusion and Future Work KEA is a semi-automatic web-based knowledge management system constructing and organizing knowledge bases for mathematical language to store, maintain, and retrieve information combining different technologies and shall be extended to manage natural scientific texts also. Metadata is generated in form of RDF/OWL
648
Nicole Natho et al.
triples and FOAF documents, which can be used in Semantic Web environments. Social tagging technologies improve our knowledge results based on tracking methods of users’ knowledge, building tags, and questionnaires. Desktop agents are provided to maintain and support the individual learning process, motivation and the interest of user. Using such an environment, the users’ and administrators’ individual preferences are preserved. During further development, a collaborative environment shall be evaluated to augment the user-friendliness of the system, to improve, and to test the system itself. Caused by the way how we learn to understand mathematics and natural sciences the KEA system is not a classical knowledge management system. In the first stage, it is primarily a system to process data in form of explicit knowledge (natural language texts). Generally, the system itself cannot impart implicit knowledge because the computer cannot draw conclusions. Thus, this task is left to the user, and can only be achieved by intensive studies of the subject2 . Therefore, the KEA system can only assist users to use the adequate explicit knowledge. For this purpose, we use social network. From the theoretical point of view, the KEA system is based on the fundamental ideas of Nonaka [Non94, NT95, NK98] and Probst [PRR06, PR97]. Currently, KEA is a testbed analyzing mathematical texts written in German and partly in English, processing selected entities and texts blocks and successfully integrating the extracted information into knowledge bases. We provide parts of the user interfaces for administrators and the user interface for a mathematical encyclopedia. We try to analyze complete textbooks for different fields of mathematics, starting with geometry because of the strong axiomatic structure of this particular field. For evaluation purposes, KEA will be integrated in a cooperative knowledge space system used for mathematical learning environments. Such systems contain numerous entries of mathematical and natural sciences language constructs. This approach confronts KEA with a wide range of examples of correct and incorrect usage of mathematical language constructs. In addition, our workgroup can create representative bookmarks for the system and further examine users’ knowledge skills through tracking methods and questionnaires. This is only the first step in documenting the users’ skills and allows for guiding the user to improve or complete his/her knowledge, e.g. by taking additional courses, hints for additional literature and check for correlations between the usage of language and actual skills. Finally, we are very interested to see which types of desktop agents will be developed by the users within the framework of KEA, and how social networks techniques improve the system.
2 Automatic reasoning systems are another active field of research, yet far from being applicable to our purposes.
M. Alavi and D. Leidner. Knowledge Management and Knowledge Management Systems: Conceptual Foundation and an Agenda for Research. MIS Quarterly, pages 107 – 136, March 2001. John R. Anderson. Kognitive Psychologie. Spektrum Akademischer Verlag, 3rd edition, March 2001. A. Asperti, I. Padovani, C. Sacerdoti, and I. Schena. HELM and the Semantic Web. In R.J. Boulton and P.B. Jackson, editors, Proceedings of the Theorem Proving in Higher Order Logics, TPHOLs 2001, volume 2152, Edinburgh, Scotland, UK, September 2001. Springer. Lect. Notes in Computer Science. A. Asperti and S. Zacchiroli. Searching Mathematics on the Web: State of the Art and Future Developments. In Proceedings of the Joint Proceedings of the ECM4 Satellite Conference on Electronic Publishing at KTH Stockholm, AMS - SM M Special Session, Houston, 2004. J. Baur. Syntax und Semantik mathematischer Texte. PhD thesis, Universität des Saarlandes, FB Computerlinguistik., November 1999. Master’s thesis. World Wide Web consortium W3C. Resource Description Framework (RDF). http:// www.w3.org/RDF/. World Wide Web consortium W3C. SPARQL. http://www.w3.org/TR/rdf-sparqlquery/. World Wide Web consortium W3C. Web Ontology Language OWL. http://www.w3. org/2004/OWL/. L. Doods. Introduction SPARQL: Querying the SemanticWeb. http://www.xml.com/ lpt/a/2005/11/16/introducing-sparql-querying-semantic-web-tutorial.html, 2006. Eclipse.org home. http://www.eclipse.org/. Stanford Center for Biomedical Informatics Research. The Protégé Ontology Editor and Knowledge Acquisition System. http://protege.stanford.edu/. C. Fellbaum. WordNet An Electronic Lexical Database. http://wordnet.princeton. edu/. C. Fellbaum. WordNet: An Electronic Lexical Database. MIT Press, Cambridge, London, 1998. Mozilla Foundation. Rhino – JavaScript for Java. http://www.mozilla.org/rhino/. The Friend of a Friend (FOAF) project | FOAF project. http://www.foaf-project.org/. GermaNet Team, GermaNet. http://www.sfs.uni-tuebingen.de/lsd/english.html. T. Gruber and G. Olse. An Ontology for Engineering Mathematics, Technical Report KSL- ,. pages 94–18, 1994. Hermann Helbig. Knowledge Representation and the Semantics of Natural Language (Cognitive Technologies). Springer, Berlin, November 2005. IRC Community Support Agent (Foafbot),. http://usefulinc.com/foaf/foafbot/. Jena. A Semantic Web Framework for Java. http://jena.sourceforge.net/. S. Jeschke, N. Natho, O. Pfeiffer, and M. Wilke. KEA - a Mathematical Knowledge Management System combining Web 2.0 with Semantic Web Technologies. In Proceedings of the 4th IEEE 4th International Conference on Innovations in Information Technology, pages 138–142, Los Alamitos, CA, 2007. IEEE Computer Society. S. Jeschke, N. Natho, O. Pfeiffer, and M. Wilke. Managing Mathematical Texts with OWL and Their Graphical Representation. In Proceedings of the HICSS 41, Waikoloa, HI, 2008. IEEE Computer Society Conference Publishing Services. M. Kohlhase and A. Franke. MBase: Representing Knowledge and Context for the Intergration of Mathematical Software Systems,. Journal of Symbolic Computation, 23:4:365 – 402, 2001. Moodle (Modular Object-Oriented Dynamic Learning Environment). http://moodle. org/.
650 [Mül] [Mül05] [Nat05]
[NK98] [Non94] [NT95]
[peo] [Pol74] [PR97]
[PRR06]
[PSBK] [rdf] [Res] [sem] [tou] [Urb05] [Urb06]
[wol]
Nicole Natho et al. S. Müller. TRALE. www.cl.uni-bremen.de/Software/Trale. S. Müller. Deutsche Syntax deklarativ: Head-Driven Phrase Structure Grammar für das Deutsche. Linguistische Arbeiten, (394), 2005. Max Niemeyer Verlag, Tübingen. N. Natho. MARACHNA: Eine semantische Analyse der mathematischen Sprache für ein computergestütztes Information Retrieval. PhD thesis, Technische Universität Berlin, 2005. I. Nonaka and N. Konna. The Concept of ’Ba’ Building a Foundation for Knowledge Creation. California Management Review, 40:3:40–54, 1998. I. Nonaka. A Dynamic Theory of Organizational Knowledge Creation. Organized Science, 5(1):14–37, 1994. Ikujiro Nonaka and Hirotaka Takeuchi. The Knowledge-Creating Company: How Japanese Companies Create the Dynamics of Innovation. Oxford University Press, USA, May 1995. The People’s Search Engine (plink). http://www.uoguelph.ca/~bcoe/. Michael Polanyi. The Tacit Dimension. University Of Chicago Press, Corr. Ed. edition, 1974. G.J.B. Probst and K. Romhardt. Building Blocks of Knowledge Management – A Practical Approach. 102, 1997. Ecole des Hautes Etudes Commericales, Universite de Geneve. G.J.B. Probst, S. Raub, and K. Romhardt. Wissen managen: Wie Unternehmen ihre wertvollsten Ressource optimal nutzen Betriebswirtschaftlicherverlag. Betriebswirtschaftlicherverlag Dr. Th. Gabler, Wiesbaden, 5. ed edition, 2006. M. Pinkall, J. Siekmann, C. Benzmüller, and I. Kruijff-Korbayova. DIALOG. http://www.ags.uni-sb.de/~dialog/. RDF Vocabulary Description Language 1.0: RDF Schema. http://www.w3.org/TR/ rdf-schema/. AT&T Research. Graphviz – Graph Visulaization Software. http://www.graphviz. org/. Semantiv Web Activity. www.w3.org/2001/sw. TouchGraph Google Browser. http://www.touchgraph.com/TGGoogleBrowser.html. J. Urban. MizarMode – An Integrated Proof Assistance Tool for the Mizar Way of Formalizing Mathematics. J. of Applied Logic, 2005. J. Urban. MoMM – Fast Interreduction and Retrieval in Large Libraries of Formalized Mathematics. Internation Journal on Artificial Intelligence Tools, 15(1):109–130, 2006. Wolfram MathWorld: The Web’s Most Extensive Mathematics Resource. http:// mathworld.wolfram.com/.
