Page 2
CHAPTER 1: INTRODUCTION TO INFORMATION SYSTEMS 1.1 What is an Information System? Most of us experience the eff...
605 downloads
6490 Views
2MB Size
Report
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!
Report copyright / DMCA form
Page 2
CHAPTER 1: INTRODUCTION TO INFORMATION SYSTEMS 1.1 What is an Information System? Most of us experience the effect of different systems every day - getting to work or school using a transport system, for example, or making a call using the telephone system. Computerised systems have transformed the world in countless ways, not least in increased connectivity between people, as provided by the Internet and the World Wide Web, for example. To many people, information systems means computers -- and computers are indeed part of the world of most information systems professionals. But there is more to the story than simple technology. Information Information is the universal commodity, available to all -- but information adds value only when it meets a real need and when it is acquired, organized and disseminated in a systematic fashion". Consider the following diagramin figure 1.1: Data Context Information Context Knowledge Figure 1.1 This suggests that there is a process by which Information and then Knowledge is produced, somewhat analogous to a manufacturing process whereby a raw material, in this case ‘Data’, is refined and worked on to produce an end-product. But what are the components, which convert data into information?
Copyright © Genetic Computer School, 2002
Page 3 Data and information are an important part of a computer system. A computer exists to convert data into information. Data are unprocessed, raw facts, figures, statistics, concepts or instructions. To be useful, they must be converted into information. Information is processed data. Data have been analyzed, organized and summarized to create useful information. Information is data that has been processed so that an action or decision can be made from it. A computer exists to create this useful information by converting data into information through a process called data processing. Data processing is the process of manipulation, organization and analysis that is used to convert data into information. Data and information in a computer system are organized using a directory structure and stored in files on the storage medium in use. Data becomes information when it is used for a purpose, which is of value to the individual or organization. Take, as an example, the following list into which animals are divided: a. Those that belong to the Emperor b. Embalmed ones c. Those that are trained d. Suckling pigs e. Mermaids f. Fabulous ones g. Those that tremble as if they were mad
Copyright © Genetic Computer School, 2002
Page 4 As outsiders, the categorization of animals under these headings has no meaning or value whatsoever but in the context of the ancient Chinese Court where they were used, it presumably provided valuable information by which the Emperor managed his livestock portfolio! The above example (figure 1.1) is rather extreme but to be an effective information manager, an understanding of the purpose or 'context' in which information is to be used is essential. That purpose may be to assist the effective management of the organisation or it may have a wider purpose, for example, to produce information which can be 'sold' either as a product or a service. This is an important way of generating profits and, indeed, information related services account for a higher, percentage of Gross Domestic Product in the U.K. than the manufacturing sector. In the same way that one manufacturer's product is more attractive to a customer because it has features which suit their specific requirements, so too, information will have characteristics which add value, depending upon the purpose for which it is required. Some of these are listed below: Quantity
The amount of information provided should be adequate for the purpose - not so much that the key information is lost, or so little that it does not present a complete picture.
Suitability
It should be appropriate to the skills and competencies of the manager who will use it and in a form which makes it 'user friendly'.
Scope
The breadth of information supplied will be in accord with the purpose for which it is to be used, for example, a population forecast will use census statistics over several decades.
Relevance
The subject matter which the information covers is the same as that which the manager is addressing.
Accuracy
As accurate as possible but, in some circumstances, not at the expense of timeliness - sometimes it's better to be 90% accurate
Copyright © Genetic Computer School, 2002
Page 5 than 100% out of date. Timeliness
It should be available when required.
Compatibility
The information is based upon standards which also apply to other information systems, for example, the accounting year as opposed to the calendar year.
Presentation
The information is presented in an appropriate style, for example, high quality printing and graphics in the case of an Annual Report.
Managing information is primarily a question of assessing the context and purpose for which it is to be used; and then deciding the relative importance of each of the above. For example, annual financial reports need to be both accurate and presented in a form that makes them intelligible to non-financial personnel. In the case of internal budgetary statements, presentation may be less important but timeliness increasingly critical if potential overspends are to be dealt with quickly and effectively. Information Systems Information systems collect, process, store, and distribute information so that it can be used by people. The information may be about people, places, things, or events inside an organization or in the environment that surrounds it. People use the information to make decisions, to keep track of resources, and to plan for the future. Information has attributes of accuracy, credibility, and timeliness - old news is not news! An information system can be defined technically as a set of interrelated components that collect (or retrieve), process, store, and distribute information to support decision making and control in an organization. In addition to supporting decision making, coordination, and control, information systems may also help managers and workers analyze problems, visualize complex subjects, and create new products. Copyright © Genetic Computer School, 2002
Page 6 Information systems contain information about significant people, places, and things with the organization or in the environment surrounding it. By information we mean data that have been shaped into a form that is meaningful and useful to human beings. Data, in contrast, are streams of raw facts representing events occurring in organizations or the physical environment before they have been organized and arranged into a form that people can understand and use. Three activities in an information system produce the information organizations need for making decisions, controlling operations, analyzing problems, and creating new products or services. These activities are input, processing, and output. Input captures or collects raw data from within the organization or from its external environment. Processing converts this raw input into a more meaningful form. Output transfers the processed information to the people or an activities where it will be used. Information systems also require feedback, which is output that is returned to appropriate members of the organization to help them evaluate or correct the input stage. Formal information systems can be either computer-based or manual. Manual systems use paper and pencil technology. These manual systems serve important needs. Computer-based information systems (CBIS), in contrast, rely on computer hardware and software technology to process and disseminate information. From this point on, when we use the term information systems we will be referring to computer-based information systems -- formal organizational systems that rely on computer technology. Although computer-based information systems use computer technology to process raw data into meaningful information, there is a sharp distinction between a computer and computer program on the one hand, and an information system on the other. Electronic computers and related software programs are the technical foundation, the tools and materials, of modern information systems. Computers provide the equipment for storing and processing information. Computer programs, or software, are sets of operating instructions that direct and control computer
Copyright © Genetic Computer School, 2002
Page 7 processing. Knowing how computers and computer programs work is important in designing solutions to organizational problems, but computers are only part of an information systems. Housing provides an appropriate analogy. Houses are built with hammers, nails, and wood, but these do not make a house. The architecture, design, setting, landscaping, and all of the decisions that lead to the creation of these features are part of the house and are crucial for finding a solution to the problem of putting a roof over one's head. Computers and programs are the hammer, nails, and lumber of CBIS, but alone they can not produce the information a particular organization needs. To understand information systems, one must understand the problems they are designed to solve, their architectural and design elements, and the organizational processes that lead to these solutions. Today's managers must combine computer literacy with information systems literacy. 1.2 Computer Information System The Computer Information System is an ensemble of hardware and software applications. The Information system is an analysis of a problem domain. It is "information oriented". For example, a banking compensation system or a telecommunication protocol are "information systems" in our sense. Their definition does not imply the use of specific software or hardware packages. Today the computer information system is called by a growing number "Information system" - maybe they find that it sounds smarter! So that company hires Oracle specialists for their "information system"! Whatever the name, the flavor is the same. Please note that in our definition information systems are not platform specific (hardware, software, language etc...). They might be called also domain systems, abstract systems, general systems etc. The interesting fact is that these systems may be seen also as "knowledge objects", one of the two views related to knowledge management. This view enables the way to "conceptual objects reuse".
Copyright © Genetic Computer School, 2002
Page 8 The Computer Information System is derived from the Information System by taking into account the platform specifics. 1.3 Why does information have to be managed? ♦
There is so much more of it these days!
The sheer volume of information available through advances in computing and communications means that managers have to be clearly focused on what information is important to them and what is not. Recent articles in newspapers have commented on the problems associated with too much information and new medical conditions such as 'information fatigue' and 'information overload’ have started to enter our vocabulary. The latest lap-top computers can hold and process vast amounts of data but the capacity of our 'neck-top' computer has not altered over the last few hundred years! ♦
It is a valuable asset in most organizations
Information-based organizations now generate more wealth than manufacturing industries in the U.K. In some organizations the only product they sell is information and that is often purchased by other organizations which add further value to the information before selling it on. It is essential, therefore, that such a valuable product is managed efficiently. ♦
It can provide a competitive advantage in the market place
Information facilitates chance Exposure to new ideas and concepts from outside the business can stimulate managers to adopt and adapt these concepts in their own business environment. There is a danger for managers, who stay in one job with one company for a long time, that they become insular in their approach and resistant to change. The working environment, as we move into the 21st Century, requires a workforce that is multi-skilled and able to adapt and keep up to date with change. There is much that organizations themselves can do to ensure that their employees have access to new information and learning opportunities. It is also the case that employees will
Copyright © Genetic Computer School, 2002
Page 9 have to take greater charge of their own personal development in the new organization structures being created in the Information Age.
♦
Information enables managers to manage effectively
If managers do not have the right information about how the business is performing then they are unlikely to be able to make the right decisions (or are more likely to, make the wrong ones). Managers need: • Essential decision making information in a usable form • Information when they want it (at a time that suits their workflow) • Real-time information (so decisions are timely and relate directly to what is happening) • To see trends developing (such as cost or time blowouts, teams not performing) • Information to manage cross functional teams doing many projects and tasks • To be notified of any changes that are likely to affect them 1.4 Roles people play in an information environment Applications support organizational goals and such goals are held by those for whom the success of the firm is critical; for this discussion, they are called stakeholders. This group includes boards of directors, shareholders, unions, communities, and suppliers. The goals represent desirable conclusions for clients. Clients may be inside or outside the organization. The classes of individuals that have direct interaction with the application are called direct users. Assisting these role players are developers, who may work for an organization, a client or a user, and they may be direct employees or contracted specialists.
Copyright © Genetic Computer School, 2002
Page 10
CHAPTER 2: FUNDAMENTAL TYPES OF INFORMATION SYSTEM 2.1 Transaction Processing Systems (TPS) Introduction A transaction Processing System is an exceptionally powerful and flexible transaction processing environment for both mobile and fixed-base data collection equipment. It provides the sophisticated connectivity needed in today's complex industrial, warehousing, distribution, and other environments. Transaction Processing System has numerous features important in creating systems and solutions for demanding transaction processing requirements. With Transaction Processing System, you can: • Develop high-performance data collection and bar code printing systems across client/server networks. • Revitalize legacy software and hardware in a state-of-the-art transaction processing and bar code environment Interface with different types of host and network computers at the same time. • Access multiple local and remote distributed databases, concurrently and simultaneously. • Select and use terminals and printers from multiple leading manufacturers, all in one system. • Combine 5250 and 3270 intelligent terminal emulation with data collection functionality using a GUI development environment that involves no host code rewrites. • Run the same transactions on your radio-frequency devices that you run on fixed devices and PCs, including intelligent emulation.
Copyright © Genetic Computer School, 2002
Page 11 • Select from a variety of automatic, dynamic servers and data backup schemes, which are available to ensure minimum downtime and maximum productivity. Transaction Processing System's modular architecture (refer to figure 2.1) allows you to choose the functionality that you need - you can always add on additional functionality as your requirements expand. Transaction processing systems [TPS] support the organization's daily activities and maintain the majority of the organization's internal data. These systems usually employ simple but highly repetitive processes but require the capability to deal with high volumes of transactions with great accuracy and high security. In the past, transactions were completed using batch processing, but today most transactions are completed immediately using on-line transaction processing [OLTP] systems.
Figure 2.1 Transaction Processing Systems Architecture To implement on-line transaction processing, master files containing key information about important business entities are placed on hard disk drives where they are directly accessible. The transaction files containing information about activity concerning these business entities, such as orders placed by customers, are also held in on-line files until they are no longer needed for everyday transaction processing activity.. This assures that the transaction data is available to all applications, and that all data is kept up-to-the-minute.
Copyright © Genetic Computer School, 2002
Page 12 Users with the proper database permissions may download data from the database to another location and upload changes to these on-line files as soon as changes are required. When the transaction files are no longer being "actively" used, they are not deleted. Instead they are archived in a special data storage facility called a data warehouse. A Transaction Processing System (TPS) is designed for Operating Management. It is also called Electronic Data Processing (EDP). It has a limited area of functionality, and is designed for one area of the business. It uses only data that is internal to the company. There are some possible exceptions, however. Inter -branch banking uses TPS as well as direct deposit of employee's pay from various companies and government institutions. Reports and other outputs from Transaction Processing Systems The outputs of TPS provide operational details, summary reports, and exception reports, which help supervise and control routine operations. Exception reporting provides a feedback loop to the operational manager about any unusual or unexpected activity that may require attention. Transaction processing systems maybe divided into major functional areas. Accounting provides systems to handle general ledger, payroll, and accounts. Marketing has TP systems to handle sales, promotion and advertising activity. Manufacturing requires transaction processing systems and operational feedback from production processes. Manufacturing reporting systems must often be designed as real-time systems to allow operational managers and supervisors to closely monitor on-going operations. Human resources needs systems to support its daily operations in recruiting, maintaining employee records, providing benefits, and monitoring occupational conditions. Finance requires maintaining information on cash reserves and investment holdings, and the monitoring of changes to tax and fiscal regulations. Many systems, such as order processing, may involve processes that cross funtional boundaries in the enterprise. For example, an order may concern sales,
Copyright © Genetic Computer School, 2002
Page 13 manufacturing, distribution, and accounting. Transaction processing systems may depend upon each other in intricate ways that are not always apparent to the users of these systems. When transaction data grows old, it can still be of value to the organization. However, it is not used frequently and takes up valuable disk space, so such data is often moved to other lower cost storage areas. In the past, archived data was difficult to access because the data archives were not well organized. Today, such archives are placed in data warehouses, which are data storage facilities that are managed by special database programs designed for that purpose. Analysts can then make use of on-line analytical processing [OLAP] tools to investigate either the current transaction data or the data that has been retired to the data warehouse facility. Using the company's data archives to determine trends and to extract other useful information is often referred to as data mining. 2.2 Management Information Systems (MIS) MIS stands for Management Information Systems. MIS is a branch of Management that involves the development, maintenance and understanding of Computer Information Systems from a business perspective. It is difficult to say exactly what this field of management entails because, like the computers and technology the field involves a rapidly changing and expanding area. Much like accounting, marketing and financing, the Management of Information Systems has become a crucial function in the operation of modern day business. Now and in the foreseeable future, the success of a company will increasingly depend on the quality of its Information Systems and technologies. MIS is a unique field in that it is able to combine the uses of technology while at the same time having an appreciation for the business environment as a whole. A widely recognized device for organizing types of management information systems uses the MIS Framework. This framework is based upon two dimensions, the level of managerial responsibility and the degree of problem structure. Structured problem solving is handled by a variety of reporting levels. Detailed operational reports are produced by transaction processing systems [TPS]. These in turn provide data to more
Copyright © Genetic Computer School, 2002
Page 14 highly summarized managerial reports used at higher levels. Management information systems [MIS] produce standard "MIS" reports in printed and on-line "soft copy' versions to assist middle management in their tactical decision making roles involving the allocation of resources and the oversight of company operations. Executives at the top rung of the corporation may have executive information systems [EIS], also called executive support systems [ESS] to supply their special needs for structured reporting. For less structured problems, and most often at higher management levels, decision support systems play a role in modeling from a wide variety of sources, both internal and external. Such problems are often called unstructured problems although that term can be misleading. The structuredness of a problem must be viewed as a continuum running from the fairly unstructured to the highly structured types of problems. No problem that can be analyzed by any technique can be considered as totally unstructured. Management Information Systems (M.I.S.) is a career area which focuses on two related topics: • Organizations - the business processes, and people as problem solvers and decision-makers. • Technology - current information technology so vital for any organization. The MIS field deals with all the information and problem solving activity of a modern, successful organization. The MIS discipline brings together the various business areas, computer science, and quantitative analysis techniques. This program provides the theory and methodology to analyze, design, implement, and manage an organization's information technology and systems. Management Information Systems (MIS) is a valuable new aspect of every business organization. Basically, MIS is looking at what types of computer systems organizations need and adapting the systems for their needs. The MIS field revolves around the fact that information and knowledge is one of the most valuable business
Copyright © Genetic Computer School, 2002
Page 15 assets. Without proper information channels, a business would lose its advantage over other companies. In the past, many people were turned off by the idea of using a personal computer in the everyday workplace. Many more people are starting to get comfortable around computers. New programs are always emerging that combine powerful applications with ease of usage. For instance, Microsoft, easily the biggest computer software provider in the world, has numerous software programs that even a beginner can tackle without much commotion. MIS specialists must take and apply their knowledge of both computers and people to the common business. They need to design programs that appeal to and motivate employees. MIS specialists must learn an overview of the organizational, strategic, and technical issues surrounding the management of information in businesses today. By building a solid base at businesses everywhere, MIS specialists help prepare future managers to manage information as a resource and to identify opportunities for using information as a competitive advantage. Management Information Systems (MIS) are used in businesses to help make decisions. They are an implementation of computer based processing and/or manual procedures yielding useful and timely information for decision making. A MIS is composed of procedures. In a MIS, procedures are sequences of steps to co-ordinate processing. Some are performed by staff and some by computer specialists. A business has many procedures, we refer to four of the common ones below: • Payroll • Personnel • Accounting • Inventory A Management Information System (MIS) is designed to help Middle Management decisions.
Copyright © Genetic Computer School, 2002
Page 16 It is an information reporting system, which receives data from all sections of the business. The system is connected to a more complex network than a Transaction Processing System. A MIS uses internal data as well as some external data. 2.3 Decision Support Systems (DSS) The group of tools and techniques, together with their definitions and descriptions of their interrelations, that assist in the specification and recording of an analyst's judgments about evidence, and inferences going into a hazard identification decision. A DSS is a tool (e.g. computer programme) or process (such as a best management practices procedure), whose utility is related to its ability to support decision-making in the real world situation and not to provide the solution itself by representing a system involving those real-world variables (physical, economic and social) and their interrelationships, whose complexities can be simplified for inclusion in the tool or process, where these variables and interrelationships can be customized to comply with individual situations and, where these variables and inter-relationships can be customized and/or interfaced with individual decision-makers, more particularly with their objectives, resources and constraints. Decision Support Systems can be differentiated from other types of information system in that they are directed towards use by users who are skilled in their subject area and who are supported rather than placed by the use of the computer. Decision support system users have a good knowledge of the problems area of the system and therefore need to be able to interact effectively with a flexible system which provides the type of support required. Decision Support Systems (DSS) is a dynamic and rapidly changing field which touches on a wide range of computing topics. Decision Support Systems have been defined as:
Copyright © Genetic Computer School, 2002
Page 17 Computer Based Systems are systems that help decision makers confront ill structured problems through direct interaction with data and analysis models. The emphasis is on problem solving tasks, which are semi-structured, ie. they combine human judgement with the use of computing tools and techniques. DSS do not replace managerial judgement but rather provide support for decision making - the final agent remains the human. Computer applications for management support are increasing and the availability of microcomputers has dramatically increased the number of systems on managers desks. 2.4 Office Automation Systems (OAS) Office Automation involves the planned application of integrated information handling tools and methods to improve the productivity of people in office operations. Although the handling of information by office people is the focus of this new technology, other aspects of the office will be affected. These include factors such as the organization of functions and lines of reporting, training for new methods, work space design, travel patterns, branch office location, home vs office work, hours of work, employee morale, and job classifications. Organizations that harness office automation products will need to deal with many more than just technological issues. About 22% of the US work force is now in the office, with that percentage rising. Labor costs account for about 70% of the total office costs in our economy and salary costs are increasing about 6% each year. During the past 15 years there has been relatively little increase in productivity of the office work force, in contras with the manufacturing sector where the average productivity has more than doubled. The cost of new technology aimed at increasing the productivity of office workers is going down, while the capabilities of office automation systems have been rapidly increasing. Office automation will impact industry and government organizations in very significant ways with both COST DISPLACEMENT and VALUE-ADDED results.
Copyright © Genetic Computer School, 2002
Page 18 These two terms are now being used by people who are considering the potential payoffs of office automation to their organizations. Cost displacement applications have the objective of achieving overall reductions in support staff costs or of increasing work volumes without adding support staff. Such applications typically center around WORD PROCESSING and provide the base for the more advanced value-added applications. Value-added applications are viewed as being directed toward improving managerial and professional staff productivity (and effectiveness) through use of more integrated office automation systems that can directly affect their work. The value-added approach deals with far more fundamental issues than the replacement of some support staff positions with word processing pools. Its focus upon individuals and groups of managers and professionals as targets for productivity improvement brings with it opportunities for significant increases in organizational effectiveness and major cost benefits in the largest segment of the office cost spectrum. For any organization, management choices at many levels will affect the balance between reducing total office costs (cost displacement) or increasing the total office effectiveness (value-added effects). Word processing applications have, until recently, been equated with the term office automation. It is interesting to note that, on the average, typing tasks comprise only 30% of the secretaries' and typists' work -- and thus account for only about 2% of the total office salaries. The next few years will see a very rapid growth in the introduction of advanced technology into offices and in applications with more impact on managers and other non-clerical people, bringing with it broadened perceptions of what office automation really includes. Office Automation is likely to become one of the fastest-growing and most significant new industries of the century. It will apply electronic technology to a broad new set of applications and bring significant change to many of the ways in which
Copyright © Genetic Computer School, 2002
Page 19 people and organizations work -AUGMENTING their capabilities and increasing both the quantity and quality of their contributions. 2.5 Executive Support Systems (ESS) An enterprise management information system encompasses the information flows in an entire organization. The applications for MIS fall into five categories; localized applications, interdepartmental applications, business process redesign, business network redesign, and business scope redefinition. Interdepartmental information systems integrate the activities of different departments into a single business system that produces appropriate coordinated responses to the enterprise's environment. Both localized and interdepartmental systems involve little redesign of the underlying business processes; they support the existing processes. Business process redesign requires the nature of business processes to be redesigned as the information system development progresses; it becomes a vehicle for change and improvement. Business network redesign concerns itself with how multiple enterprises work together; consideration is given as to how information is structured and handled as it crosses organisational boundaries. Business scope redefinition consists of applications that change the nature of the business. Interdepartmental information systems use a shared or corporate database facility. Procedures and practices are consistent and coordinated by the nature of the database structure. Departments are provided with different views of the information but the underlying data is consistent. Data captured in one department may be the source of information for the activities of another department. Because such systems are standardized, changes to data and the information produced have to be carefully controlled. There is a need for enterprise systems management.
Copyright © Genetic Computer School, 2002
Page 20 Interdepartmental management information systems differ from workgroup systems in several ways. Workgroups will know each other whereas users of an interdepartmental system are distant and the only common factor may be the use of an information source, they are separated both physically and organisationally. Workgroups may number in the tens of members, interdepartmental systems could supply the information needs of hundreds. Data contained in interdepartmental systems tends to be more complex and heterogeneous in nature. Whilst workgroup systems tend to be highly functional, indepartmental systems cross functional boundaries coordinating the information needs of several process related departments. Business process design is emerging as a necessary step to optimize the use of information technology. Rather than automating existing processes, the underlying business processes are redesigned to take advantage of the new technologies. Business process redesign can have dramatic impact for an enterprise; This is attributed to the change of culture, the redesign of work, the retraining or the loss of staff. Business network redesign refers to the use of information systems to enable groups of enterprises to interact more productively. Four types of network redesign can be identified; electronic data interchange (EDI), in which organisations agree on a common set of data standards, interenterprise system access in which organisations use each other's systems, interenterprise system integration in which enterprises develop shared information systems and knowledge networks use information technology. Business scope redefinition produces applications, which changes the nature of an enterprise's activities. New technologies can make it possible to enter new markets, enable new products or extend or enhance the capabilities of existing ones.
Copyright © Genetic Computer School, 2002
Page 21
CHAPTER 3: ORGANIZATIONS AND INFORMATION SYSTEM Introduction Organization
MIS Framework - Nolan & Wetherbe Figure 3.1 Effectiveness of Organizations It could be said that the role of the organization is to support creative individuals while providing an environment where they can create knowledge. But why is it important for organizations to provide this supportive environment? Well, one only has to look at the changes that have taken place on a macro-economic scale, in the creation of wealth; the reduced contribution by the manufacturing-based industry sector and corresponding expansion of the service and information sectors. This shift was identified in the early 1960's, and the term "knowledge worker" was born. Since that time, knowledge has increasingly been viewed as a resource with knowledge workers playing a pivital role in many organizations. An organization is defined as a systematic arrangement of people to accomplish some specific purpose. Organizations share three common characteristics.
Copyright © Genetic Computer School, 2002
Page 22 a. Has a distinct purpose. b. Is composed of people. c. Develops a systematic structure that defines and limits the behavior of organizational members. All effective organizations are not necessarily efficient. Effectiveness is concerned with goal attainment, while efficiency is concerned with resource usage (refer to figure 3.1). An organization can achieve its goals, but do so by being wasteful and using an inordinate amount of input resources. 3.1 Workgroup Information Systems A workgroup is a collection of people who work together to achieve a common goal. In homogeneous workgroups, everyone in the workgroup fulfills the same role. As such workroups grow so does productivity, in 'a" linear fashion. Providing information systems for these workgroups can be very cost effective since everyone carries out similar work. In heterogeneous workgroups there are several roles and jobs covered by individuals within the group. For this reason productivity does not increase in a linear progression when additional people join the group. The increase in productivity is more related to the previous knowledge of the individuals joining the group. Training and communication becomes an important issue. Since there are many roles and jobs covered by a heterogeneous group the provision of information systems can be expensive and difficult to support. Workgroup norms can play an information role in the acceptance or rejection of certain information systems. The influence of workgroup attitudes has a large part to play, peer pressure can persuade or deter acceptance, fears can be allayed or reinforced. Workgroups can be permanent or temporary. They can exist on single sites or be distributed throughout an organization.