Enterprise Application Integration für die virtuelle Produktion Daniel Schilberg, Tobias Meisen, Philippe Cerfontaine, Sabina Jeschke
Zusammenfassung Möglichst präzise und detailliert virtuell abzubilden, wie aus unterschiedlichsten Rohstoffen ein fertiges Produkt hergestellt wird, ist das Ziel des Produktplanungsprozesses. Um ihn ganzheitlich anzulegen, müssen viele verschiedene Faktoren beachtet werden. So muss das Wissen von Experten unterschiedlicher Disziplinen zusammengebracht werden und der Prozess selbst muss dynamisch und flexibel angelegt sein, um an wechselnde Anforderungen adaptiert werden zu können. Da im Produktplanungsprozess viele heterogene und als Insellösungen entwickelte Softwarewerkzeuge eingesetzt werden, die jedoch lediglich Teile der Produktion vorab virtualisieren, ist es für eine durchgängige Darstellung des Prozesses notwendig, diese zu verbinden, um langfristig wettbewerbsfähige Produkte zu planen und herzustellen. Am Beispiel des Herstellungsprozesses eines Stahlrohrs für eine Pipeline wird deutlich, wo die Probleme bei einer durchgängigen Abbildung liegen. So müssen z. B. Veränderungen der Form eines Bauteils mit Veränderungen in der Werkstoffstruktur in Verbindung gebracht werden. Dies kann mit Hilfe der Enterprise Application Integration (EAI) erfolgen. In diesem Beitrag wird ein auf EAI basierendes Datenintegrationswerkzeug vorgestellt, das die Verknüpfung heterogener Simulationen ermöglicht. Schlüsselwörter Enterprise Application Integration · Data Integration · Simulation · Semantic Interconnection · Production
zur Planung und zur Prozessbeschreibung prozessorientiert, ohne den Quellcode der eingesetzten Softwaresysteme zu verändern. Alle benötigten Schnittstellen werden durch Adapter des Frameworks realisiert. Dies ermöglicht es, die einzelnen Softwaresysteme als Kette in nahezu beliebiger Reihenfolge und über eine einzige Mensch-Maschine-Schnittstelle zu verwenden und entsprechende Geschäftsprozesse abzubilden. EAI wird zurzeit in produzierenden Unternehmen primär für die Abbildung häufig wiederkehrender teil- oder vollautomatisierter Geschäftsprozesse eingesetzt. Die Softwaresysteme, die in den Planungsbereichen Material und Verarbeitungsprozesse Anwendung finden, werden bei den gängigen Frameworks nicht berücksichtigt. So wird bei der Abbildung eines Geschäftsprozesses die eigentliche Herstellung eines Produktes nicht dargestellt, was zur Folge hat, dass Implikationen vom Herstellungsprozess auf den Geschäftsprozess nicht betrachtet werden und dadurch wichtige Details verloren gehen. Wenn jedoch das Ziel verfolgt wird, die Produktion vom Rohstoff bis zum fertigen Produkt virtuell abzubilden, um eine große Anzahl an Implikationen zu identifizieren, ist eine ausschließliche Betrachtung des Geschäftsprozesses nicht ausreichend. Vielmehr ist eine möglichst durchgängige Modellierung des Prozesses vom Rohstoff bis zum fertigen Produkt durch Simulationen erforderlich.
2 Problemstellung Der Großteil der Herstellungsschritte bei der Produktion von Gütern ist durch einzelne Simulationen beschreibbar. Die Problematik dieser Verfahrensweise liegt darin, dass jeder dieser Schritte singulär betrachtet und nicht unmittelbar in den Kontext der gesamten Herstellungsprozesskette gesetzt wird, wie in Abbildung 1 am Herstellungsprozess einer Line-Pipe abgebildet. Dieser besteht aus einer Reihe von Wärmebehandlungen, Umformprozessen, spanenden und fügenden Verfahren.
Abb. 1 Herstellungsprozess einer Line-Pipe
Enterprise Application Integration für die virtuelle Produktion
653
Jeder dieser Produktionsschritte ist durch eine Simulation abbildbar. Die Modelle, die den Simulationen zugrunde liegen, basieren auf dem Expertenwissen einer Domäne wie beispielsweise der Wissensdomäne des Fügens, und stellen i. d. R. Insellösungen dar, die keine gemeinsamen Schnittstellen oder Datenformate aufweisen. Darüber hinaus arbeiten die Simulationen auf unterschiedlichen Skalen. So betrachten die Umform-, Spanenden- und Fügesimulationen Veränderungen der Makrostruktur, während sich die Wärmebehandlungsprozesse auf die Mikrostruktur und damit auf die Materialeigenschaften auswirken. In bestehenden Systemen ist eine Verknüpfung von Simulationen zu Simulationsketten für die Abbildung von Herstellungsprozessen, die derart verschiedene Skalen und Wissensdomänen umfasst, nicht ausreichend berücksichtigt. Um die Simulationen in Simulationsketten verwenden zu können, wird ein Integrationswerkzeug benötigt, das es ermöglicht, die einzelnen Simulationen miteinander zu verknüpfen [Sch10]. Im Folgenden wird der aktuelle Stand der Technik in Bezug auf EAI Frameworks vorgestellt. Im Anschluss daran wird ein Datenintegrationswerkzeug vorgestellt, das für die Kopplung heterogener numerischer Simulationen entwickelt wurde. Anhand des Herstellungsprozesses einer Line-Pipe wird das Werkzeug validiert. Der Beitrag schließt mit einer Zusammenfassung und Ausblick auf weitere Entwicklungsmöglichkeiten.
3 Stand der Technik EAI kann als umfassendes Konzept verstanden werden, das ein vollständiges System der Geschäftsprozesse, Management Funktionen und organisatorischen Interaktionen beinhaltet. Als ganzheitlicher Prozess gestaltet es übergangslose hoch-agile Prozesse und organisatorische Strukturen, die der strategischen und wirtschaftlichen Ausrichtung eines Unternehmens entsprechen. Diese Systeme zeichnen sich durch integrative, progressive und iterative Zyklen beim Einsatz von Technologien, Arbeitskräften, Wissen und operativen Prozessen aus [Mye02]. Bei Betrachtung der eingesetzten Informationssysteme zur Umsetzung von EAI wird der Begriff verständlicher und konkreter: EAI wird als Softwarewerkzeug zur Integration heterogener Systeme verstanden [gul10]. EAI Frameworks umfassen zur Bereitstellung der Funktionalität meist die Komponenten Metadatenbank, Middleware, Adapter, Nachrichtenmanagement und Prozessmanagement (vgl. Abbildung 2) [Hoh02]. Weitere wichtige Arbeiten bezüglich der EAI Modellierung sind den Werken von Linthicum, Ruh, Cummins, O’Rourke und Whitten zu entnehmen ([Lin00][RMea01][Cum02][OFS03][WBD04]). Die Middleware dient zur Verwaltung von Adaptern und Ressourcen sowie zur Bereitstellung von Zusatzdiensten. Die Adapter realisieren die Verbindung zwischen den zu integrierenden Applikationen und der Middleware. Es werden statische (fest implementiert) und dynamische (konfigurierbar) Adapter unterschieden, die die Inkompatibilitäten, die sowohl aus der syntaktischen als auch aus der strukturellen Heterogenität der Applikationen resultieren können, überwinden. Die Metadatenbank dient der zentralen Speicherung der Verteilung von Komponenten - Informationen bezüglich der Sicherheitsparameter, Verantwortlichkeiten, technologischen Infrastruktur, Nachrichtenschemata, Transformationen, Regeln und Logiken für die Verarbeitung von Nachrichten,
654
Daniel Schilberg et al.
Abb. 2 EAI-Komponenten
Architektur und des Designs werden hier abgelegt und verwaltet. Das Nachrichtenmanagement schließlich stellt Transformations- und Synchronisationsdienste bereit und gewährleistet die Transaktionalität von Transformationsoperationen. Neben der syntaktischen und strukturellen Heterogenität ermöglicht es die Überwindung der semantischen Heterogenität. Das Prozessmanagement steuert und administriert die Prozessmodellierung, die Prozesssteuerung sowie die Prozesskontrolle. Laufende und unterbrechbare Geschäftsprozesse mit manuellen Anteilen werden vom Prozessmanagement unterstützt. Des Weiteren übernimmt das Prozessmanagement die Aufgabe des Monitoring und Reporting [Hoh02]. Im Folgenden werden aktuelle EAI Frameworklösungen vorgestellt, zwei proprietäre von Oracle und IBM und zwei Open-Source Frameworks, und anschließend verglichen. Die Frameworks wurden aufgrund der weiten Verbreitung ausgewählt, weshalb die Auswahl keine inhaltliche Bewertung darstellt.
Oracle Application Server Interconnect Das von Oracle entwickelte EAI Framework Application Server Interconnect verwendet zur Applikationsintegration ein auf XML basierendes gemeinsames Datenschema und -modell. Die zu integrierenden Applikationen werden unter Verwendung einer Integrationslogik erstellt und über Integrationspunkte, die in einer Metadatenbank abgelegt sind, mit dem gemeinsamen Datenmodell verknüpft. Die Integration der jeweiligen Applikation geschieht ereignisgesteuert. Hierbei werden die Applikationen zu einem definierten Ereignis einem Integrationspunkt zugewiesen, wodurch die Abbildung eines Gesamtprozesses durch die über das EAI Framework verknüpften Anwendungen realisiert wird [20110a].
IBM WepSphere Product Suite Zur Verknüpfung von Anwendungen stellt die WebSphere Product Suite von IBM eine Sammlung von Werkzeugen wie beispielsweise Message Broker, Event Broker, Message Queue und Process Server bereit. Auch diese nutzt ein gemeinsames
Enterprise Application Integration für die virtuelle Produktion
655
Datenschema und –modell, das auf XML basiert. Für Protokolle wie FTP, HTTP usw. werden seitens der Suite Adapter zur Anbindung auf Datenebene angeboten. Die Benutzeroberfläche stellt eine Form der graphischen Programmierung dar, bei der konfigurierbare Blöcke untereinander verknüpft werden können, um eine Applikationsintegration zu programmieren [20110b].
JBoss Auch das in Java entwickelte Open-Source EAI Framework arbeitet mit XML Nachrichten, die für die Kommunikation innerhalb des Frameworks verwendet werden. Anwendungen werden über Adapter in das Framework integriert. Die jeweiligen Adapter erzeugen JBoss konforme XML Nachrichten, wodurch die Kommunikation entlang eines zu beschreibenden Prozesses sichergestellt wird. Eine Anbindung an Ressourcen wie das Internet, Datenbanken, etc. erfolgt über Konnektoren [20110c].
OpenAdaptor OpenAdaptor stellt kein Framework zur Applikationsintegration zur Verfügung, sondern einzelne Adapter, mit denen einzelne Anwendungen miteinander gekoppelt werden können. Es handelt sich um Standardadapter, die für den jeweiligen Anwendungsfall über XML konfiguriert werden können. Der Datenaustausch über den Adapter zwischen den Anwendungen findet über Java Objekte statt. Abbildung 3 zeigt schematisch die Kopplung zweier Anwendungen mit OpenAdaptor unter Verwendung einer Lese-, einer Schreib- und einer Prozessorkomponente [20110d]. Die vorgestellten Frameworks und Lösungen zur Applikationsintegration haben das Ziel, Anwendungen entlang eines Geschäftsprozesses zu verknüpfen und als integrierte Lösung nutzbar zu machen. XML wird dabei häufig verwendet, um ein gemeinsames Datenschema und –modell abzubilden. Für den in Abschnitt 2 skizzierten Herstellungsprozess ist XML aufgrund der großen Menge an strukturierten Finite Element Daten nicht geeignet, um ein darauf basierendes Schema oder Modell zu verwenden. Des Weiteren sind die EAI Frameworks darauf aus-
Abb. 3 OpenAdaptor Kopplungsschema
656
Daniel Schilberg et al.
gerichtet, semantische Verknüpfung aus jeweils derselben oder einer verwandten Wissensdomäne zu realisieren [LOP05]. Für den in Abschnitt 5 beschriebenen Anwendungsfall ist es jedoch notwendig, eine semantische Integration über viele Wissensdomänen hinweg zu realisieren. Im folgenden Kapitel wird der auf semantische Kopplung ausgelegte domänenübergreifende Datenintegrator dargestellt. Der Datenintegrator wird am ZLW/IMA der RWTH Aachen University entwickelt.
4 Datenintegration Der entwickelte Datenintegrator verwendet zur Kopplung von Applikationen deren in Dateiform vorliegende Ein- und Ausgaben. Die Ausgaben der Applikationen sowie ergänzende manuelle Eingaben werden über einen Extract-Transform-LoadTransform-Prozess (ETLT-Prozess) in eine gemeinsame Datenhaltung integriert. Die für eine Anwendung benötigten Eingabedaten werden mit einem EnrichExtract-Transform-Load-Prozess (EETL-Prozess) aus der gemeinsamen Datenhaltung extrahiert. Abbildung 4 zeigt schematisch, wie zwei Simulationen auf diesen Weg verknüpft werden. Die gemeinsame Datenhaltung wird durch eine Datenbank realisiert, die analog zu den Datenschemata und–modellen der in Abschnitt 3 vorgestellten Frameworks zu verstehen ist [MSH09]. Die ETLT- und EETL-Prozesse übernehmen unter anderem auch die Aufgaben der in Abschnitt drei erwähnten Adaptern und Konnektoren. Im Folgenden werden die Funktionsweisen und die Aufgabe der beiden Prozesse eingehend erläutert. Im ersten Schritt des ETLT-Prozesses werden die Daten aus den vorliegenden Quellen extrahiert (Extract). Extraktionsquellen können Dateien oder Eingaben über eine Mensch-Maschine-Schnittstelle sein. Anschließend werden die extrahierten Daten in das der Datenbank zugrunde liegende Modell transformiert (Transform). Bei Finite-Elemente-Daten bedeutet dies, dass die Knoten-, Element- und Attributdaten einheitlich indiziert und bezeichnet werden, was ein semantisches Mapping der jeweiligen Bedeutung der Bezeichner erfordert. Hiernach werden die Daten in der Datenbank gespeichert (Load). Die erfolgreiche Validierung der integrierten Daten in einem Postprocessing bildet den Abschluss des ETLT-Prozesses. Im EETL Prozess werden über eine Abfolge von Teilprozessen Daten aus der Datenbank für eine Quelle extrahiert. Im ersten Teilprozess werden die Daten entweder in das erforderliche Format umgerechnet oder aus vorhandenen Daten neu berechnet (Enrich). Abbildung 5 zeigt die Funktionsweise des Enrich-Prozesses.
Abb. 4 Datenintegrationsschema des Datenintegrators
Enterprise Application Integration für die virtuelle Produktion
657
Abb. 5 Funktionsweise Enrich-Prozess
In der Datenbank liegen Informationen zu Temperatur, Dehnung, Spannung, Querschnittsfläche und Länge eines Bauteils vor. In dem in Abbildung 5 beschriebenen Beispiel benötigt die Zielquelle die Temperatur in K, das E-Modul und die Federkonstante. Die Temperatur kann direkt übernommen werden, das E-Modul und die Federkonstante müssen aber über den Einsatz von Transformationen generiert werden. Danach werden die Daten aus der gemeinsamen Datenhaltung extrahiert (Extract), um anschließend in die benötigte strukturelle Form gebracht zu werden (Transform). Der letzte Teilprozess ist das Schreiben der von der Zielquelle benötigten Eingabedatei (Load). Das Mapping zwischen den einzelnen Daten geschieht über die in der Datenbank hinterlegten Bedeutungen der einzelnen Felder, die Anwendungsbezogen abgelegt sind. Hiermit sind die vom Datenintegrator angewandten Prozesse zur Applikationsintegration beschrieben und es folgt nun eine kurze Darstellung des Integrationsprozesses heterogener numerischer Simulationen zur Abbildung eines durchgängigen Herstellungsprozesses anhand des Anwendungsfalls Line-Pipe.
5 Anwendungsfall Als Anwendungsfall für die Validierung des in Abschnitt 4 vorgestellten Datenintegrators wird die Simulation des Herstellungsprozesses einer Line-Pipe (vgl. Abbildung 1) betrachtet. Hierbei sollen die Anwendungen die Veränderungen auf der Makrostrukturebene betrachten und Mikrostruktursimulationen zu einer Simulationskette verknüpft werden. Die Simulationskette soll automatisiert ablaufen, alle Eingangsdaten sollen entweder zum Initialisierungszeitpunkt eingegeben oder zur Laufzeit aus den vorliegenden Eingabe- und Ausgabedaten durch den Datenintegrator generiert werden. Nach jedem Prozessschritt auf der Makrostrukturebene wird eine Mikrostruktursimulation durchgeführt, deren Ergebnisse von den
658
Daniel Schilberg et al.
Makrostruktursimulationen als Eingabedaten verwendet werden. Um die Ergebnisse aus der Mikrostrukturebene auf der Makrostrukturebene verwenden zu können, müssen diese mittels einer Anwendung zur Homogenisierung aufbereitet werden. Die Simulationen wurden jeweils als Insellösungen entwickelt und haben keine gemeinsamen Schnittstellen. Des Weiteren unterscheiden sich die jeweiligen Ein- und Ausgaben der Anwendungen in Syntax (Unterschiede in der technischen Darstellung von Informationen), Struktur (Unterschiede in der strukturellen Repräsentation von Informationen) und Semantik (Unterschiede in der Bedeutung verwendeter Begriffe und Konzepte). Zur Simulation des gesamten Herstellungsprozess werden fünf Anwendungen in folgender Reihenfolge eingesetzt (vgl. Abbildung 1): Zunächst wird mit der Software CASTS (entwickelt von ACCESS e.V.) das Glühen simuliert. Im Anschluss daran wird mit der vom Institut für Bildsameformgebung (IBF) der RWTH Aachen entwickelten Umformsimulation LARSTRAN der Warmwalzprozess dargestellt. Das folgende Abkühlverfahren und die spanende Bearbeitung des gewalzten Blechs zur Schweißvorbereitung werden erneut mit CASTS simuliert. Das Umformen der Bleche zu einem Rohr wird mit der Dassault Entwicklung ABAQUS abgebildet. Das Institut für Schweißtechnik und Fügeverfahren (ISF) der RWTH Aachen liefert mit SimWeld eine Schweißsimulation zur Darstellung des vorletzten Prozessschritts. Der letzte Schritt im simulierten Herstellungsprozess der Line-Pipe ist das Expandieren, das wiederum mit der Anwendung CASTS abgebildet wird. Nach jedem der genannten Schritte wird mit der Software MICRESS (ACCESS) eine Mikrostruktursimulation durchgeführt, um bessere Eingangsdaten für die Makrostruktursimulationen zu schaffen. Mit der Anwendung HOMAT (ACCESS) werden die Mikrostrukturdaten homogenisiert. Der Datenintegrator ist, wenn er automatisiert mittels einer Middleware aufgerufen wird, in der Lage, die Ein- und Ausgabedateien der zuvor erwähnten Simulationen nach den in Abbildung 4 und 5 dargestellten Schemata einzulesen, zu verstehen und entsprechend neue Dateien zu generieren. Zu diesem Zweck müssen neben der Eingabedatei der ersten Simulation (CASTS - Glühen) alle Parameter, die nicht durch den Prozess generiert werden können, bekannt sein.
6 Zusammenfassung und Ausblick In diesem Beitrag wurde ein Werkzeug zur automatisierten durchgängigen Kopplung von Anwendungen vorgestellt. Der Datenintegrator basiert auf den allgemeinen Konzepten der EAI [Hoh02] und wurde anhand des simulierten Herstellungsprozesses einer Line-Pipe validiert. Er stellt ein wichtiges Element bei der Etablierung einer virtuellen Produktion dar, da neben der virtuellen Darstellung von Fabriken und Anlagen auch Fertigungsprozesse präzise und Schritt für Schritt dargestellt werden können. So können mit dem Datenintegrator beispielsweise skalenübergreifend verschiedene Simulationsmodelle zu einem Gesamtprozess verknüpft werden. Ein nächster Schritt in Richtung virtueller Produktion ist nun, alle Simulationen, die im Bereich der Planung und Herstellung von Produkten eingesetzt werden, zu verknüpfen, d. h. nicht nur Prozess- sondern auch Maschinensimulationen und
Enterprise Application Integration für die virtuelle Produktion
659
Fabrikplanungstools zu berücksichtigen. Dabei stellen sich die folgenden zentralen Herausforderungen: Integration weiterer Wissensdomänen entlang der vertikalen Achse und Erweiterung der Möglichkeit zum Mappen von Syntax, Struktur und Semantik. Langfristiges Ziel ist es, neben der horizontalen eine vertikale Integration der Simulationen zu ermöglichen, um eine Betrachtung vom Rohstoff über dessen Verarbeitung bis hin zur verarbeitenden Fabrik durch verknüpfte Einzelsimulationen darzustellen.
www.oracle.com, downloaded 19.03.2010, 2010. www.ibm.com, downloaded 19.03.2010, 2010. www.jboss.org, downloaded 19.03.2010, 2010. www.openadaptor.org, downloaded 2010-03-19, 2010. F. Cummins. Enterprise Integration. Wiley Computer Publishing, 2002. Gulp Knowledge Base: Definition EAI. www.gulp.de/kb/pt/techexpert/mysap.html, downloaded 2010-03-19, 2010. [Hoh02] G. Hohpe. Enterprise Integration Patterns. In 9th Conference on Pattern Language of Programs, 2002. USA. [Lin00] D. S. Linthicum. Enterprise Application Integration. Addison-Wesley, 3rd edition, 2000. Boston. [LOP05] F. Losavio, D. Ortega, and M. Pérez. Comparison of EAI Frameworks. Journal of Object Technology, 4(4), June 2005. [MSH09] T. Meisen, D. Schilberg, and K. Henning. Planner Based Data Integration for Simulation chains in Virtual Production. In International Conference on Science, Technology and Innovation for Sustainable Well-Being (STISWB), 2009. Mahasarakham University, Thailand. [Mye02] M. J. Myerson. Enterprise Application Integration. 2002. Boca Raton. [OFS03] C. O’Rourke, N. Fishman, and W Selkow. Enterprise Architecture Using the Zachman Framework. Thomson Course Technlogy, 2003. [RMea01] W. Ruh, F. Maginnis, and et al. Enterprise Application Integration - A Wiley Tech Brief. John Wiley & Sons, Inc., 2001. New York. [Sch10] D. Schilberg. Architektur eines Datenintegrators zur durchgängigen Kopplung von verteilten numerischen Simulationen. VDI Fortschrittsberichte, 2010. Düsseldorf. [WBD04] J. Whitten, L. Bentley, and K. Dittman. Systems Analysis and Design Methods. McGraw-Hill, 6th edition, 2004.