Copyright © Genetic Computer School, 2002
Page 23 Workgroup effectiveness can be measured by output, personal satisfaction of the members, and by the group capacity for further cooperation. Effectiveness is determined by group effort, knowledge and skill. The strategies and approaches used to perform work are important. The major difference between personal and workgroup information systems is the need for workgroup systems to control the shared use of resources. This must be achieved without inconvenience to individual members of the group. Granularity refers to the size of the shared information resource. Large granularity means that group members share large information sources with the possibility of delays due to contention problems but with small information systems adminstration overheads. Conversely, small granularity produces few contentions but administration overheads rise accordingly. The major categories of a workgroup information system are hardware sharing and data sharing facilities. Hardware sharing applications allow members of the workgroup to share many hardware devices, some of these devices may be uneconomical for a single user system. Systems that focus on communications are applications such as group E-mail systems that provide a number of mailboxes that can be for individual mail or group and project mail. These systems can be passive or dynamic, the user or group has to access incoming mail in the static system, whereas, in the dynamic system the user is told that mail has arrived. Group E-mail systems are designed to enhance group coordination.
Group
conferencing
including
electronic
bulletin
boards,
videoconferencing with white boarding, help to ensure coordination and often support group dynamics in problem-solving settings. Collaborative writing systems let users share the writing of documents. These applications coordinate and control the users efforts. The tedious copying and altering of documents in a single user environment is avoided.
Copyright © Genetic Computer School, 2002
Page 24 Analytical support for workgroups is typically provided by workgroup spreadsheets. These are often composed of spreadsheet templates, the coordination of these spreadsheets can be complex. To make these work effectively it is essential that good documentation and training is provided. Group Decision Support Systems are applications which allow individual users to use group workboards in their own decision making process. Group decision making can be facilitated by a controller with the group using support applications. Another important category in workgroup information systems is the retrieval of information, including group database management systems, workflow automation systems, group scheduling systems, group project management systems and shared textbase systems. Group database management systems have similar functionality to personal database systems but are more complex and tend to concentrate on workgroup coordination. Workflow automation applications provide automatic routing of forms used in routine work processing. They also track the progress of forms. Groups can often improve processes by looking at the output from workflow systems. Trucking and monitoring is provided by similar systems provided for personal information systems but they are more complex. Project management has more complexity with the focus on group coordination. Group textbase systems are similar to personal systems but shared documents are indexed for group or individual retrieval. 3.2 Personal Information Systems Personal computer systems are used to help an individual facilitate their work. This facilitation can be considered as support for communication, analysis and tracking and monitoring.
Copyright © Genetic Computer School, 2002
Page 25 Typical communication support is provided by word processors, which can be used to produce simple text documents or more sophisticated documents containing a wide range of colours, fonts and even graphics. The functionality of these programs can often provide templates, boilerplate and automated document production. They also feature mailmerge and support scripting and macro languages to extend the customization of document production. Within the scope of communication support lies the desk-top publishing application and the presentation application. Both can produce professional quality output that can be shown on computer slideshows or be printed. Analytical support is given by such applications as the electronic spreadsheet capable of modelling quite complicated business activities. Many such programs are supported by scripting and macro languages and can support graphics and database type functionality. These applications are often enhanced to provide connectivity to various internal and external data sources to enable data extraction. Other types of analytical programs are the operational research ones providing linear programming and other methods. Tracking and monitoring support is given by database and project management applications. A database application includes a database, some means of data entry, reporting facilities and possibly applications programs. Users can interface with many databases using default, customized or programmed access. A common interface standard is SQL, structured query language. Project management applications keep track of a project's progress with regard to the allocation of resources and the management of tasks. A variety of techniques are used, typically, Gantt charts, critical path graphs and the like. Integrated application provide an environment in which all applications can work statically or dynamically with each other. Changes made to database application will dynamically update speadsheet models. In the Microsoft Windows environment this is achieved using DDE, dynamic data exchange and OLE, object linking and embedding.
Copyright © Genetic Computer School, 2002
Page 26 3.3 The System Development Process
Systems Development Life Cycle Define Problem
Develop/ Code Figure 3.2 The System Developement Life Cycle System Development Life Cycle From concept to production, you can develop a database by using the system development life cycle, which contains multiple stages of development. This topdown, systematic approach to database development transforms business information requirements into an operational database. (Refer to figure 3.2) Strategy and Analysis Study and analyze the business requirements. Interview users and managers to identify the information requirements. Incorporate the enterprise and application mission statements as well as any future system specifications. Build models of the system. Transfer the business narrative into a graphical representation of business information needs and rules. Confirm and refine the model with the analysts and experts.
Copyright © Genetic Computer School, 2002
Page 27 Design Design the database based on the model developed in the strategy and analysis phase. Build and Document Build the prototype system. Write and execute the commands to create the tables and supporting objects for the database. Develop user documentation, help text, and operations manuals to support the use and operation of the system. Transition Refine the prototype. Move an application into production with user acceptance testing, conversion of existing data, and parallel operations. Make any modifications required. Production Roll out the system to the users. Operate the production system. Monitor its performance, and enhance and refine the system. The SDLC procedures ensure that timely and accurate information concerning the progress of system development is available to stakeholders and others in the university community. System Development Life Cycle refers to a methodology for developing systems. It provides a consistent framework of tasks and deliverables needed to develop systems. The SDLC methodology may be modified to include only those activities appropriate for a particular project, whether the system is automated or manual, whether it is a new system or an enhancement to existing systems. The SDLC methodology tracks a project from an idea developed by the user, through a feasibility study, system analysis and design, programming, pilot testing,
Copyright © Genetic Computer School, 2002
Page 28 implementation and post-implementation analysis. Documentation written during project development is used in the future when the system is reassessed for its continuation, modification, or deletion. Alternative Quality Process - an alternative quality process is a systematic sequence of tasks for planning, organizing, and monitoring for quality performance improvement. It begins with customer requirements, focuses on root causes or barriers to improvements, and ensures that decisions and actions are based on real data; it includes problem solving processes such as the plan-do-check-act model. Detailed descriptions are available in the materials located in the quality center or available through team training. System - an organized collection of independent tasks and processes that is designed to work together in order to accomplish specific objectives. The processes and tasks typically receive input(s) from and provide output(s) to other processes and tasks and even other systems. The tasks and processes may or may not be supported by automation. The system development life cycle (SDLC) is an organized approach to obtaining an information system. The SDLC is formalized in many organizations, with detailed instructions outlining reporting requirements, specific tasks that must occur in each phase, and individual responsibilities. Major objectives of system development are to create a system: with the desired capabilities, within budget, and on time. The SDLC assessment process involves three components: 1. Understanding the current level of achievement (baseline assessment), 2. Defining the goal or achievement objective, and 3. Developing a plan to achieve the defined goal. The System Development Life Cycle (SDLC) is intended to provide a set of guidelines for the successful completion of application system development projects. The SDLC consists of seven distinct Phases as shown in Figure 3.3 below: Copyright © Genetic Computer School, 2002
Page 29 Phase 1
Phase 2
Planning
Definition
Phase 3 Analysis
Phase 4 Design
Phase 5 Build
Phase 6
Phase 7
Transition Warehouse
Each Phase contains one or more individual deliverables associated with the Phase. Figure 3.3 Planning This Phase of the SDLC is required to determine the feasibility of a particular project proceeding, or not. This Phase will produce a high-level overview document of the proposed project. It will contain information relating to the project's requirements and will enable the formalization and definition of the scope of the project. This is the first Phase in the SDLC. The Project Charter deliverable is the first major deliverable in the project. The Project Charter is to be produced prior to the commencement of a project. Due to fiscal year budget allocations and fiscal contract restrictions the Project Charter will typically refer to project activity up to the end of the fiscal year in which the project starts. There may however be exceptions and special circumstances when a Project Charter will span more than one fiscal year. In these cases, this should be specifically stated in the Project Charter document. For projects spanning more than one fiscal year, an updated Project Charter will be required for each year that the project is undertaken, prior to the start of the fiscal year (normally done in March of each year). The updated Project Charter for a subsequent fiscal year(s) will mainly focus on identifying the scope and deliverables of the project for that particular fiscal year. The updated Project Charter will also include revisions (as appropriate) to project team, budget, schedule and project status. The Business Champion is responsible for producing the Project Charter. However, the Business Champion may delegate this
Copyright © Genetic Computer School, 2002
Page 30 activity for completion. Together with appropriate members of the User Team, the Business Champion will provide information for the completion of the Project Charter. The Project Sponsor is responsible for reviewing and approving the Project Charter and securing the necessary funding for the project The draft Project Charter must be submitted to the Information Systems Branch (ISB) for Business Analyst review and QA against standards. The completed Project Charter will be submitted to the Executive Management (Steering _ Committee) for approval and budget allocation authorization. Definition This Phase of the SDLC defines exactly what, who, when and how the project will be carried out. This Phase will take the deliverable from the previous Phase (Project Charter), expand on the high-level project outline and provide a specific and detailed project definition. This Phase is the first activity of the project after obtaining approval and funding to proceed. The description of this Phase here assumes that the following associated activities have been completed: • the preparation and distribution of an RFP; • the selection of a contract development team; and • the appointment of a Project Manager. This Phase provides an effective way of communicating to project stakeholders, the project scope and schedule as well as any risks or constraints related to the project. This is the second Phase in the SDLC but the first Phase of the project itself. The Project Statement deliverable from this Phase must be completed and signed-off prior to commencing the next phase of the project. Copyright © Genetic Computer School, 2002
Page 31 Agreement to the Project Statement ensures that everyone involved in the project is clear on the project scope, objectives, goals and outcomes. The Project Manager is responsible for producing the Project Statement. However, the Project Manager may delegate this activity to the Development team for completion. The Business Champion, Business Analyst and appropriate members of the User Team will provide project information which will support the completion of the Project Statement. The draft Project Statement must be submitted to the Information Systems Branch for Business Analyst review and QA against standards. The completed Project Statement will be submitted to the Business Champion and Project Sponsor for approval and sign-off. Problems can be viewed as unsatisfied needs. They can be seen as opportunities to develop new and better methods, procedures and if appropriate computer systems. The definition phase should be used to examine all aspects of the problem and to come up with a clear definition of the problem. The scope of problem definition varies. In an informal setting simply specifying business application requirements may be sufficient. In a more formal case the product of the definition phase is a FUNCTIONAL SPECIFICATION - a precise description of the functional requirements of a computer system. The definition phase can be viewed as a series of steps; • Getting to know the application. • Deciding whether a computer system can meet the application's requirements and how it will do so. • Defining processing requirements constraints, costs and benefits. • Developing functional specifications and prototypes. Copyright © Genetic Computer School, 2002
Page 32 • Reviewing the specifications and prototypes. In some instances the steps leading to the functional specification and evaluation might be quite informal. The systems analyst uses the simple techniques of questioning and comparison. By asking questions, the analyst can consult references that have already addressed the problem or can talk to people with more experience in the problem area under consideration. By making comparisons the analyst will be able to recognize and isolate those parts of the problem that are familiar. In other instances, as in a large organization where many projects are in competition for management's attention more formal analytical techniques, such as feasibility studies, top-down design and structured analysis are required. The more people that need to interact in the decision-making process the more formal the analysis and specification of the functional requirements will have to be. One important aspect of functional specifications is their value as a communication tool between Analysis This Phase of the SDLC is required to understand and document the users' needs for the system. This Phase will document, in significantly more detail than the Project Statement, the scope, business objectives and requirements of the current/proposed system. The emphasis throughout this Phase is on what the system is to do. During the analysis and specification, the technical aspects and constraints should be considered, but should not be influenced by implementation characteristics. The technical aspects of the system are addressed in the Design Phase. During this Phase the Data Conversion requirements, at a high level, will become known. This will commence a parallel set of SDLC Phases for the Data Conversion associated with the system. Data Conversion will follow Phases 3 to 6 of the SDLC. Depending on the size and complexity of the total project (system and data conversion) Copyright © Genetic Computer School, 2002
.w
Page 33 the Data Conversion Application could be either incorporated as a section or component of the main system Design deliverable, or a separate Data Conversion deliverable of its own. In addition, the Data Warehousing requirements will also be identified during this Phase and a parallel set of Data Warehouse SDLC Phases should commence. Depending on the business requirements of the system, it is possible that the Data Warehousing aspects will only be considered once the system has been operational for a while. These parallel streams of SDLC are shown in more detail in Figure 3.4 below. Figure 3.4: SDLC Parallel Processing Streams
System Development Life-Cycle Phases
Data Conversion SDLC Phases
Data Warehouse SDLC Phases
The Detailed System Analysis is to be stated in a language that reflects the background of the user base. Most requirements are written as natural language sentences supplemented with diagrams, graphics and tables of detailed information. The Project Manager is responsible for producing the deliverables associated with the Detailed System Analysis. However, the Project Manager usually delegates this to Copyright © Genetic Computer School, 2002
I
Page 34 the Business Analyst. This deliverable has significant input from the Business Champion and User Team. In cases where the production of some or all of the Phase's deliverables have been delegated, the Project Manager will still maintain overall responsibility for the production of quality deliverable(s) submitted to the Business Champion for review and sign-off and ISB for Quality Assurance. The Project Manager will provide initial Quality Assurance of the deliverable prior to review by ISB QA, User Team and the Business Champion. The draft Detailed System Analysis must be submitted to the Information Systems Branch for Data Administration review and QA against standards. The completed Detailed System Analysis will be submitted to the Business Champion and Project Sponsor for approval and sign-off. Design This Phase of the SDLC continues on from the Detailed System Analysis and describes how the proposed system is to be built. The Design is specific to the technical environment that the system will be required to operate in and the tools to be used in building the system. The results of this Phase will significantly impact the Build and Transition Phases of the system. The Project Manager is responsible for producing the deliverables associated with the Detailed System Design. However, the Project Manager usually delegates responsibility for some or all of these deliverables to the Development Team. In cases where the production of some or all of the Phase's deliverables have been delegated, the Project Manager will still maintain overall responsibility for the production of quality deliverable(s) submitted to the Business Champion and ISB for Quality Assurance. The Project Manager will provide initial Quality Assurance of the deliverable(s) prior to review by ISB QA, User Team and the Business Champion.