Ontology Based Semantic Interconnection of Distributed Numerical Simulations for Virtual Production Daniel Schilberg, Tobias Meisen, Klaus Henning
Abstract The focus of this work is to promote the semantic interconnection of distributed numerical simulations for virtual production systems. The product planning phase usually requires a huge range of various numerical simulations that are used by different departments of a company. To increase productivity on the one hand and reduce complexity of simulation chains on the other hand, it is essential to be able to interconnect the differing syntax and semantics of the distributed numerical simulations. The interconnection is realized by an ontology based tool for data integration that is used within a simulation platform. Keywords Semantic Interconnection · Simulation · Ontology
1 Introduction A manufacturing company in a high-wage country has to position itself in two dilemmata. The first dilemma lies between mass production with a very limited product range (scale) and the manufacturing of a large variety of products in small quantities (scope). The second dilemma involves the dichotomy between planning and value orientation. The vertices of this dilemma shape the so-called polylemma of production technology (Figure 1) [SKBS07]. In order to create a lasting competitive advantage over low-wage countries, it is necessary to develop methods to reduce or dissolve the polylemma. With the intent to satisfy customer’s desires by offering a tailored product at the reduced costs of a mass production product, it is not enough to examine and optimize just a singular vertex; every change on one of the four vertices affects according to the dilemma the corresponding vertex. The Cluster of Excellence “Integrative Production Technology for High-Wage Countries” at RWTH Aachen University is engaged in the reduction of the D. Schilberg (B) ZLW/IMA & IfU, RWTH Aachen University, Dennewartstr. 27, 52068 Aachen, Germany e-mail: [email protected]
mentioned polylemma. In the integrative cluster domain (ICD) “Virtual Production Systems”, the issue of the second dilemma is discussed. One objective of the ICD is the development of a simulation platform for the interconnection of simulations that are used in product or production planning. The simulation platform builds up whole simulation chains according to a specific not predefined manufacturing process. The use of automatically interconnected simulations reduces the dilemma from planning orientation towards value orientation, whilst ruling out that there will be a reduction in the quality of planning. The challenge lies in combining dynamically monolithic simulations to represent an entire production process sensible for virtual production [SGH08]. This paper shows the software architecture that enables the simulation platform to interconnect simulations syntactically and semantically by using an ontology. Here process simulations, such as welding or forming of the assemblies and microstructure simulations designed to look at the fabric of the assembly, are considered. The focus lies on the components needed for the syntactic and semantic interconnection. Then on the basis of the following use case the application of the platform is described. The paper closes with a summary of the idea of the simulation platform.
2 Use Case The interconnection of simulations used in the manufacturing process of a gear wheel is considered for the validation of the simulation platform. The gear wheel is manufactured in several working steps (Figure 2). Two levels of detail are examined in the manufacturing process of a gear wheel [SP09]. The first level of detail
Ontology Based Semantic Interconnection
663
Fig. 2 Simulation chain “gear wheel”
is linked to macro-structure processes like forming, heat treatment and welding. The second level of detail is linked to the examination of the micro-structure. The micro-structure data is homogenized on a transfer level to allow its usage on the macro-structure level. Each simulation that is used in the manufacturing process must be docked to the simulation platform. That means that all used parameters of the five simulations must be represented by meaning and value. The meaning is stored in the simulation platform, while the values are provided by the simulations. By meaning the following is intended: unit, value range and context. In the use case, the blank will be prepared with a heat treatment for the recast process. In order to use the best material data, a microstructure analyses will be made after every step in the macrostructure process. With the results from the heat treatment the micro-structure data for a few representatives points of the blank will be determined. In order to use this information for the whole macrostructure, a homogenization tool is used to prepare the micro-structure data for the macro level. The results of the heat treatment and the microstructure simulation can be used separately or together as input for the following recast process simulation. This procedure is repeated for the next heat treatment and the welding process [SGH09]. The simulation platform gathers all parameters that are needed for a complete and automated run of the whole simulation chain through the user and the simulations. The user of the platform has to select the steps of the manufacturing process that should be simulated via the simulations. The user has to provide all data that is not generated by the simulations before the simulation chain can be started. A list of the missing data is provided by the simulation platform. The simulation platform will organize the order of the simulations to match the manufacturing process. Furthermore, it will provide the input data for each simulation. Via the components of the simulation platform, the syntactic and semantic simulation interconnection is realized. On the basis of this interconnection many manufacturing processes can be represented with the simulation platform simply by
664
D. Schilberg et al.
using the simulations in different orders. All parameters and the values generated by the simulations are available for every downstream simulation in the chain. Thus, in any further process step which will be simulated, the entire process history is recorded by the simulation platform; this ultimately leads to a better data set for all simulations. In the following section the software architecture of the simulation platform, with its main components for the semantic interconnection, is introduced.
3 Simulation Platform The system architecture for the simulation platform consists of the platform itself and the simulations that are linked to it. Figure 3 shows the system architecture of the platform. The platform consists of the components Web Interface, Visualization, Data Integrator and Middleware. The Web Interface is part of the Human Machine Interface which also includes the visualization. The user utilizes the Web Interface to operate the platform. The user can choose the process steps that should be simulated and the order on which these will be implemented. Data that is not generated by the simulations or producible by the Data Integrator component out of the simulations results must be entered by the user via the Web Interface. The visualization presents the simulation results, either individually or comparatively. For example, the results of the simulations can be displayed as single results or as interlinked results. For the interlinked representation all or a collection of results can be used. Through this representation, the user is able to visualize in parallel the effects of a change in the macro and microstructure at every time step of the manufacturing process. The middleware is used among other things for data exchange between the platform components and the platform, as well as the simulations. The components and the simulations are distributed through a network. The Data Integrator is required to create the syntactic and semantic interconnection between the simulations and
Fig. 3 Simulation Platform
Ontology Based Semantic Interconnection
665
to ensure that the chosen manufacturing process can be represented correctly by the simulation platform. The described components of the simulation platform are realized as services in a service oriented architecture [Erl08], this allows to distribute components of the platform in the network. In the following sections the components of the Data Integrator that are used for the syntactic and semantic interconnection are described. The components of the Data Integrator are a Data Management, a Data Enrichment, a Converter, a Data Repository and a Data Integration Control. The Data Repository is used to store the input and output data of the simulation and process related information. As an interface for the communication between the Data Integrator and the other components of the platform, the Data Integrator uses a Data Integration Control component. All the incoming and out going data from the Data Integrator is administrated by this component. The Data Integration Control coordinates the data transfer between internal components of the Data Integrator and the order and time of execution of these components. Because of the minor scientific interest in the functionality of the Data Repository and the Data Integration Control, the following sections will not describe these components in detail.
3.1 Converter Figure 4 shows the software architecture of the Data Integrator subsystem Converter. The Data Integration Control communicates the Converter Core component. The main objectives of the Converter are to convert simulation data in and out of a common format that is described in the following. Most simulations are developed as monolithic software with specific input and output data, format and structures. In order to compute the data of another simulation, the data must be converted to match the requirements of the subsequent simulation. Because the simulations are used in different manufacturing processes or in different sequences it is not practical to use a converter for each connection. This would increase the number of converters with every integrated simulation quadratic (n²-n). Instead of the adoption of many converters, two configurable converters are used. A parser converts the simulation data, which is provided in specific formats into a common format. A sequencer converts from the common format to the simulation specific formats. The Visualization Toolkit (VTK) [SML04] is used as the common format for the Data Integrator. By using the common data format, all simulation data is provided by the Data Integrator and can be processed by the Data Enrichment component without further conversions. To achieve these tasks the Converter needs the following components: A Converter Core as an Interface to the Data Integration Control which also serves to control the subsystem. A Configuration File Reader to read and interpret a file that contains information regarding any conversions which should be performed. A Data Input Reader to read and interpret the files that are generated by the simulations. A VTK Reader to read and interpret VTK files. A Data Output Writer to provide the files that are needed by the simulations. A VTK Writer to provide the resulting data
666
D. Schilberg et al.
Fig. 4 Converter
to the simulation platform. The Conversion Routines contained all needed routines to get from one format into another. Based on the data provided by the Data Integration Control, the Converter Core decides if a simulation file or VTK data must be computed. Depending on this the needed configuration file is processed. For every file format one configuration file is needed. After processing the configuration file, the Data Input Reader or VTK Reader gathers data that will be converted by the Conversion Routines. If new conversions are needed, because a new simulation was integrated in the platform, this component needs to be extended. The previously mentioned generation of data or any integration of former simulation results to a simulation input file on base of the semantic is done in the VTK format. In the following section the Data Enrichment component is described. The Data Enrichment realizes the semantic interconnection.
3.2 Enrichment The Data Enrichment (Figure 5) subsystem of the Data Integrator consists of the following components: A Query Processor, a Reasoner, a Knowledge Base Parser, a Knowledge Base that is represented by an ontology, a Compiler and a Planner. To interconnect the simulations with different semantics, a component is needed to realize a comparison of the used parameters; the Data Enrichment will provide this. All information concerning the parameters used by the simulations is stored in the Knowledge Base with the additional information value range, dependencies
Ontology Based Semantic Interconnection
667
Fig. 5 Data Enrichment
and context. Via the Data Integration Control the Query processor in cooperation with Data Management Control (cf. C. Data Management) will check if a certain parameter is generated by a simulation, is introduced manually or if it can be generated and made accessible through the Data Enrichment. For example one simulation generates a value for the parameter amperage (I), another simulation generates a value for the parameter voltage (U), but the next simulation needs a value for the parameter resistance (R). The Data enrichment can than provide the information that amperage and voltage can be transformed to resistance. The needed formula U/I = R is stored in the Transformation Library of the Data Management (cf. C. Data Management). The stored parameters are abstractly represented by an ontology [Tud08]. The Knowledge Base Parser provides the Reasoner with information out of the Knowledge Base. To compute a certain query, the Query Processor has to gather all necessary information from the Knowledge Base by using the representation structure of the ontology. This is done by the Reasoner. After all needed information is available; the Planner is used to identify and order the next steps to gain a result for the query. The Compiler computes the results of the Planner. Then the Query Processor can provide the Data Integration Control with the information that the requested data can be generated. A different case can be the identification of identical data this is represented by the simulations in different ways. In one simulation a material is represented by its chemical composition and in another simulation by its name (e.g. 20MnCr5 = Saarstahl). Without the necessary knowledge as to what this means, no direct conversion is possible. By using an ontology which represents this additional information, a transformation method can be chosen in order to provide the next simulation with this parameter [CN99]. Through the cooperation of the above mentioned components of the subsystem, Data Enrichment enables the Data Integrator to connect simulations not only on the level of syntax but also on the level of semantics. The ontology, from which the
668
D. Schilberg et al.
importance of individual parameters is derived, makes semantic mappings possible. Parameters will be linked for comparison respectively their values, based on their meanings. The comparison performed by the Data Enrichment will run through the following steps. First it will be checked whether the meanings directly match. If this is the case, the parameter values will be directly circulated to the Converter. The second step is only used when the first does not generate a result. If the Data Enrichment does not find a direct match, it must be verified if the required data can be generated out of existing data via the use of the Transformation Library. If so, the parameter values will be generated. Should this not be possible, the missing data must be made available by the user via the Web Interface. The data that is provided by the user is also stored in the Data Base of the Data Management [Lei05]. The next section describes the subsystem Data Management.
3.3 Data Management The component architecture of the subsystem Data Management is shown in Figure 6. The Data Management consists of a Database, a Data Management Control, a Storage Manager, an Indexer, a Data Extractor and the Transformation Library. Via the Data Integration Control, the core component of the Data Management, the Data Management Control, is executed. The task of the Data Management is to provide data and the transformation methods to generate missing data. To be able to solve this task, the Data Management has to index the data and to store the indices related to the data and to the information. If certain data and information has to be provided, the Data Management uses the Data Extractor component to identify the needed data and information, and extract these from the Database via the indices. After that the transformation methods like U/I = R from the example are used to generate the values of the parameters. The needed values will be integrated by the Converter (cf. A. Converter) in the files, that are used by the simulation. The next section will give a brief conclusion of the promoted Data Integrator as an ontology based semantic interconnection of distributed numerical simulations for virtual production.
4 Conclusion The need for semantic interconnection of distributed numerical simulations derives from the need of reducing the planning effort of production processes and therefore the requirement of flexible simulation chains. Investigation in this field means increasing the insight and understanding of the scientific and the technical processes, the product quality or the market advantages, as well as reducing production costs. This paper has introduced a concept of semantic interconnection of distributed nu-
Ontology Based Semantic Interconnection
669
Fig. 6 Data Management
merical simulations. It also presents how it can be realized by using a Data Integrator. The approach shown of semantic interconnection by using an ontology that represents the simulations parameter, value range and its meaning, has the advantage that all input and output data of the whole simulated process chain are correlated. This will prevent the loss of semantic information, as all data is linked to a certain step in the process of the simulation chain. This circumstance allows to identify the completeness of the requested data. If it is not complete, the system is able to localize the missing data and also specify where the missing data can be generated. The main research goal is to increase the understanding of how the semantic interconnection of distributed numerical simulations can be successfully implemented as basis for the realization of virtual production systems. The development of tools and techniques to support the interoperability of simulations along the process chain supports the achievement of the main goal. The simulation platform is the first step in implementing a virtual production system that contains the product planning, the process planning, the process itself, the process chain and the factory design planning [Paw08]. The simulation platform enables the user to observe the manufacturing process on a microstructure scale from the used material, to the linked manufacturing process steps. Acknowledgements The depicted research has been funded by the German Research Foundation DFG as part of the Cluster of Excellence “Integrative Production Technology for High-Wage Countries”.
670
D. Schilberg et al.
References [CN99] [Erl08] [Lei05] [Paw08] [SGH08]
[SGH09] [SKBS07] [SML04] [SP09] [Tud08]
M. Ciocoiu and D. S. Nau. Ontology-Based Semantics. NIST, 1999. T. Erl. SOA Principles of Service Design. 2008. Y. Lei. An Instance Mapping Ontology for Semantic Web. In Proceedings of the 3rd International Conference on Knowledge Capture, Banff Canada, 2005. G. Pawellek. Ganzheitliche Fabrikplanung. Springer-Verlag, Berlin Heidelberg Germany, 2008. D. Schilberg, A. Gramatke, and K. Henning. Semantic interconnection of distributed numerical simulations via SOA. In Proceedings of the World Congress on Engineering and Computer Science, page 894, Hong Kong, 2008. Newswood Limited. D. Schilberg, A. Gramatke, and K. Henning. Verkettung von Prozesssimulationen für die virtuelle Produktion. In 12. IFF Wissenschaftstage, Magdeburg Germany, 2009. G. Schuh, F. Klocke, R. Schmidt, and C. Brecher. Excellence in Production. Apprimus Verlag, Aachen Germany, 2007. W. Schroeder, K. Martin, and B. Lorensen. The Visualization Toolkit, Kitware, Inc. USA, 2004. G.J. Schmitz and U. Prahl. Toward a Virtual Platform for Materials Processing. JOM, pages 19–23, 2009. T. Tudorache. Ontologies in Engineering. VDM Verlag, Saarbrücken Germany, 2008.
Verkettung von Prozesssimulationen für die virtuelle Produktion Daniel Schilberg, Arno Gramatke, Klaus Henning
Zusammenfassung Ein Unternehmen in einem Hochlohnland muss sich in zwei Spannungsfeldern positionieren. Das erste Spannungsfeld liegt zwischen der Massenproduktion mit einem sehr eingeschränkten Produktspektrum (Scale) und der Fertigung von Produkten mit einem sehr großen Variantenreichtum bei geringen Stückzahlen (Scope). Das zweite Spannungsfeld umfasst die Dichotomie zwischen Planungs- und Wertorientierung. Im Rahmen dieses Beitrags wird eine Plattform entwickelt, mit der Simulationen, die im Produkt- oder Fertigungsplanungsprozess eingesetzt werden, zu Simulationsketten verknüpft werden können. Diese Verknüpfung soll den Aufwand für den Einsatz der Simulationen verringern und damit die Ausrichtung von der Planungsorientierung hin zur Wertorientierung ohne Einbußen bei der Planung verstärken. Schlüsselwörter Virtuelle Produktion · Semantische Kopplung · Simulation · Simulationsplattform
einen Eckpunkt singulär zu betrachten und zu optimieren, um den Wunsch der Kunden nach einem individualisierten Produkt zu den Kosten eines Massenprodukts zu erfüllen. Der Exzellenzcluster „Integrative Produktionstechnik für Hochlohnländer“ der RWTH Aachen University beschäftigt sich mit der Reduzierung des Polylemmas. In dem Teilcluster „Virtuelle Produktionssysteme“ wird die Problematik des zweiten Spannungsfelds thematisiert. Es wird eine Simulationsplattform entwickelt, mit der Simulationen, die im Produkt- oder Fertigungsplanungsprozess eingesetzt werden, zu Simulationsketten verknüpft werden können. Diese Verknüpfung soll den Aufwand für den Einsatz der Simulationen verringern und damit die Ausrichtung von der Planungsorientierung hin zur Wertorientierung verstärken, ohne dass es zu einer Reduzierung der Planungsqualität kommt. Die Herausforderung liegt in der Verknüpfung monolithischer Simulationen, um sinnvoll ganze Produktionsprozesse für die virtuelle Produktion abzubilden [SGH08]. In diesem Beitrag wird gezeigt, welche Anforderungen die Simulationsplattform erfüllen muss. Hierbei werden Prozesssimulationen wie z. B. das Schweißen oder das Umformen von Bauteilen sowie Mikrostruktursimulationen zur Betrachtung der Gefüge im Bauteil betrachtet, die im Anwendungsfall des folgenden Kapitels verwendet werden. Nach Festlegung der Anforderungen werden aus diesen die für die Plattform benötigten Komponenten abgeleitet. Mit einer Zusammenfassung schließt das Beitrag die Vorstellung der Simulationsplattform ab.
2 Anwendungsfall Der Herstellungsprozess eines Getriebezahnrads soll durch eine Simulationskette abgebildet werden. Das Getriebezahnrad wird in mehreren Arbeitsschritten gefertigt (Abbildung 2). Ein Rohling wird mehrfach wärmebehandelt und umgeformt, bis ein
Verkettung von Prozesssimulationen
673
Abb. 2 Anwendungsfall Getriebezahnrad
Zahnrad gefertigt ist. Anschließend erfolgen weitere Wärmebehandlungen, und ein Synchronring wird auf das Zahnrad aufgeschweißt; damit ist der Fertigungsprozess des Getriebezahnrads abgeschlossen. Die für diesen Fertigungsprozess benötigte Simulationskette wird in zwei Ebenen Simulationen erfordern. In der Makrostrukturebene werden Umform-, Wärmbehandlungs- und Schweißsimulationen verwendet. In der Mikrostrukturebene werden Simulationen zur Betrachtung der Mikrostrukturen eingesetzt. Des Weiteren werden auf einer Transferebene Mikrostrukturdaten für die Makrostrukturebene homogenisiert. Die eingesetzten Softwarewerkzeuge, die für die Simulationskette verwendet werden, sind monolithische Softwareanwendungen ohne gemeinsame Schnittstellen, Datenformaten, -strukturen und -repräsentation. Im folgenden Kapitel wird gezeigt welche Anforderungen sich aus der Kopplung dieser monolithischen Simulationen ergeben. Der Fokus liegt hierbei bei der Umsetzung der syntaktischen Verknüpfung (Datenformate und -strukturen) und der semantischen Verknüpfung (Datenrepräsentation und Bedeutung).