Copyright © Genetic Computer School, 2002
Page 35 The draft Detailed System Design must be submitted to the Information Systems Branch for Database Administration review and QA against standards. The completed Detailed System Design will be submitted to the Business Champion and Project Sponsor for approval and sign-off. Build This Phase of the SDLC deals with the development, unit testing and integration testing of the system (application) modules, screens and reports. In addition, this Phase will address the preparation and establishment of the technical environment for development, testing and training of user representatives. This Phase is usually carried out in parallel with the development of user procedures and user documentation from the Transition Phase. Both of these will be required for module testing, upon the completion of the Build Phase. Coordination of the activities of the Build and Transition Phases is a key responsibility of the Project Manager at this time. Any special procedures for data conversion and/or data warehousing are also developed and tested. The processes of developing and testing of data conversion and data warehousing modules is no different from those required for the system itself. This is the fifth Phase in the SDLC. The Project Manager is responsible for producing the deliverables associated with the Build Phase. However, the Project Manager usually delegates responsibility for some or all of these deliverables to the Development Team. In cases where the production of some or all of the Phase's deliverables have been delegated, the Project Manager will' still maintain overall responsibility for the production of quality deliverable(s) submitted to the Business Champion, User Team and ISB for Quality Assurance. The Project Manager will provide initial Quality Assurance of the deliverable(s) prior to review by ISB QA, User Team and the Business Champion. Copyright © Genetic Computer School, 2002
Page 36 The draft System Build must be submitted to the Information Systems Branch for technical review and QA against standards. The completed System Build will be submitted to the Business Champion and Project Sponsor for approval and sign-off. Transition This Phase of the SDLC is to prepare for and carry out the transition of the developed system through user and acceptance testing to a full production system. This is the sixth (and in some cases last) Phase in the SDLC. This will be the last Phase only for those Business Systems that (for specific documented reasons) will not make its data available in the Data Warehouse. This Phase will provide users with the documentation and training to effectively use the system. Although the Data Conversion will only to be done once, user documentation will also be required. Individual system components that successfully completed unit and integration testing during the Build Phase are now subjected to a more rigorous system and acceptance testing as defined by the testing plans. In addition user and operation procedures are to be tested for the system. As appropriate, the Data Conversion will be performed prior to the finalization of the system into Production. The detailed Data Conversion and Transition Plans will define in more detail exactly how this will be accomplished. The Project Manager is responsible for producing the deliverables associated with the Transition Phase. However, the Project Manager usually delegates responsibility for some or all of these deliverables to the Development Team. In cases where the production of some or all of the Phase's deliverables have been delegated, the Project Manager will still maintain overall responsibility for the production of quality deliverable(s) submitted to the Business Champion, User Team and ISB for Quality Assurance. Copyright © Genetic Computer School, 2002
Page 37 The Project Manager will provide initial Quality Assurance of the deliverable(s) prior to review by ISB QA, User Team and the Business Champion. The draft deliverables must be submitted to the Information Systems Branch for technical review and QA against standards Warehouse This Phase of the SDLC addresses the publication of the system's data into the Ministry's Data Warehouse for business manipulation and decision support. Although described as one Phase here, the Warehouse Phase actually comprises, as appropriate, all the deliverables associated with SDLC Phases 2 [Definition] through 6 [Transition]. This is the final Phase in the SDLC. The Project Manager is responsible for producing the deliverables associated with the Warehouse Phase. However, the Project Manager usually delegates responsibility for some or all of these deliverables to the Development Team. In cases where the production of some or all of the Phase's deliverables have been delegated, the Project Manager will still maintain overall responsibility for the production of quality deliverable(s) submitted to the Business Champion, User Team and ISB for Quality Assurance. The Project Manager will provide initial Quality Assurance of the deliverable(s) prior to review by ISB QA, User Team and the Business Champion. The draft deliverables must be submitted to the Information Systems Branch for review and QA against standards.
Copyright © Genetic Computer School, 2002
Page 38
CHAPTER 4: EVOLVING APPLICATIONS OF INFORMATION TECHNOLOGY The application of IT has evolved and is continuing to evolve through 3 stages: Automation of work Information management Business transformation This evolution involves major leaps in the complexity of tasks that it is being designed to perform. As we review this evolution, a consistent pattern of change emerges in the business application of IT. As we evolve automation of work through information management to business transformation, the strategic importance of IT applications increases and that amount of organizational change required to realize the benefits of an application is also greater. Specifically, an increasing number of changes are being made to elements of the business system beyond IT such as business processes, organisational structure and even business culture. At the same time, the number and complexity of applications (or potential applications) also increases. The three stages of evolution are summarised in the table 4.1 below. 4.1 Stages of IT Evolution
Stage Automation of work
Impact
Benefit
• Getting work Done •
Doing
Examples
Operational
• Payroll
efficiency
• Check processing
the
same things more efficiently
• Basic order processing • Basic airline reservation systems
Copyright © Genetic Computer School, 2002
Page 39 Information
• Restructuring
management
work
and
Operational and
• Customer
work Tactical effectiveness information systems
processes
• Airline yield
• Doing things
management systems
differently
• Executive information systems
Business
• Defining the
Strategic effectiveness • JIT inventory
transformation
business
and positioning
systems
• Doing different
• Electronic
things
commerce
• Changing the
•
OLAP
business/industry rules
Table 4.1 • Automation of work The first application of IT in business involved the automation of work tasks such as census data calculations, check processing and payroll as well as basic order processing and reservation systems. An automation application such as payroll, is not a strategic application, and while it is certainly necessary, and failure to pay employees would have serious consequences, it is not an application that provides strategic advantage. The benefits were largely in the area of operational efficiency. A few new jobs were created to program, operate and support the technology itself and some manual jobs, such as pay calculation and check processing, were replaced. There was also limited change to people's jobs or to business processes, but the overall change to the nature of work was not significant. Learning requirements were relatively simple and narrow, focused on how to use the technology. Change was generally limited to one or a small number of functional areas. In the case of payroll, little if any change was experienced Copyright © Genetic Computer School, 2002
Page 40 outside of the payroll department. The most important thing was that the payroll application ran correctly. The self-sufficiency of computerization was reflected by the physical reality of the mainframe, where computers were isolated behind data center walls - and operated invisibly by IT experts. Applications were often limited to those conceived by same experts, with little understanding of technology by broader business community, or how it could be applied. • Information Management Automation applications created information as a by-product of automating work. In the early years of the automation stage, this information was not generally used, certainly not in widespread format way. As we moved into the information stage, the opportunities to use this information began to be recognized. With the wider distribution of desktop computer terminals, IT was increasingly applied to provide information to support improved decision making, to move it "close to the customer" and to support new service and product design. Here the introduction of advanced order processing systems, airline yield management systems, customer information systems as well as the start of so-called executive information systems (EIS). Benefits moved beyond operational efficiency to operational and tactical effectiveness. Information could be used to make tactical, and in some cases, strategic decision. Initially, information was used to enable workers to do their jobs better. Their jobs changed somewhat, but primarily they were required, and had to be trained to take largely predetermined action based on the information provided. As the information stage advanced, benefits were premised on workers improving how they analyzed and applied information to their work. In the case of order processing, seasonal variations in demand might be noticed, and adjustments made to order levels. In the case of customer information systems (or customer information files in financial institutions), information was used to increase the value of "customer moments" by cross-selling and target marketing of certain services. Airline systems moved beyond basic reservation systems to sophisticated Copyright © Genetic Computer School, 2002
Page 41 yield management systems. In the later steps of the information stage, automation information bases provided opportunities to design new products, such as today's multitude of mutual funds and numerous volume-based discount plans for valued customers. It was no longer sufficient to simply provide the application and make sure that it worked as specified. For these benefits to be realized, the nature of people's work had to change. Business processes had to be restuctured and better integrated. Reward systems had to change. Significant learning was required. The changes crossed functional boundaries, and in some cases, changed or eliminated them. Physically, personal computers emerged from behind the walls of - the central data center. PCs began to appear everywhere in organizations and to be operated by nonexperts. The number of potential applications of technology increased dramatically. Many of these were conceived outside of the IT world, by the broader community of business managers and front-line technology users. • Business transformation Information management applications enable organisations to rethink and redesign their business processes and how they carry out their business. As more and more computing power is distributed, and as advanced communications capabilities continue to erase the constraints of time and distance, the very nature of businesses, and even entire industries, is being redefined. Benefits have moved beyond operational and tactical effectiveness to strategic effectiveness and positioning. Business transformation applications, such as just-in-time (JIT) inventory systems and advanced electronic commerce, enable organizations to rethink not just how they do things, but also what they choose to do. For example, JIT ordering and inventory management systems are fundamentally changing the value/supply chain and shifting the balance of power among stakeholders. The emergence of Internet and virtual banking is redefining the financial industry by removing century-old barriers to entry and blurring financial product boundaries. Airlines are now offering passengers direct access to reservations
Copyright © Genetic Computer School, 2002
Page 42 systems and more fighting for ownership of the client with travel agents, and thus redefining the travel agent business. Amazon.com is helping to redefine the book industry. It is not only selling books electronically and offering a wider selection than is possible in physical bookstores, it is using the power of computers to repackage - and eventually transform - a range of services that were historically spread across multiple businesses, including the reference capabilities of libraries, the retail display and selection expertise of bookstores, the efficiency of volume discount distributors and the knowledge of professional book reviewers. All these applications have significant strategic implications. While technology enables these benefits, most of the required change is beyond the realm of IT. Changes are required in the mission, the very raison d'etre of organizations. The organizations carrying out these changes will, in many cases, be redrawing traditional industry boundaries or, at minimum, changing industry structures and rewriting industry ground rules. In doing so, they will harness IT with the aim of "competing for the future". The potential of transformational applications is tremendous, but to realize it will present organizations with new and significant challenges. Automating payroll processing was primarily an engineering question, whereas creating a virtual bank branch or bookstore is primarily a business one. As we moved beyond automation of work to information management and business transformation applications, sound management of IT projects remains necessary but is no longer by any means sufficient. In the case of financial customer information files, for example, employees have to learn new skills, assume new responsibilities and accept different reward systems. Cross-selling to banking customers means astutely interpreting customer profile information and cultivating personal relationships, rather than efficiently handling transactions and answering routine questions to shorten waiting time in a branch line-up. Such changes in management practice and work place culture are just as important as
Copyright © Genetic Computer School, 2002
Page 43 new IT - though perhaps less visible - to any organization's ability to make a successful transition to Knowledge Economy. Organizations that have recognized this, and changed their management practices accordingly, have been relatively successful with their IT investments. In, those organizations that have not - the majority, unfortunately - the failure to recognise these changes has caused management practices to lag behind the evolving nature of the application of IT. While the application of Information Technology has evolved significantly over the past three decades, our approach to managing it has not. When the primary application was automation, management thinking was still rooted in the industrial age. The mind-set meant that you were ahead and built the system, plugged it in and made sure it was running, like a new machine on a manufacturing assembly line. Unfortunately, as the application of IT has moved beyond automation of work all the way to business transformation, our management approach has remained rooted in industrial-age thinking. Management thinking has failed to understand the implications of the evolving role of IT in business system beyond technology. Many still think in terms of payroll processing systems that can begin depositing money in employee accounts on day one. In fact, we are delivering customer information systems that will only produce results gradually on day 50, day 100 or day 365, after people are trained and motivated to use the new application when serving customers. 4.2 Management blind spots: Four critical dimensions of complexity Current management practice fails to adequately address the impact and resource implications of four critical dimensions of complexity. These blind spots in traditional management mind-sets are: linkage, reach, people and time.