3 Anforderungen an eine Simulationsplattform Die Anforderungen an eine Simulationsplattform zur Verkettung von den in Kapitel 2 vorgestellten Simulationen teilen sich in funktionale und nichtfunktionale Anforderungen. Die wichtigsten Anforderungen fokussieren sich auf die syntaktische und semantische Kopplung der beteiligten Softwareanwendungen [SGH08]. Darüber hinaus werden eine Mensch-Maschine-Schnittstelle und eine Möglichkeit zur Anbindung von Simulationen an die Plattform benötigt. Die Anforderungen zu den beiden letztgenannten Punkten werden in diesem Beitrag nicht thematisiert. Basis der Kopplung der Simulationen ist es, die Ausgaben und Eingaben der eingesetzten Simulationen lesen, schreiben und deren Inhalt deuten zu können. Dafür muss auf der Syntaxebene ein Verständnis für die Dateiformate und -strukturen der Simulationsdaten (sowohl Eingabe als auch Ausgabe) vorhanden sein. Denn die
674
D. Schilberg et al.
Simulationsdaten können z. B. binär oder im ASCII-Format vorliegen und Schlüsselworte oder Offsets zur Sortierung der Inhalte verwenden. Auf der semantischen Ebene müssen die Bedeutungen von Parametern bekannt sein. Das folgende Beispiel zeigt, dass für eine Verknüpfung der Simulationen die folgenden beschriebenen Anforderungen essentiell sind. Parameter wie z. B. das Material werden in einer Simulation durch eine Zahl (z. B. 5) und in einer anderen Simulation durch die Materialbezeichnung (z. B. X2CrNi12) gekennzeichnet. Innerhalb der zu simulierenden Prozesskette muss auf alle bereits vorliegenden Simulationsdaten zugegriffen werden können, um Daten auszuwählen und zu manipulieren. Für die Speicherung der Simulationsdaten soll ein allgemeines Datenformat verwendet werden, damit Datenmanipulationen nicht die originalen Simulationsdaten verändern. Daher müssen alle Simulationsdaten in das und aus dem allgemeinem Datenformat konvertiert werden können. Hierfür müssen die Datenstrukturen der Dateien analysiert und verglichen werden. Innerhalb des allgemeinen Datenformats ist es für die syntaktische Kopplung notwendig, die von den Simulationen verwendeten bzw. erzeugten Parameter und deren Werte zu vergleichen. Für die semantische Kopplung müssen darüber hinaus die Parameter-bedeutungen gegenübergestellt werden können. Die Generierung von fehlenden Parametern, die aus bekannten Parametern errechnet werden können, muss ermöglicht werden. Die Kopplung der Simulationen zu einer Simulationskette muss automatisch ablaufen, um den Arbeitsaufwand für den Einsatz von Simulationsketten zu reduzieren. Daher ist es erforderlich, dass die zuvor beschriebenen Konvertierungen, Generierungen und Vergleiche innerhalb der Simulationsplattform ohne manuelle Arbeitsschritte vonstattengehen. Die nichtfunktionalen Anforderungen an die Simulationsplattform wurden in Workshops mit den Anwendern ermittelt und werden im Folgenden kurz zusammengefasst: Ein modularer Aufbau soll die Verwendung alternativer Komponenten ermöglichen. Erweiterbarkeit und Flexibilität sollen das Anbinden von weiteren Simulationen an die Plattform gestatten. Eine automatisierte Bedienung sowie ein hoher Grad an Konfigurierbarkeit der Plattform sind neben einer geringen Datenverarbeitungsdauer und einer hohen Zuverlässigkeit Anforderungen, die erfüllt werden müssen. Kapitel 4 stellt basierend auf den Anforderungen dar, wie die Systemarchitektur der Plattform gestaltet ist und mit welchen Komponenten die Erfüllung der Anforderungen konkret umgesetzt wird.
4 Simulationsplattform zur Verkettung von Prozesssimulationen 4.1 Architektur der Simulationsplattform Die Systemarchitektur für die Simulationsplattform wird in Abbildung 3 dargestellt. Die Komponenten der Plattform sind das Web Interface, die Visualisierung, der Daten-Integrator und die Middleware. Das Web Interface ist Teil der
Verkettung von Prozesssimulationen
675
Abb. 3 Systemarchitektur der Simulationsplattform
Mensch-Maschine-Schnittstelle zu der auch die Visualisierung gehört. Es dient der Bedienung der Plattform durch den Anwender; hier werden die Prozessketten, die simuliert werden sollen, zusammengestellt und Simulationsdaten, die nicht durch die Plattform bereitgestellt werden können, abgefragt. Die Visualisierung stellt die Simulationsergebnisse einzeln oder vergleichend dar. So können z. B. die Ergebnisse von parallel ablaufenden Simulationen auch parallel und verknüpft dargestellt werden. Die Middleware dient zum Datenaustausch zwischen der Plattform und den über ein Netzwerk verteilten Simulationen. Der Daten-Integrator wird benötigt, um die in Kapitel 2 spezifizierten Anforderungen an die syntaktische und semantische Kopplung zu realisieren. Des Weiteren speichert der Daten-Integrator über eine Datenmanagement-Komponente alle Simulationsdaten ab. Die Daten werden mit einer eindeutigen Identifikationsnummer (ID) markiert. Die Prozess-ID, der Prozessschritt und der jeweilige Simulationszeitschritt werden ergänzt, um die Produktionshistorie nachverfolgen zu können. Die Konverter und das Enrichment greifen über die Datenmanagement-Komponente auf die gespeicherten Informationen und Daten zu. Die Kommunikation des Daten-Integrators mit den übrigen Komponenten der Plattform wird durch eine Datenintegrationskontrolle gesteuert, die wie ein Datenbus die Ansteuerung der Daten-Integrator Komponenten administriert. Da die Komponenten Datenmanagement und Datenintegrationskontrolle administrative Aufgabe erfüllen, werden sie in diesem Paper nicht näher beleuchtet. Die in Abbildung 3 dargestellten Basiskomponenten des Daten-Integrators für die syntaktische und semantische Kopplung werden in den Kapiteln 4.2 und 4.3 in ihrer Funktionsweise detaillierter betrachtet. Der Daten-Integrator verwendet für diese Aufgabe Konverter (Parser und Sequenzer), die durch Konvertierung der Datenformate und -strukturen die syntaktische Kopplung realisieren. Die Enrichment-Komponente ermöglicht durch den Einsatz einer Wissensbasis und eines semantischen Mappers die semantische Kopplung [ABH+ 06].
676
D. Schilberg et al.
4.2 Konverter Monolithisch entwickelte Simulationen verwenden spezielle Ein- und Ausgabeformate mit bestimmten Datenformaten und -strukturen. Damit eine Simulation die Daten einer anderen Simulation einlesen kann, müssen diese aus den Ursprungsdatenformaten und -strukturen in die erforderlichen Formate und Strukturen konvertiert werden [ZG05]. Da je nach zu simulierender Prozesskette die Simulationen in unterschiedlichen Reihenfolgen innerhalb der Plattform eingesetzt werden können, ist es nicht sinnvoll, für die syntaktische Kopplung der einzelnen Simulationen untereinander jeweils einen Konverter zu verwenden, da dadurch die Anzahl der Konverter mit jeder weiteren zu integrierenden Simulation exponentiell (n2 -n) steigen würde. Stattdessen werden zwei konfigurierbare Konverter verwendet. Ein Parser konvertiert die Ausgabedaten der einzelnen Simulationen, die in speziellen Formaten vorliegen, in ein allgemeines Format. Ein Sequenzer wiederum konvertiert aus dem allgemeinen Format die speziellen Eingangsformate der jeweiligen Simulationen. Als allgemeines Format wird für den Daten-Integrator das Visualization Toolkit (VTK)-Format [kit] verwendet. Durch die Verwendung des allgemeinen Datenformats hat man darüber hinaus alle Simulationsdaten in einem Format vorliegen und kann damit wiederholt auf diese ohne weitere Konvertierungen für Datenmanipulationen zugreifen. In Abbildung 4 ist die Funktionsweise der Konverter dargestellt. Alle Konvertierungsroutinen, die für die im Daten-Integrator integrierten Simulationen benötigt werden, sind in einer Bibliothek für Parser und Sequenzer hinterlegt. Über eine Konfigurationsdatei, die für jede durchzuführende Konvertierung vorliegen muss, wird festgelegt, welche Konvertierungsroutinen verwendet werden. Der Parser liest die Konfigurationsdatei und die entsprechende Simulationsausgabedatei ein und schreibt alle Daten in das VTK-Format. Im VTK-Format können die Daten durch das im folgenden Kapitel beschriebene Enrichment weiterverarbeitet werden. Der Sequenzer liest analog zum Parser eine Konfigurationsdatei ein und generiert auf Basis dieser Konfiguration aus den Daten im VTK-Format die geforderte neue Simulationseingabedatei.