Copyright © Genetic Computer School, 2002
Page 44 Linkage This refers to the linkage between the expected results of an IT investment and business strategy, and between the IT investment and investments required in other areas of the business in order to realise the benefit. Understanding and addressing linkage requires a clear appreciation of the ultimate benefits and of the full scope of the investment required to achieve the benefits. Reach Reach refers to the breadth of change required by an IT investment, meaning how much of an organization is impacted. It also refers to the depth of change - the degree of impact and of organizational change required to realize the benefit. Addressing each requires understanding what areas of the organization, other organizations and stakeholders will be affected, what the impact will be and how it will be managed. People A large number and diversity of people must be motivated and prepared to change. This critical factor in business transformation is often underestimated. We need to understand who these people are today, how they will have to change and what interventions will be required to effect the change. We need to ask how these interventions will be managed for people with different starting points, attitudes and motivations. Time In business transformation, time is always of essence, but realistic time frames are notoriously hard to estimate. We need to ask - and ask again and again - what the realistic length of time is for all the necessary changes to occur and for the full benefits to be realized. We must base these estimates on understanding the previous three dimensions.
Copyright © Genetic Computer School, 2002
Page 45 Evolving complexity of IT applications These dimensions of change have become increasingly complex as the applications of IT have advanced through the stages of automation of work, information management and business transformation. In the automation stage, the four dimensions were fairly straightforward and posed few problems. In the case of automated payroll systems, for example, there few linkages, organizational reach was limited and few people were affected. Time was required to deliver the benefits was short; or, at least, the time frames were easily predictable in advance. Finally, benefits were easy to measure. As we moved through the information stage, there were more linkages, not all of which were, obvious 4.3 Business transformation and the knowledge economy The potential risks and rewards associated with such cases of business transformation show, what is involved in engineering out transition to a knowledge economy. The opportunities include expanding geographic scope, expanding electronic commerce and creating virtual companies. We are, moving toward an economy that is on-line, interactive, instantaneous, inter-networked and knowledge based. It is an economy that will require new organizational forms and which will dramatically change the nature of organizations and work. The emergence of this new economy involves business transformation fundamental change in value chain management and the application of new technologies to support "networked organizations that share knowledge, insight and experience effectively. Some experts predict chief executives will become knowledge capitalists who manage the knowledge assets of their organizations. Knowledge will not just be limited to your organization; it will come from outside well. In addition to managing investments in IT-enabled change in your own business system, you will have to manage change in an extended business system which includes customers, supplier financial institutions, regulators and many other intermediaries, all of whom will themselves be in state of change.
Copyright © Genetic Computer School, 2002
Page 46 While the opportunities created by business transformation are awesome the risks can be daunting to investment decision-makers. Today's large-scale IT projects and organizational change programs will be viewed as relatively simple initiatives compared to the sophisticated business transformation ones that will be required in the knowledge economy. These will raise significant new issues of linkage, reach, people and time. To manage these dimensions of complexity successfully, business transformation initiatives can no longer be viewed as traditional projects. They will need to be treated almost like mid-size businesses within the business, as programs that are managed continuously and proactively over long periods of time.
Copyright © Genetic Computer School, 2002
Page 47
CHAPTER 5: MANAGEMENT AND DECISION-MAKING 5.1 What is Management? Introduction Every organization, regardless of size, type, or location needs managers. The characteristics of managers vary. Managers may come from any nationality or be of either gender. They're just as likely to be women, particularly in middle management and supervisory management positions, as they are men. For instance, the manager described in the chapter opening Manager's Dilemma, Cynthia Glenn, chief operating officer of Oxford Health Care, is a good example of a successful manager in action. She's involved with a major change that will radically transform her organization. As the COO, she epitomizes the planning, organizing, leading, and controlling activities that must be performed in order for the company to meet its goals. Management is the process of getting activities done efficiently with and through other people. The process includes the functions or primary activities performed by managers. Efficiency refers to the relationship between inputs and outputs and refers to efforts to minimize resource costs. Effectiveness refers to goal attainment. Managers seek to be both efficient and effective. An organization is a systematic arrangement of people to accomplish some specific purpose. Managers are important to an organization's success because they direct and coordinate activities so the organization can reach its goals. Management is important in our society today. Accordingly, there are two reasons for studying management. We interact with organizations every day of our lives. Every product we use and every action we take is provided or affected by organizations. These organizations require managers. Managers earn more than operators because their decisions have a significant effect on organization's performance and because of the inadequate supply of effective managers. Copyright © Genetic Computer School, 2002
Page 48 Managers come in all shapes and sizes. First of all they exist at various levels within the (hierarchic) structure of an organization. Secondly, they are found in a variety of functional areas, such as marketing, finance and manufacturing. Nevertheless, while there are obvious differences between the different levels and functional areas, all managers perform the same functions and play the same roles. A successful manager must possess many skills, of which communication and problem solving are the most basic. Managers transmit and receive information in both written and in oral form. Written communications include reports, letters, memos, books and magazines, electronic mail and electronic bulletin boards like the World Wide Web. Oral communication comes from telephone calls, formal and informal meetings, and voice mail. Problem solving is the set of activities that leads from the recognition of a problem to its solution, where a problem is a condition or event that is (potentially) harmful or alternatively, a condition or event that is (potentially) beneficial. While trying to find a solution to a problem, managers engage in a sub-process of decision making, where a decision is a particular selected course of action. In reaching a solution to a given problem, it is generally necessary to make decisions. Management and Leadership Management and leadership, they are often used to mean the same thing. But what many people don't realize is that they are quite different. What is management and what is leadership? How do they relate? Management is composed of five major functions; planning, organizing, staffing, directing and controlling. These five functions are used to define what is management and the role of a manager. These functions are easily defined and match the definition of management which is; an activity undertaken by those who are responsible for the
Copyright © Genetic Computer School, 2002
Page 49 success of organizations. Other key words that are used to define management are coordinating, resources, accomplishment, and desired results. Leadership is not very easy to define. One thing that is clear is that leaderships function is to influence. The process of leadership is to influence. Influence in leadership is defined as the ability to modify or change behavior of people. Influence has five bases; legitimacy, coercion, rewards, expertise and referent. All these bases help to exert influence. Management and leadership are the arts of influence over people. They both are influencing people. Leadership is dealing solely with influence over others. Leadership is dealing exclusively with making people do something they otherwise would not do unless influenced to do. Management is much the same but is not so aggressive as leadership is. Management is a way of making sure people stay on task when they may not be on task. The role of management is to guide an organization toward goals and management is the process of reaction to goals by working with or through people and other resources. Leadership is guiding people in a certain direction. They both sound alike, but there is a difference that is important. Management is a function where as leadership is a process which is not structure as well as management. In management, the idea is to accomplish a goal that is set forth and agreed upon. There isn't any real persuasion that goes into this because the goals are set and decided on and the only thing need is to just do the work. In leadership, influence must be exercised. Influence is a process of making people do something they otherwise would not do. In management, influence is not very great because the goals are established and agreed on. Influence is not a major part of management like it is in leadership. Management and leadership are not the same. Management is a way to accomplish goals that have been agreed upon. Leadership is influencing people. The two have many areas that seem the same but they are very different because of the level of influence involved.
Copyright © Genetic Computer School, 2002
Page 50 Management Functions Henri Fayol, a French industrialist from the early part of the 1900s, proposed that managers perform five management functions: POCCC (plan, organize, command, coordinate, control). These functions still provide the basis around which popular management textbooks are organized. However, the functions have been condensed to four. 1. Planning includes defining goals, establishing strategy, and developing plans to coordinate activities. Formulation of long and short term plans for goals. 2. Organizing is determining what tasks are to be done, who is to do them, how the tasks are to be grouped, who reports to whom, and where decisions are to be made. Creation of structures and frameworks of standards, procedures and policies. 3. Leading includes motivating subordinates, directing others, selecting the most effective communication channels, and resolving conflicts. 4. Controlling is monitoring activities to ensure that they are being accomplished as planned and correcting any significant deviations. Monitoring performance and the environment. Although the functional approach is clear and simple, critics have suggested that it does not provide an accurate description of what managers actually do. Management Roles In the late 1960s, Henry Mintzberg conducted a precise study of managers at work. His findings challenged several long-held beliefs about the manager's job. He concluded that managers perform ten different, but highly interrelated roles. Management roles refers to specific categories of managerial behavior. 1. Interpersonal roles included figurehead, leadership, and liaison activities. In Mintzberg's framework, there are three interpersonal roles that manager play. Figurehead
Copyright © Genetic Computer School, 2002
Page 51 The manager performs ceremonial and other duties, such as greeting visiting dignitaries and signing legal documents. Leader The manager maintains the unit by hiring and training the staff and providing motivation and encouragement Liaison The manager makes contact with persons outside the manager's own unit peers and others in the unit's environment - for the purpose of attending to business matters. In many organizations it is the practice for all mail to be addressed to or signed by the manager. 2. Informational roles included monitoring, disseminating, and spokesperson activities. In Mintzberg's framework, there are three informational roles that manager play. Monitor The manager constantly looks for information bearing on the performance of the unit, scanning both the internal activity of the unit as well as its environment. Disseminator The manager passes on valuable information along to others in the unit. Spokeperson The manager passes on valuable information along to those outside the unitsuperiors and persons in the environment. Common activities are issuing press releases, giving media interviews, reporting to board meetings and so on. 3. Decisional roles included those of entrepreneur, disturbance handler, resource allocator, and negotiator. In Mintzberg's framework, there are four decisional roles that managers play. Copyright © Genetic Computer School, 2002
Page 52 Entrepreneur The manager makes rather permanent improvements to the unit, such as changing the organizational structure. Disturbance handler The manager reacts to unanticipated events, such as a currency devaluation in an overseas market. Resource allocator The manager controls the purse strings of the unit, determining which subsidiary units get which resources. Negotiator The manager resolves disputes both within the unit and between the unit and its environment. Follow-up studies of Mintzberg's role categories in different types of organizations and at different managerial levels within organizations have generally supported the notion that managers perform similar roles. However, the more traditional functions have not been invalidated. In fact, the functional approach still represents the most useful way of classifying the manager's job. Management Skills Managers need certain skills to perform the varied duties and activities associated with being a manager. Robert L. Katz found through his research in the early 1970s that managers need three essential skills or competencies. 1. Technical skills are skills that include knowledge of and proficiency in a certain specialized field. Technical skills became less important as a manager moves into upper levels of management. However, top managers still need some proficiency in the organization's specialty.
Copyright © Genetic Computer School, 2002
Page 53 2. Human skills include the ability to work well with other people both individually and in a group. The importance of human skills remains consistent, regardless of level within the organization. 3. Conceptual skills include the ability to think and to conceptualize about abstract situations, to see the organization as a whole and the relationships among the various subunits, and to visualize how the organization fits into its broader environment. All managers perform essentially the same functions, but lower-level managers emphasize leading while upper-level managers spend more of their time planning, organizing, and controlling. For the most part, the manager's job is the same in both profit and not-for-profit organizations. Managers in small businesses tend to emphasize the spokesperson role and are generalists. Also, the formal structure and nature of a manager's job in a large organization is replaced by more informality in a small firm. When managers work in different countries, they often need to modify their practices. 5.2 Managers and their Information needs
Figure 5.1 Management Structure Management Structure The model above (see figure 5.1) is rather simplistic but clearly, in any substantial organization, there will be a top layer of senior management dealing with high level, strategic and long-term issues for the organization. Below them will be a middle management layer with responsibilities for performance
Copyright © Genetic Computer School, 2002
Page 54 monitoring and shorter time horizons. Finally, there will be an administrative level, ensuring that the day-to-day operational activities are effectively managed. Managers at all levels of the organization need information in order to ensure that they are achieving the desired objectives or targets for their area of the business. Those particular objectives will be different at the various levels and business units in the organization (although they should all be geared to achieving its corporate objectives) and, hence, their information requirements will also be different. Nevertheless, there will be a flow of information both up and down the organization structure as each level receives information for its purposes and adapts it for the next level. We will be looking in more detail at these information systems in a later session but for the moment the focus is on what information is required. It is common to classify managers in a hierarchic organization as operating at one of three basic levels (refer to figure 5.2) Figure 5.2 Hierarchic Organization of Managers
Copyright © Genetic Computer School, 2002
Page 55 1. The strategic planning level (Top management) Upper management is usually comprised of the CEO and Vice Presidents of a company. The decisions that they make have the broadest scope and cover the longest time frame. Upper management decisions are very unstructured. Upper Management uses a ‘Decision Support System’ to help make its decisions. 2. The management control level (Middle management) Decisions by middle management, usually the Directors, cover a broader range of time than operational, management but less broad than upper management. These decisions involve experience using historical data to plan and control operations and implement the policies of upper management. Middle management decisions are semi-structured. Middle management uses a 'Management Information System', MIS, to produce periodic reports. 3. The operational control level (Operating management) Departmental or operating management is responsible for maintaining business records and improving the flow of work on a daily basis. Decisions here cover a narrow time-frame and are structured, i.e. predictable, and made by following a welldefined set of procedures using the current status of activities. Operating Management produces Daily Reports using a 'Transaction Processing System'. The characteristics of these three levels of management are quite different as shown in the following table by Kanter, with top management being more involved with planning and concentrating on a longer time horizon. Top management focuses more on external information sources, and deals with complex, relatively unstructured activities involving few people, but having broader scope than lower levels of management.