Abb. 4 Konfigurierbare Konverter
Verkettung von Prozesssimulationen
677
4.3 Enrichment Die Enrichment-Komponente des Daten-Integrators realisiert mit einem semantischen Mapper (Abbildung 5) die semantische Kopplung der Simulationen. Die Simulationen werden jeweils zur Darstellung durch eine Klasse repräsentiert. Die Klasse besteht aus dem Simulationsnamen, den Simulationsattributen (Parameterwerte) und den Simulationsalgorithmen, mit denen aus den Eingabeparametern die Ausgabeparameter berechnet werden. Diese Informationen werden in eine Wissensbasis überführt. Der semantische Mapper greift für Vergleiche auf diese Wissensbasis, die im Enrichment hinterlegt ist, zu [NM01]. Der Zugriff kann über einen Planer, eine Inferenzmaschine oder einen Hybriden aus Planer und Inferenzmaschine realisiert werden. Für den vorliegenden semantischen Mapper wird eine hybride Lösung verwendet. In der Wissensbasis sind Parametertabellen mit ihren Abhängigkeiten für alle im Daten-Integrator integrierten Simulationen in Form einer Ontologie hinterlegt [Tud08]. Aus dieser Ontologie kann die Bedeutung einzelner Parameter abgeleitet werden, womit es dem semantischen Mapper möglich ist, die jeweiligen Parameterwerte ihrer Bedeutung zuzuordnen und die Parameterbedeutungen der Simulationen, die verknüpft werden sollen, zu vergleichen. Für diesen Vergleich wird zuerst überprüft, ob die Bedeutungen unmittelbar übereinstimmen. Sollte dies der Fall sein, können die Parameterwerte direkt an die Konverter weitergegeben werden. Bei keiner unmittelbaren Übereinstimmung wird geprüft, ob die geforderten Parameterwerte aus den bereits vorliegenden Daten generiert werden können. Ist dies der Fall, so werden die Parameterwerte erzeugt. Sollte dies nicht möglich sein, müssen die Daten über das Web Interface durch den Anwender zur Verfügung gestellt werden. Die über das Web Interface ergänzten Daten werden ebenfalls gespeichert. Sämtliche Operationen des Enrichment verwenden Daten die im VTK-Format vorliegen. Die simulationsspezifischen Informationen Parametereinheit und -bedeutung und Prozesszusammenhang werden bei der Integration in die Simulationsplattform in die Wissensbasis überführt und durch das Enrichment auf sinnvoll übertragbare Werte überprüft. Damit liegen den Simulationen alle zuvor ermittelten Parameter und
Abb. 5 Semantische Kopplung
678
D. Schilberg et al.
Parameterwerte und die damit verbundenen Erkenntnisse vor. So wird bei jedem weiteren Prozessschritt, der simuliert wird, die gesamte Prozesshistorie berücksichtigt. Dies führt letztendlich zu einer besseren Datenbasis für alle eingesetzten Simulationen.
5 Zusammenfassung Der Bedarf nach einer semantischen Kopplung von verteilten, numerischen Simulationen entsteht aus der Notwendigkeit, den Planungsaufwand für Produktionsprozesse zu reduzieren. Hierzu müssen die geplanten Produktionsprozesssimulationen ein Höchstmaßan Flexibilität aufweisen. Die Forschung an der semantischen Kopplung zur Realisierung der Simulation von Prozessketten verspricht, den Einblick und den Erkenntnisgewinn über die technischen und wissenschaftlichen Zusammenhänge in der Produktion zu erhöhen und damit eine Verbesserung der Produktqualität und eine Reduzierung der Produktionskosten zu ermöglichen. In dieser Arbeit wurde die Architektur einer Simulationsplattform zur Verkettung von Prozesssimulationen vorgestellt. Die Realisierung der semantischen Verknüpfung der eingesetzten Simulationen reduziert die Gefahr von Erkenntnisverlusten. Die Simulationsplattform, in der eine erweiterbare modular aufgebaute Wissensbasis hinterlegt ist, kann um einzelne Simulationen und damit Produktionsprozessschritte ausgebaut werden. Dies ermöglicht es, weitere Herstellungsprozesse komplett zu simulieren.
Literaturverzeichnis [ABH+ 06]
[kit] [NM01]
[SGH08]
[SKBS07]
[Tud08] [ZG05]
K. Arnarsdóttir, Arne-Jørgen Berre, Axel Hahn, Michele Missikoff, and Francesco Taglino. Semantic mapping: ontology-based vs. model-based approach Alternative or complementary approaches? In Proceedings of the Open Interop Workshop on Enterprise Modelling and Ontologies for Interoperability, Luxembourg, 2006. Kitware Inc. - leading edge, high-quality software. http://www.kitware.com/. Natalya F. Noy and Deborah L. McGuinness. Ontology Development 101: A Guide to Creating Your First Ontology. Technical Report KSL-01-05, Stanford University, Stanford, 2001. D. Schilberg, A. Gramatke, and K. Henning. Semantic Interconnection of Distributed Numerical Simulations Via SOA. In Proceedings World Congress on Engineering and Computer Science 2008, Lecture Notes in Engineering and Computer Science, pages 894–897, Hong Kong, China, 2008. International Association of Engineers. Günther Schuh, Fritz Klocke, Christian Brecher, and Robert Schmitt. Excellence in Production: Festschrift für Univ.-Prof. Dr.-Ing. Dipl.-Wirt. Ing. Dr. techn. h.c. Dr. oec. h.c. Walter Eversheim. Apprimus-Verlag, 1st edition, August 2007. Tania Tudorache. Ontologies in Engineering: Modeling, Consistency and Use Cases. VDM Verlag, Saarbrücken, Germany, June 2008. L. Zhang and J. G. Gu. Ontology based semantic mapping architecture. In Proceedings of 2005 International Conference on Machine Learning and Cybernetics, Guangzhou, volume 1-9, pages 2200–2205, New York, USA, August 2005. IEEE.
Appendix A Monographs and Book Contributions Published at the ZLW/IMA & IfU
References [BHLJ10]
[BL09a] [BL09b] [Bra09a]
[Bra09b] [Bra10]
[BRH09]
[BWL10] [FHM09]
[GI09]
Ursula Bach, Frank Hees, Ingo Leisten, and Sabina Jeschke. Innovative Förderstrukturen im Förderschwerpunkt Präventiver Arbeits- und Gesundheitsschutz – das Metaprojekt StArG als Netzwerkmanager. In Klaus Hennning, Ursula Bach, and Frank Hees, editors, Präventiver Arbeits- und Gesundheitsschutz 2020: Prävention weiterdenken!, volume 63 of Aachener Reihe Mensch und Technik ARMT, pages 264– 277. Wissenschaftsverlag Mainz, 2010. Ursula Bach and Ingo Leisten. Leitfaden zur Gestaltung von Internetseiten im Präventiven Arbeits- und Gesundheitsschutz. online: http://www.starg-online.de/, 2009. Ursula Bach and Ingo Leisten. Leitfaden zur Pressearbeit im Präventiven Arbeits- und Gesundheitsschutz. online: http://www.starg-online.de/, 2009. Stefan Brall. In der Kürze liegt die Würze: Mit Microtrainings Lernen befördern. In Katharina Wiegmann, editor, Innovationen aus Europa: Zur Bilanz der Leonardo da Vinci-Pilotprojekte - Teil II. Impuls, volume 39, pages 216–220. Medienhaus Plump, Bonn, 2009. Stefan Brall. Überfachliche Kompetenzanforderungen in den Ingenieurwissenschaften. Eine Literaturanalyse. RWTH Aachen University, Aachen, 2009. Stefan Brall. Arbeitsbegleitende hochschuldidaktische Kompetenzentwicklung. In Ralf Schneider and Birgit Szczyrba, editors, Hochschuldidaktik aufgefächert - vernetzte Hochschubildung. Bildung - Hochschule - Innovation, Festschrift für Johannes Wildt., pages 140–149. LIT Verlag, Berlin, 2010. Stefan Brall, Anja Richert, and Frank Hees. Wissensaktualisierung durch Kurzlerneinheiten. In Klaus Henning and Christiane Michulitz, editors, Unternehmenskybernetik 2020 – Betriebswirtschaftliche und technische Aspekte von Geschäftsprozessen. Wirtschaftskybernetik und Systemanalyse., volume 25, pages 339–344. Duncker & Humblot, Berlin, 2009. Ursula Bach, Till Weinert, and Ingo Leisten. Transfermethoden im Präventiven Arbeits- und Gesundheitsschutz. im Druck, 2010. Aachen. Paul Flachskampf, Eckart Hauck, and Christiane Michulitz. Systematik zur prozessbegleitenden betriebswirtschaftlichen Gestaltung und Absicherung technischer Entwicklungsprojekte. In Klaus Henning and Christiane Michulitz, editors, Unternehmenskybernetik 2020 – Betriebswirtschaftliche und technische Aspekte von Geschäftsprozessen. Wirtschaftskybernetik und Systemanalyse, volume 25, pages 109–114. Duncker & Humblot, 2009. Harald Gerking and Ingrid Isenhardt. Kaizen (KVP) im Handel. In Klaus Henning and Christiane Michulitz, editors, Unternehmenskybernetik 2020 – Be-
S. Jeschke et al. (eds.), Automation, Communication and Cybernetics in Science and Engineering 2009/2010, DOI 10.1007/978-3-642-16208-4, Springer-Verlag Berlin Heidelberg 2011
679
680
References