Copyright © Genetic Computer School, 2002
Page 56
Characteristic
Top management Middle management Operating management
Focus on planning Heavy
Moderate
Minimum
Focus on control
Moderate
Heavy
Heavy
Time frame
One to five years
Up to a year
Day to day
Scope of activity Extremely broad
Entire functional area
Single subfunction or task
Nature of activity Relatively unstructured
Moderately structured Highly structured
Level of complexity
Less
Very complex, many variables
defined variables
Job measurement Difficult
Result of activity
Type of information utilised
Mental attributes
Number of people involved Departmental/ divisional interaction
Less difficult
Plans, policies and strategies
External
Creative, innovative
complex,
better
Straightforward Relatively easy
Implementation schedules,
performance End product
yardsticks Internal,
reasonable Internal historical, high
accuracy
level of accuracy
Responsible, persuasive,
Efficient, effective
administrative
Few
Moderate number
Many
Intra-division
Intra-department
Inter-department
Table 5.1 Table 5.1 shows Job content of management levels.
Copyright © Genetic Computer School, 2002
Page 57 Measuring Performance As stated previously, effective information managers are those who have a clear understanding of what information they require to tell them how they are performing against their objectives. That is not always as easy as it sounds. Managing a busy department, for example, is a 'messy' business, with problems coming from all directions and the manger constantly having to reassess priorities to deal with them. It is easy to lose sight of the real goals and then, as a consequence, having to take drastic short-term actions to regain course. Critical success factors (CSF's) are crucial areas of responsibility where 'things must go right' requiring 'constant and careful attention from management'. For example, responding to new business enquiries within two working days. 1. Identify the managers 'Critical Success Factors' (CSF's) in your organization. 2. Determine what information is required to show whether or not that CSF is being achieved. 3. Set up the systems necessary to deliver that information as and when required. This technique proved to be particularly effective in those organizations which had clearly expressed corporate and departmental objectives as these ought to tie in closely with CSF's. Additional benefits were also generated by applying the technique throughout the organization whereby it helped to focus on improving communications between levels of management. This process has been applied and can be seen in many large multi-national companies today, where subsidiaries operate with a high degree of autonomy, reporting regularly to the parent company or Board on their performance in a number of 'key' areas. ‘Hard’ and ‘Soft’ Information When we talk about information in the context of the business environment there is a tendency to think in terms of facts and figures, reports and statements,
Copyright © Genetic Computer School, 2002
Page 58 print-outs or whatever appears on our computer terminal. It is relatively easy to obtain this type of information, which may be generated as a by-product of formal systems and processes already in place. While these forms of information are important, there are other forms, less formal, which can also play an important role in the decision making process. Previous research has shown that the informal channels for collecting, what is termed, 'soft' information can be very important, particularly in more senior positions within the organization. Indeed, less regard is likely to be given to 'hard' factual information. This tends to reflect the way in which managers work in the real world, where informal soft information (over the telephone or in the canteen) is collected and stored until other items of information (either soft or hard) reinforce or add to that information, like pieces in a jig-saw. The human mind is extremely effective at this form of 'information management' far more so than computers, where attempts to mirror this methodology have been relatively unsuccessful. So, an important lesson for the manager is not to focus solely on hard information and associated systems but also to consider how to develop and nurture the informal contacts and networks, which operate in and between organizations. 5.3 Problem Solving and Decision Making Decision Making
Figure 5.3
Copyright © Genetic Computer School, 2002
Page 59 Management decision making can be organized according to the level of managerial responsibility (refer to figure 5.3). At the top sits the strategic management level. Their strategic decision making requires a wide variety of information sources and the flexibility to model and present ideas in support of their strategic mission. Information systems technology at the strategic level must help to clearly communicate ideas and concepts. Information usually needs to be summarized more highly than at lower levels of management support, and also tends to use more "external" information coming from outside sources. At the middle range, tactical matters require management to reach into corporate records and obtain external data to implement and support company strategy. This tactical management level of management decision making is responsible for allocating resources to reach objectives and therefore requires specific feedback concerning performance of operational units. Control of corporate resources, such as performance monitoring and budget planning are part of their job. At the lower management levels, daily operations are the chief concern. At this operational management level, transaction processing and process control data are the major concern, and information systems often highly automated and critical for managing competitively. Beneath these systems, in the "basement" under the management information system [MIS], the transaction processing and other automated production processes churn out masses of data that must be stored, processed, and converted into business intelligence that feeds into every level of management. Effective decision making requires an understanding of decision-making processes including behavioral and communications aspects, the ability to identify, obtain, store, and retrieve critical information necessary for making sound decisions and the skill to use mathematical, statistical and simulation tools in analyzing information in order to make decisions. Structured decisions have a known and well-defined solution that uses fully available data. The outcomes will always be the same. Examples of structured
Copyright © Genetic Computer School, 2002
Page 60 decisions are calculating net pay, calculating interest on a loan, etc. In nonstructured (or unstructured and semi-structured) decisions, there is no agreement upon procedure and not all the necessary data may be available. It requires judgement, evaluation and intuition. There are few unstructured business decisions, many are semi-structured or parts of the problem are structured and parts require judgement and intuition. Most business decisions are structured or semi-structured. Budget analysis is structured and can be done by a spreadsheet program. Budget preparation is semistructured. Computers can analyze, past data and projections. Humans must judge what they think is most likely to happen in future. Each level has different types of decisions, therefore, measures the value of information differently. General Model of Decision Making 1. Intelligence [Finding out what is going, what the significant problems are] 2. Design [Identifying the decision(s) that need to be made] 3. Choice [Making the decision] 4. Implementation Go through an example of this process (Choosing classes for the Fall) Problem Solving ♦
Structured Problems
• solved by known methods and models • all relevant data available • require little judgement, evaluation or insight • easily automated Copyright © Genetic Computer School, 2002
Page 61 ♦
Unstructured Problems ("Less structured" would be a better term)
• standard solution methods unknown • some desirable data may not be available • requires considerable human judgement, evaluation and creative insight • difficult to automate ♦
Semi-structured Problems
• Some parts are structured, others are not • Remember that there is a continuum from structured to unstructured. Two important points to remember: 1. Structuredness is not the complexity of the problem -- complex problems may be decomposed and approached in a highly structured manner. 2. All problems that can be approached in a rational manner are capable of some degree of structuring. MIS are often categorized according to the type of approach that they employ in helping resolve managerial problems related to these activities. Managerial problem solving range from the highly structured to the relatively unstructured approachs. Structured problems are sometimes called "programmable" decisions because they are easily formalized with models. Structured problems predominate in operations, where data processing on transaction data can be coupled with process control data to permit highly automated decision making. When such processes become very repetitive, they may be fully automated, leaving humans to simply monitor the decision and outcomes. On the other end of the range, unstructured problems require tools that permit managers to model the dimensions of their problems, honing in on the best solutions to difficult, and often unique, situations. However, no problem that confronts management
Copyright © Genetic Computer School, 2002
Page 62 is ever totally unstructured. If a problem can be identified, it must contain elements that allow it to be structured to some degree. The amount of structuring that is done depends upon the nature of the problem and the time and skill the manager has to devote to it. It also depends upon the manager's own decision making style. In the middle between the highly structured and the unstructured problems lie many semi-structured problems in which parts, but not all of the problem can be approached using formal decision making models. For example, the decision of AquaPenn to acquire additional manufacturing facilities on the West Coast was based partly upon a structured analysis of the logistics of supplying customers in that regoin of the country. Solid data and knowledge to use it was available concerning the logistics. However, parts of the problem, such as those involving the activities of their competitors or the future demand for bottled water, required a less structured analysis of the market for bottled spring water. Not all factors were known and not all data was "hard”. 5.4 Coping with Changes How managers can help employee deal with change Different employees have different needs. The most effective learned how to adapt their style based on the individual's needs. ♦
"Surviving" (Innocent) Perspective:
• Employees in this level are very fragile and should be protected from change as much as possible. Keep as much of their routine the same since any type of changes - even what seem to be small ones - will be terrifying to them and their reactions may be extreme. ♦
"Learning" (Rule-based) Perspective:
• Employees at this level need a "parent" figure to help them deal with the changes - someone who can help them feel safe and secure with the changes as well as help them adjust to the changes at a pace that they can handle. Otherwise, they will create all sorts of problems for management. Copyright © Genetic Computer School, 2002
Page 63 ♦
"Competing" (Striving) Perspective:
• People at this level need to maintain their sense of self-worth through the changes. If they feel threatened, they will fight back in whatever ways they can find. If they can be recruited to have a feeling of personal importance in the change, they will find it much easier to deal with. It is important to help them understand that fighting the changes are not in their best interest, and that working with others is the way they can benefit them the most. ♦
"Relating" (Partnership) Perspective:
• The challenge for people at this level is to feel that everyone's needs are being taken care of. They will feel the pain of their co-workers' struggles to cope with change, yet they may not know how to deal with their own feelings of insecurity. They most value being part of a team and will usually want to help others adjust to the changes. They may participate actively in the gossip about changes. People at this level can assist by being asked to "partner" with someone who is expected to have more difficulty in coping with changes. ♦
"Teaching (Integration) Perspective:
• People at this level will be able to understand the reasons behind the changes and can be very helpful as a calming presence for people at other levels. They can be most helpful to management if they are told about upcoming changes and have time to help prepare others. They will be most effective in this role when they can see the change(s) as a positive one and translate that message to those who are more fearful. They will usually take a more philosophical attitude toward the changes when they are told the larger picture and overall strategy behind the changes. As with people in the "relating" perspective, employees at this level can be very helpful if they are asked to help others cope with the changes through a "partner" or "buddy" arrangement, whether formal or informal.
Copyright © Genetic Computer School, 2002
Page 64 Note: "employee" can be any person who works in an organization - line, staff, management or executive Organizational Responses to Change ♦
Top Management:
Top management has a hard time coming to grips with the direct implications of the change. They often underestimate the impact that change has on their employees. They tend to isolate themselves. Often they engage in strategic planning sessions and gather information in survey reports. They avoid communicating or seeking "bad news," because it is difficult for them to admit "they don't know." They expect employees to "go along" when a change is announced and they blame their middle managers if people resist or complain about the change. They often feel betrayed when employees don't respond positively. ♦
Middle Management:
Managers in the middle feel the pressure to "make organization change" according to the wishes of top management. They feel pulled in different directions. Middle managers often lack information and leadership direction to focus on multiple priorities. They are caught in the middle, and often fragmented because they don't have clear instructions. They feel besieged with upset, resistant or withdrawn employees who no longer respond to previous management approaches, and deserted, blamed or misunderstood by their superiors. ♦
Employees/Workers/Associates:
Workers often feel attacked and betrayed by changes announced by management. They are often caught off guard, not really believing that "my company could do this to me." Many respond with resistance, anger, frustration and confusion. Their response can solidify into a wall of "retirement on the job." They become afraid to take risks, be innovative or try new things. They experience a loss of traditional relationships, familiar structure and predictable career advancement patterns.
Copyright © Genetic Computer School, 2002
Page 65
CHAPTER 6 - KNOWLEDGE REPRESENTATION 6.1 Introducton Artificial intelligence techniques do not necessarily reproduce human thought processes. Rather, they seek to apply computers to problem solving tasks requiring intelligence. These techniques need to produce results consistent with human activity if they are to be useful, however, they do not necessarily reproduce the reasoning of an expert in their analysis. Expert systems represent the logical extreme of the trend away from large generalized artificial intelligence models. By adopting a specific problem domain, the tasks of knowledge representation and problem description are rendered as simple as possible. This enables a more powerful model to be developed, since the knowledge base applicable to the chosen problem domain can be made much more comprehensive. A detailed and specific knowledge base is the source of the problem solving power of an expert system (Waterman [1986]). The manner in which the knowledge base is used will be determined by the system's meta knowledge. Meta knowledge can be seen as a set of rules governing how the knowledge base is applied. Meta knowledge determines what case specific knowledge is obtained from the user and what knowledge from the knowledge base is incorporated in the problem solving process. It also provides the means to combine these two sources of knowledge for an expert system to generate its output. Meta knowledge and knowledge complement one another. Knowledge cannot be applied in the absence of meta knowledge. However, reliance on meta knowledge may be offset to the degree of reliance on a knowledge base. An advanced expert systems shell facilitates the construction of expert systems based models that embody a high degree of declarative meta knowledge. Such a model solves problems while making "economical" use of the system's knowledge base. It limits its use to those pieces of knowledge pertinent to the problem.
Copyright © Genetic Computer School, 2002
Page 66 A model with a less advanced meta knowledge structure may produce the same conclusion, but it would be less discriminating in its use of the knowledge base. As a result, it may incorporate many pieces of knowledge that were not strictly necessary in its analysis. A procedurally coded computer model will necessarily place a heavier reliance on its knowledge base than it will on a meta knowledge structure. This is not necessarily a failing. Knowledge is typically easier to capture and represent than meta knowledge. Expert systems are commonly constructed using proprietary shell packages. These packages provide the tools to construct the knowledge base required by an expert systems model and the inference engine and meta knowledge required to access that knowledge base. The construction of an expert system requires the skills of a knowledge engineer, as opposed to a conventional programmer. In contrast, decision support systems can be constructed using only a conventional program code compiler. A compiler embodies less functionality than an expert system shell, but as a result, needs to be much less structured and can therefore allow the developer much more freedom over the form of a decision support system. The programming skills required to use a conventional program compiler are much more readily available and therefore potentially more cost effective than the skills of a knowledge engineer. The remainder of this paper is organized as follows. The second section discusses the distinction between expert systems and decision support systems. The third section reviews the application of expert systems to audit risk assessment. The fourth outlines the constructiort of two computer models, one expert systems based and the other procedural. The final section provides a summary and conclusion. A well-known definition of artificial intelligence (AI) is the following: AI is the science of making machines do things that would require intelligence if done by humans. An important branch of Al research is expert systems (ES) or more generally, knowledge based systems (KBS). In the 1980s, there were a number of highly-publicized, successful expert systems put into operation around the world (e.g., MYCIN, XCON, Prospector) Despite these successes and the many operational expert systems in service today,
Copyright © Genetic Computer School, 2002
Page 67 the field is widely seen as having arrived, failed, and disappeared. However, it is important to keep in mind that much this perceived failure is due to unrealistic expectations fueled by hype rather than a lack of achievement by ES researchers and developers. In fact, such a fate is not uncommon in information technology. The basic idea behind expert systems is to create small, practical "intelligent" systems by eliciting a knowledge base of rules from experts in a particular field and providing non-expert users with means of accessing this knowledge easily. For example, an early (circa 1975) expert system called MYCIN helps physicians to diagnosis infectious blood diseases and prescribe medication. The knowledge base of MYCIN was elicited from a large number of experts in the field and contained an enormous amount of information about different types of bacteria associated with blood diseases. Although MYCIN provided diagnosis as good or better than any human expert, the technology itself never achieved widespread acceptance in the medical community. 6.2 Knowledge Representation Knowledge representation is crucial. One of the clearest results of artificial intelligence research so far is that solving even apparently simple problems requires lots of knowledge. Really understanding a single sentence requires extensive knowledge both of language and of the context. For example, today's (4th Nov) headline "It's President Clinton" can only be interpreted reasonably if you know it's the day after the American elections. (Yes, these notes are a bit out of date). Really understanding a visual scene similarly requires knowledge of the kinds of objects in the scene. Solving problems in a particular domain generally requires knowledge of the objects in the domain and knowledge of how to reason in that domain - both these types of knowledge must be represented. Knowledge must be represented efficiently, and in a meaningful way. Efficiency is important, as it would be impossible (or at least impractical) to explicitly represent every fact that you might ever need. There are just so many potentially useful facts, most of which you would never even think of. You have to
Copyright © Genetic Computer School, 2002
Page 68 be able to infer new facts from your existing knowledge, as and when needed, and capture general abstractions, which represent general features of sets of objects in the world. Knowledge must be meaningfully represented so that we know how it relates back to the real world. A knowledge representation scheme provides a mapping from features of the world to a formal language. (The formal language will just capture certain aspects of the world, which we believe are important to our problem - we may of course miss out crucial aspects and so fail to really solve our problem, like ignoring friction in a mechanics problem). Anyway, when we manipulate that formal language using a computer we want to make sure that we still have meaningful expressions, which can be mapped back to the real world. 6.3 What is Artificial Intelligence? Artificial intelligence (AI) is a broad field, and means different things to different people. It is concerned with getting computers to do tasks that require human intelligence. However, having said that, there are many tasks,. which we might reasonably think require intelligence - such as complex arithmetic - which computers can do very easily. Conversely, there are many tasks that people do without even thinking - such as recognizing a face - which are extremely complex to automate. Al is concerned with these difficult tasks, which seem to require complex and sophisticated reasoning processes and knowledge. People might want to automate human intelligence for a number of different reasons. One reason is simply to understand human intelligence better. For example, we may be able to test and refine psychological and linguistic theories by writing programs, which attempt to simulate aspects of human behaviour. Another reason is simply so that we have smarter programs. We may not care if the programs accurately simulate human reasoning, but by studying human reasoning we may develop useful techniques for solving difficult problems.
Copyright © Genetic Computer School, 2002
Page 69 AI is a field that overlaps with computer science rather than being a strict subfield. Different areas of Al are more closely related to psychology, philosophy, logic, linguistics, and even neurophysiology. 6.4 Is AI Possible? Artificial intelligence research makes the assumption that human intelligence can be reduced to the (complex) manipulation of symbols, and that it does not matter what medium is used to manipulate these symbols - it does not have to be a biological brain! This assumption does not go unchallenged among philosophers etc. Some argue that true intelligence can never be achieved by a computer, but requires some human property which cannot be simulated. There are endless philosophical debates on this issue (some on comp.ai.philosophy), brought recently to public attention again in Penrose's book. The most well known contributions to the philosophical debate are Turing's "Turing test" paper, and Searle's "Chinese room". Very roughly, Turing considered how you would be able to conclude that a machine was really intelligent. He argued that the only reasonable way was to do a test. The test involves a human communicating with a human and with a computer in other rooms, using a computer for the communication. The first human can ask the other human/computer any questions they like, including very subjective questions like "What do you think of this Poem". If the computer answers so well that the first human can't tell which of the two others is human, then we say that the computer is intelligent. Searle argued that just behaving intelligently wasn't enough. He tried to demonstrate this by suggesting a thought experiment (the "Chinese room"). Imagine that you don't speak any Chinese, but that you have a huge rule book which allows you to look up chinese sentences and tells you how to reply to them in Chinese. You don't understand Chinese, but can behave in an apparently intelligent way. He claimed that computers, even if they appeared intelligent,
Copyright © Genetic Computer School, 2002
Page 70 wouldn't really be, as they'd be just using something like the rule book of the Chinese room. Many people go further than Searle, and claim that computers will never even be able to appear to be really intelligent (so will never pass the Turing test). There are therefore a number of positions that you might adopt: • Computers will never even appear to be really intelligent, though they might do a few useful tasks that conventionally require intelligence. • Computers may eventually appear to be intelligent, but in fact they will just be simulating intelligent behaviour, and not really be intelligent. • Computers will eventually be really intelligent. • Computers will not only be intelligent, they'll be conscious and have emotions. Computers can clearly behave intelligently in performing certain limited tasks, full intelligence is a very long way off and hard to imagine. However, these philosophical issues rarely impinge on Al practice and research. It is clear that AI techniques can be used to produce useful programs that conventionally require human intelligence, and that this work helps us understand the nature of our own intelligence. 6.5 Some AI Tasks Human intelligence involves both "mundane" and "expert" reasoning. By mundane reasoning I mean all those things which (nearly) all of us can routinely do (to various abilities) in order to act and interact in the world. This will include: • Vision: The ability to make sense of what we see. • Natural Language: The ability to communicate with others in English or another natural language. • Planning: The ability to decide on a good sequence of actions to achieve your goals. Copyright © Genetic Computer School, 2002
Page 71 • Robotics: The ability to move and act in the world, possibly responding to new perceptions. By expert reasoning I mean things that only some people are good at, and which require extensive training. It can be especially useful to automate these tasks, as there may be a shortage of human experts. Expert reasoning includes: • Medical diagnosis. • Equipment repair. • Computer configuration. • Financial planning. Expert Systems are concerned with the automation of these sorts of tasks. AI research is concerned with automating both these kinds of reasoning. It turns out, however, that it is the mundane tasks that are by far the hardest to automate. 6.6 Knowledge Engineering Having decided that your problem is suitable you need to extract the knowledge from the expert and represent it using your expert system shell. This is the job of the knowledge engineer, but involves close collaboration with the expert(s) and the end user(s). The knowledge engineer is the AI language and representation expert. He/she should be able to select a suitable expert system shell (and other tools) for the project, extract the knowledge from the expert, and implement the knowledge in a correct and efficient knowledge base. The knowledge engineer may initially have no knowledge of the application domain. To extract knowledge from the expert the knowledge engineer must first become at least somewhat familiar with the problem domain, maybe by reading introductory texts or talking to the expert. After this, more systematic interviewing of the expert begins. Typically experts are set a series of example problems, and will explain aloud their
Copyright © Genetic Computer School, 2002
Page 72 reasoning in solving the problem. The knowledge engineer will abstract general rules from these explanations, and check them with the expert. As in most applications, the system is wasted if the user is not happy with it, so development must involve close collaboration with potential users. As mentioned in the introduction, the basic development cycle should involve the rapid development of an initial prototype and iterative testing and modification of that prototype with both experts (to check the validity of the rules) and users (to check that they can provide the necessary information, are satisfied with the systems performance and explanations, and that it actually makes their life easier rather than harder!). In order to develop the initial prototype the knowledge engineer must make provisional decisions about appropriate knowledge representation and inference methods (e.g., rules, or rules+frames; forward chaining or backward chaining). To test these basic design decisions, the first prototype may only solve a small part of the overall problem. If the methods used seem to work well for that small part it's worth investing the effort in representing the rest of the knowledge in the same form. Expert system development was very trendy around 5-10 years ago, with unrealistic expectations about the potential benefits. Now some cynicism has set in. Expert system shells are in fairly wide use, but are often used to solve fairly simple problems, and are chosen as much for their user interface and development environments as for their inferential abilities.
Copyright © Genetic Computer School, 2002
Page 73
CHAPTER 7: PLANS AND STRATEGIES 7.1 Strategy Planning for Information Systems History Perhaps, because of the relative 'youth' of computing in the business environment and coupled with its associated pace of change, the track record of IS, as far as effective strategic planning is concerned, has been patchy. There are many welldocumented examples of poor strategic decisions and costly failures. Surveys indicate that, until comparatively recently, a high percentage of organizations failed to plan strategically in this area and those that did were not particularly effective. The strategic planning process demands regular review and reassessment. However, as stated in the text, major corporate changes i.e. acquisitions, alliances, etc. or radical changes in the market which the organization operates in, will hasten these. Consider the following questions: • Where does an IS strategy fit within the wider set of strategies? • What has been the history of IS strategy planning? • What circumstances demand major reassessment of IS strategy plans? • Who might be employed to do the actual planning? • What might an IS strategy plan contain? Firstly, in its supporting role, the IS strategy has to be clearly linked with the corporate and business goals of the organization. Its constituent parts will consider specific issues, often viewed as sub-strategies, in considerable detail. As an example, the software sub-strategy may include a detailed software replacement programme, identifying which systems will be replaced and when, together with the resources required at each stage. This sub-strategy will have been determined by reference to the overall business strategy, as well as determining factors at the IS strategy level Copyright © Genetic Computer School, 2002
Page 74 such as 'a decision to source future systems from off the self rather than developed inhouse'. This would also be reflected in the sub-strategy dealing with I.T. staffing. Key Players Much has been written about who the key players are in IS strategy formulation and a preliminary stakeholder analysis may give the clue as to who should be involved. The culture of the organization - centralized or devolved - will also act as a determining factor. IT directors have seen their fortunes wax and wane over the years as more and then less were afforded Board level status. The concept of the 'hybrid manager' has been adopted in many organizations. Components of an IS Strategy Undoubtedly, there is no one arrangement, which best suits all. Similarly, the content of an IS strategic plan may very well be a 'mix and match' most appropriate to the organization concerned - although elements of 'best practice' are well described in the text and include the following: • A clear statement of the IS objectives. • An inventory and assessment of current organizational capabilities. • An implementation plan. 7.2 IS and Bisiness Strategy- Alignment and Intergration Rationale There is little contrary argument to the proposition that working to a welldeveloped plan offers greater chance of achieving desired -objectives than working in an intuitive, reactionary mode. Not least, a structured plan provides a map to refer to when the actual position is different to that anticipated or desired. Information systems are costly to develop and manage, in common with the other resources the organization has to use. More importantly perhaps than with other assets, its life-cycle is more
Copyright © Genetic Computer School, 2002
Page 75 indeterminate and potentially has a greater impact on the well-being of the parent organization. Consider an overview of the strategic planning process as it relates to the organization and you will (will have) encountered many of the models elsewhere in the Programme. Determining the business strategy is fundamental to the subsequent development of an appropriate supporting IS strategy. As a consequence, before developing an organization-wide IS strategy, there must be a clear understanding of the external business environment together with an assessment of the organization's internal 'condition'. Having established the credence for organizations adopting an IS strategy, we need to consider which planning process is the most appropriate for a particular organization. The basic components of an IS strategy will be common to most organizations. However, the methodology or framework by which it is developed is not. Framework The framework or methodology adopted to develop the IS strategic plan will contain the following elements: • A structure that gives guidance on what to do and when to do it. • A definition of techniques to do what needs to be done. • Advice on how to manage the quality of the results. • Tools to automate the process. It has to be emphasized at this point that frameworks are not always mutually. However, a particular approach may lend itself to the style and culture of an organization.