triebswirtschaftliche und technische Aspekte von Geschäftsprozessen. Wirtschaftskybernetik und Systemanalyse., volume 25, pages 243–248. Duncker & Humblot, Berlin, 2009. [HGS09] Klaus Henning, Arno Gramatke, and Daniel Schilberg. Informatik im Maschinenbau - Software -Entwurf, Realisierung und Management. Aachener Reihe Mensch und Technik (ARMT). Wissenschaftsverlag Mainz, Aachen, 9 edition, 2009. [HH09] Frank Hees and Renate Henning. Entwicklung eines internetbasierten Strategiemanagers für KMU - verdichtete Identifikation unternehmerischer Handlungsbedarfe. In Jörn-Axel Meyer, editor, Management - Instrumente in kleinen und mittleren Unternehmen. Jahrbuch der KMU Forschung und - Praxis 2009, pages 51–64. JOSEF EUL VERLAG, Lohmar/Köln, 2009. [HHBH09] Klaus Henning, Frank Hees, Ursula Bach, and Alan Hansen. Yes, we can! Warum Deutschland den Kopf nicht in den Sand stecken sollte. In Dieter Spath, editor, Festschrift Gerhard Ernst: Arbeits- und Dienstleistungsforschung als Innovationstreiber. Bilanzen, Herausforderungen, Zukünfte., pages 51–59. Fraunhofer Verlag, Stuttgart, 2009. [HHL09] Klaus Henning, Frank Hees, and Ingo Leisten. Unternehmenskybernetik 2020 Betriebswirtschaftliche und technische Aspekte von Geschäftsprozessen. In Klaus Henning and Christiane Michulitz, editors, Unternehmenskybernetik 2020 - Betriebswirtschaftliche und technische Aspekte von Geschäftsprozessen. Wirtschaftskybernetik und Systemanalyse., pages 25–37. Duncker & Humblot, Berlin, 2009. [HJM09] Frank Hees, Sebastian Jursch, and Colin Messerschmidt. Veränderungen strategisch planen wie die Großen: Ein internetbasierter Strategieplaner für KMU. In Klaus Henning and Christiane Michulitz, editors, Unternehmenskybernetik 2020 - Betriebswirtschaftliche und technische Aspekte von Geschäftsprozessen. Wirtschaftskybernetik und Systemanalyse., volume 25, pages 249–253. Duncker & Humblot, Berlin, 2009. [HK09] Max Haberstroh and Peter Kochalski. Innovation im Netz - Produktentwicklung am Beispiel von Crystal Space. In Klaus Henning and Christiane Michulitz, editors, Unternehmenskybernetik 2020 - Betriebswirtschaftliche und technische Aspekte von Geschäftsprozessen. Wirtschaftskybernetik und Systemanalyse., volume 25, Berlin, 2009. Duncker & Humblot. [HLBH09] Klaus Henning, Ingo Leisten, Ursula Bach, and Frank Hees. Präventionsforschung und unternehmerische Praxis: Zwei Seiten einer Medaille. In Klaus Henning, Ingo Leisten, and Frank Hees, editors, Innovationsfähigkeit stärken - Wettbewerbsfähigkeit erhalten. Präventiver Arbeits- und Gesundheitsschutz als Treiber., volume 60 of Aachener Reihe Mensch und Technik ARMT, pages 12–30. Wissenschaftsverlag Mainz, Aachen, 2009. [HLHJ10] Max Haarich, Ingo Leisten, Frank Hees, and Sabina Jeschke. Langfristiges Verstehen durch kurzfristiges Missverstehen. Die Bedeutung der interaktiv-transkriptiven Störungsbearbeitung für den Transfer von Wissen. In Klaus Henning, Ursula Bach, and Frank Hees, editors, Präventiver Arbeits- und Gesundheitsschutz 2020: Prävention weiterdenken!, volume 63 of Aachener Reihe Mensch und Technik ARMT, pages 278–297. Wissenschaftsverlag Mainz, Aachen, 2010. [HM09] Klaus Henning and Colin Messerschmidt. Supportprozesse global effizient managen: Das jährliche IT-Assessement der Knorr Bremse AG. In Klaus Henning and Christiane Michulitz, editors, Unternehmenskybernetik 2020 - Betriebswirtschaftliche und technische Aspekte von Geschäftsprozessen. Wirtschaftskybernetik und Systemanalyse., volume 25, pages 163–168. Duncker & Humblot, 2009. [HR09] Frank Hees and Anja Richert. Strategischer Forschungstransfer - Aktuelle Herausforderungen und interdisziplinäre Lösungsansätze. In Klaus Henning and Christiane Michulitz, editors, Unternehmenskybernetik 2020 - Betriebswirtschaftliche und technische Aspekte von Geschäftsprozessen. Wirtschaftskybernetik und Systemanalyse., volume 25, pages 153–158. Duncker & Humblot, Berlin, 2009.
References [Her10]
681
Annette Hermanns. „House of Learning“ als Basis für Qualitätsmanagement gemäß EFQM für die praxisnahe universitäre Ausbildung in den Ingenieurwissenschaften, volume 197 of Fortschritt-Berichte VDI. VDI-Verlag, Düsseldorf, 2010. [IB09] Ingrid Isenhardt and Stefan Brall. Integration von Arbeiten und Lernen im Studium. Handlungskompetenz als Schlüssel zum Erfolg. In Robertson von Trotha, editor, Schlüsselqualifikationen für Studium, Beruf und Gesellschaft Problemkreise der Angewandten Kulturwissenschaft, volume 14, pages 497–508. Universitätsverlag Karlsruhe, Karlsruhe, 2009. [IP09] Ingrid Isenhardt and Leonie Petry. Innovation & Entrepreneurship - Kann man Quantensprünge managen? In Klaus Henning and Christiane Michulitz, editors, Unternehmenskybernetik 2020 - Betriebswirtschaftliche und technische Aspekte von Geschäftsprozessen. Wirtschaftskybernetik und Systemanalyse., volume 25, pages 377–381. Duncker & Humblot, 2009. [ISH+ 09] Johannes Isensee, Mischa Seiter, Tim Höhne, Sebastian Jursch, Daniel Neuhäuser, Olaf Sauer, and Andreas Trautmann. RFID-Innovationen in Produktion und Logistik - Eine Analyse zu Potenzialen des RFID-Einsatzes in Produktion und Logistik. In International Performance Research Institute. Research Paper Nr. 20, pages 28–34. IPRI gGmbH, Stuttgart, 2009. [Jae10] Verena Jänen. Ein transponderbasiertes Informationssystem für die Produktion hybrider Mikrosysteme in wandelbaren Produktionsnetzwerken, volume 808 of Fortschritt-Berichte VDI. VDI-Verlag, Düsseldorf, 2010. [Jak09] Sabine Julia Jakobs. Agiles Software-Reengineering mit Erfolg. In Klaus Henning and Christiane Michulitz, editors, Unternehmenskybernetik 2020 - Betriebswirtschaftliche und technische Aspekte von Geschäftsprozessen. Wirtschaftskybernetik und Systemanalyse., volume 25, pages 145–149. Duncker & Humblot, Berlin, 2009. [MM09] Thilo Münstermann and Christiane Michulitz. Culture Based Change. In Klaus Henning and Christiane Michulitz, editors, Unternehmenskybernetik 2020 - Betriebswirtschaftliche und technische Aspekte von Geschäftsprozessen. Wirtschaftskybernetik und Systemanalyse., volume 25, pages 237–241. Duncker & Humblot, Berlin, 2009. [RBH09] Uschi Rick, Wolfgang Backhaus, and Klaus Henning. AsIsKnown - Ein semantisch gestütztes Wissensflusssystem. In Klaus Henning and Christiane Michulitz, editors, Unternehmenskybernetik 2020 - Betriebswirtschaftliche und technische Aspekte von Geschäftsprozessen. Wirtschaftskybernetik und Systemanalyse., volume 25, pages 363–368. Duncker & Humblot, Berlin, 2009. [RTJ10] Anja Richert, Kerstin Thöing, and Sabina Jeschke. Erzeugung agiler Wissensprozesse in Förderschwerpunkten mittels einer interaktiven Diskussions- und Arbeitsplattform. In Klaus Henning, Ursula Bach, and Frank Hees, editors, Präventiver Arbeits- und Gesundheitsschutz 2020: Prävention weiterdenken!, volume 63 of Aachener Reihe Mensch und Technik ARMT. Wissenschaftsverlag Mainz, Aachen, 2010. [SBH08] Sanaz Schröder, Wolfgang Backhaus, and Klaus Henning. RELOAD - Die Wissensplattform für die Bau- und Heimwerkerbranche., booktitle = TrendBook E-Learning 2008/2009, Überblick über bedeutende E-Learning-Lösungen im deutschsprachigen Raum. Haufe Fachmedia, Freiburg, 2008. [Sch10] Daniel Schilberg. Architektur eines Datenintegrators zur durchgängigen Kopplung von verteilten numerischen Simulationen, volume 807 of Fortschritt-Berichte VDI. VDI Verlag, Düsseldorf, 2010. [SHB+ 09] Max Skorning, Christian Herrmanns, Stefan Beckers, Daniel Rörtgen, Sebastian Bergrath, Michael Protogerakis, Tadeusz Brodziak, Jörg Brokmann, and Rolf Roissant. Med-on-@ix: Aachener Forschungsprojekt zur Verbesserung der Patientenversorgung in lebensbedrohlichen Situationen. In Achim Jäckel, editor, Telemedizinführer Deutschland 2009, volume 10, pages 323–326. Minerva, Bad-Nauenheim, 2009.
682 [Sch10]
[SVBH09]
[TSHJ10a]
[TSHJ10b]
[WVRI10]
References S. Schröder. Eine Untersuchung zum Lernverhalten der heutigen Studierenden: Die Net Generation - Fallstudien zum Einsatz einer Wissenslandkarte als E-LearningTool im Ingenieurstudium -, volume 62 of Aachener Reihe Mensch und Technik, Wissenschaftsverlag Mainz, Aachen, 2010. Sanaz Schröder, Tobias Valtinat, Wolfgang Backhaus, and Klaus Henning. Ein interaktives Lern- & Wissensmanagement-Instrument: Die web- und semantik-basierte Wissenslandkarte. In Klaus Henning and Christiane Michulitz, editors, Unternehmenskybernetik 2020 - Betriebswirtschaftliche und technische Aspekte von Geschäftsprozessen. Wirtschaftskybernetik und Systemanalyse., volume 25, pages 369–374. Duncker & Humblot, Berlin, 2009. Sven Trantow, Katharina Schuster, Frank Hees, and Sabina Jeschke. Spannungsfelder der Innovationsfähigkeit. Internationales Monitoring im BMBF-Forschungs- und Entwicklungsprogramm A-L-K. In Klaus Henning, Ursula Bach, and Frank Hees, editors, Präventiver Arbeits- und Gesundheitsschutz 2020: Prävention weiterdenken!, volume 63 of Aachener Reihe Mensch und Technik ARMT, pages 310–332. Verlag Mainz, Aachen, 2010. Sven Trantow, Katharina Schuster, Frank Hees, and Sabina Jeschke. Spannungsfelder der Innovationsfähigkeit. Internationales Monitoring im BMBF-Forschungsund Entwicklungsprogramm A-L-K. In Klaus Henning, Ursula Bach, and Frank Hees, editors, Präventiver Arbeits- und Gesundheitsschutz 2020: Prävention weiterdenken!, volume 63 of Aachener Reihe Mensch und Technik, pages 310–332. Wissenschaftsverlag Mainz, Aachen, 2010. Florian Welter René Vossen Anja Richert and Ingrid Isenhardt. Network Management for Clusters of Excellence: A Balanced-Scorecard Approach as a Performance Measurement Tool. In The Business Review, volume 15, pages 171–178. Cambridge, 2010.