Copyright © Genetic Computer School, 2002
Page 76 7.3 Analytical Tools for IS Strategic Planning Introduction There is a real danger in attempting to use too many tools and techniques in formulating an IS strategy and the predominant framework, as discussed in the previous session, ought to give a firm steer as to which ones may be most appropriate. Tools The value of each of the tools described is in their ability to clarify, to a greater or lesser extent, a complex situation. Often, they also force the planner to consider a range of issues from a different perspective. As a consequence, 'Beauty' - the value of a particular tool - is very much 'in the eye of the beholder'. For example, the layered structure and inter-connectedness of Rockart's Critical Success Factors is immediately attractive to someone coming from a highly stratified and bureaucratic organizational background but would be alien to someone from an informal and innovative organization such as 3M. 7.4 MIS Strategic Planning Strategic Planning and Information Technology Strategic planning is the first step in initiating organizational change, as well as an essential element of business process reengineering. The strategic planning process includes an organizational assessment and the development of strategic foundations: mission, future vision, and guiding principles. Analyzing the gap between the current state of the organization and its future vision provides essential information for developing strategic goals, specific strategies, and objectives. Each strategy may have several objectives. It is at the level of objectives that it is reasonable to introduce performance measures to gauge progress toward achieving the objectives and, therefore, the strategies and goals.
Copyright © Genetic Computer School, 2002
Page 77 Projects are typically the links between plans and the budgeting process. As they are used in this document, projects are undertakings directed at the accomplishment of an objective. Projects, with cost and schedule estimates, may implement new information systems or major improvements to existing systems. The completion of projects should mark progress toward reducing the gap between an organization's current state and its future vision. Strategic Planning Process The literature on the strategic planning process is extensive; the references provide some recommended starting points. The ITSC has found two documents to be especially helpful. Wells, [1995] have evolved a structured strategic planning process that is appropriate for public sector organizations. While many variations of this process are possible, it does provide a basic framework. An overview of the planning process, as described by Wells, is presented in figure 7.1. The process begins with pre-planning activities designed to prepare senior management for strategic planning. These activities focus on gathering data about the organization and assessing the current state of the organization. In a workshop setting, senior leaders develop, through a consensus building process, the mission, the future vision, and the guiding principles of the organization. These elements represent the strategic foundation, but the elements by themselves are insufficient as a plan. Strategic goals, strategies, and objectives are necessary to guide deployment and implementation. To develop the goals, strategies, and objectives, managers work through a gap analysis to identify what needs to be changed to move the organization from its current state to the new desired state. This gap analysis further illustrates the relationships between strategic planning and typical business process reengineering (BPR) activities, during which the "as is" and "to be" states are described.
Copyright © Genetic Computer School, 2002
Page 78
Adapted From Denise Wells, et al., Strategic Management: Using the Strategic Plan to Build Organizational Budgets, Workshop on Performance Measuring and Monitoring in Government San Francisco, 1995. Figure 7.1 Figure 7.1 shows that developing goals, strategies, and objectives marks the completion of the strategic plan. Understandably, the calendar time and effort to develop the plan will vary among organizations. In the Wells approach, the strategic foundations -- the vision, mission, and guiding principles -- are developed in a 3-day workshop with trained facilitators. After the workshop, top management usually seeks the assistance of other members of the organization to complete the goals, strategies, and objectives, based on the strategic foundations. After completion of the plan, it is published and distributed throughout the organization. Implementation of the strategic plan occurs when action plans (i.e., projects, with cost and schedule estimates) are developed to accomplish the objectives. The projects are developed and implemented by personnel throughout the organization.
Copyright © Genetic Computer School, 2002
Page 79 Implementation efforts are monitored and measured so that the organization can evaluate progress toward achievement of its goals, strategies, and objectives. Information from this evaluation is used as input to the next iteration (see Figure 7.1) of the strategic planning process where the strategic plan is validated and updated based on changed conditions. An important by-product of this process is team-building and the breaking down of functional barriers. Team building is a natural by-product because senior managers are discussing the future of the organization and how they can cooperate towards achievement of that future. The managers have agreed to work together for the overall aim of the organization rather than their departments. This leads to organizational alignment and the breaking down of barriers among departments. Strategic Goals The strategic goals are broad statements about where the organization wants to be at some point in the future. These goals work towards achieving the overall mission of the organization and help achieve the vision of the future. The strategic plan usually contains goals in multiple strategic areas. The strategic goals are sometimes confused with the future vision. Whereas the future vision is a single statement or paragraph that describes a desired organizational end-state at some point in the future, the strategic goals are general statements of activities that if pursued would lead to the attainment of the future vision. In this context the future vision is a statement about where to go, and the strategic goals are statements about how to get there. Progress in achieving these goals should be measurable, hence enabling quantitative performance measurement. Objectives and Performance Measures Objectives represent specific courses of action that are bounded by and support the strategies. They contain a target and a performance measure. In the absence of targets and performance measures, it is impossible to measure progress towards achieving the strategic goals. Copyright © Genetic Computer School, 2002
Page 80 Objectives follow from the "top-down" approach of IM strategic planning. Goals describe the end state to be achieved; strategies outline how to get there; and objectives describe who does what and when. Once the outcomes and the methods for achieving those outcomes have been established, indicators to measure performance must be developed. Implementation, however, is a "bottom-up" activity; it is around the achievement of objectives that data are, collected and performance is measured. Measuring progress in achieving strategic goals is never easy; however, for government to be accountable, performance measurement is important. Performance measures are quantifiable expressions that measure the achievement of program objectives. Multiple indicators are often needed to capture the intent of a program. The measures must reflect the desired result intended, and they must possess the qualities of validity and sufficiency. A valid performance measure is one that reasonably represents the program objective. A valid performance measure should not measure workload data, and it should minimize the measurement of change not attributable to the program. Sufficiency implies that the number of indicators adequately reflects the intent of the program objective. A sufficient measure provides a comprehensive measure of effectiveness. Additional indicators are needed if they provide new information that is pertinent to the assessment of program effectiveness. 7.5 Strategic Information Systems Introduction The topic of 'strategic information systems' is concerned with systems which contribute significantly to the achievement of an organisation's overall objectives. The body of knowledge is of recent origin and highly dynamic, and the area has an aura of excitement about it. It is risky to attempt a historical exposition of such a recently emerged topic. On the other hand, the line of development which the conventional wisdom has followed is itself interesting and instructive. This paper is prepared as an introduction to the literature, but embodies interpretation in both its structure and its
Copyright © Genetic Computer School, 2002
Page 81 expression, and should therefore be read with at least as critical a disposition as any other paper in the area. It may also be compared with other critical interpretations such as Swatman & Swatman (1992), Galliers (1993) and Ciborra (1994). The notion and its origins are first discussed. The emergence of the key ideas is then traced. The process whereby strategic information systems come into being is assessed. Finally, areas of weakness are identified, and directions of current and future development suggested. Origins The role of Information Systems (IS) has developed during the years. The original conception was of the automation of existing manual and pre-computer mechanical processes. This was quickly succeeded by the rationalisation and integration of systems. In both of these forms, IS was regarded primarily as an operational support tool, and secondarily as a service to management. During the 1980s, an additional potential was discovered. It was found that, in some cases, information technology (IT) had been critical to the implementation of an organisation's strategy. The dominant sense in which the term is used is that a strategic information system (SIS) is an information system which supports an organisation in fulfilling its business goals. An alternative interpretation of the term is that it is not necessarily a particular IS, but rather the combination of those parts of an organization's cluster of information systems which provide information into its strategic planning processes (Higgins & Vincze 1993.p.93). The functions involved include the gathering, maintenance and analysis of data concerning internal resources, and intelligence about competitors, suppliers, customers, government and other relevant organisations. A variety of interpretations of strategy exist, most of which have a great deal to do with competition between corporations. Chamberlin's theory of monopolistic competition sees corporations as being heterogeneous, and competing on the basis of asset differences, such as technical knowledge, reputation, ability for teamwork, Copyright © Genetic Computer School, 2002
Page 82 organizational culture and skills, and other 'invisible assets' (Chamberlin 1933, Itami 1987). Competition therefore means cultivating unique strengths and capabilities, and defending them against imitation by other firms. Another alternative sees competition as a process linked to innovation in product, market, or technology (Schumpeter 1950). Porter's Strategic Theory The context within which SIS theory emerged was the competitive strategy framework put forward by Porter (1980, 1985), which was based on industrial organisation economics. For developments along that path, see Kaufmann 1966, Kantrow 1980, Pyburn 1981, Parsons 1983, EDP Analyzer 1984a, 1984b, McFarlan 1984, Benjamin et al 1984, Wiseman & Macmillan 1984, Ives & Learmonth 1984, Cash & Konsynski 1985, Porter & Millar 1985, Keen 1986, King 1986). This first section outlines the basis of that theory. It will then be shown how Strategic information systems theory is concerned with the use of information technology to support or sharpen an enterprise's competitive strategy. Competitive strategy is an enterprise's plan for achieving sustainable competitive advantage over, or reducing the edge of, its adversaries. In Porter's view, the performance of individual corporations is determined by the extent to which they cope with, and manipulate, the five key 'forces' which make up the industry structure: • the bargaining power of suppliers; • the bargaining power of buyer; • the threat of new entrants; • the threat of substitute products; and •
rivalry among existing firms.
Porter's classic diagram representing these forces is reproduced in Figure 7.1. Enterprises, through their strategies, can influence the five forces and the industry structure, at least to some extent.
Copyright © Genetic Computer School, 2002
Page 83 There are two basic strategic stances that enterprises can adopt: • low cost; and • product differentiation. In the long run, firms succeed relative to their competitors if they possess sustainable competitive advantage in either of these two, subject to reaching some threshhold of adequacy in the other. Another important consideration in positioning is 'competitive scope', or the breadth of the enterprise's target markets within its industry, i.e. the range of product varieties it offers, the distribution channels it employs, the types of buyers it serves, the geographic areas in which it sells, and the array of related industries in which it competes. Under Porter's framework, enterprises have four generic strategies available to them whereby they can attain above-average performance. They are: • cost leadership; • differentiation; • cost focus; and • focused differentiation. Porter's representation of them is reproduced in Figure 7.2. Figure 7.2
Copyright © Genetic Computer School, 2002
Page 84 According to Porter, competitive advantage grows out of the way an enterprise organises and performs discrete activities. The operations of any enterprise can be divided into a series of activities such as salespeople making sales calls, service technicians performing repairs, scientists in the laboratory designing products or processes, and treasurers raising capital. By performing these activities, enterprises create value for their customers. The ultimate value an enterprise creates is measured by the amount customers are willing to pay for its product or services. A firm is profitable if this value exceeds the collective cost of performing all of the required activities. To gain competitive advantage over its rivals, a firm must either provide comparable value to the customer, but perform activities more efficiently than its competitors (lower cost), or perform activities in a unique way that creates greater buyer value and commands a premium price (differentiation). (Refer to figure 7.3)
Figure 7.3 Many differentiation bases exist, classified into four major groups (Border 1964, quoted in Wiseman 1988): • product (quality, features, options, style, brand name, packaging, sizes, services, returns); • price (list, discounts, allowances, payment period, credit terms); • place (channels, coverage, locations, inventory, transport); and
Copyright © Genetic Computer School, 2002
Page 85 • promotion (advertising, personal selling, sales promotion, publicity). IT can be used to support or sharpen the firm's product through these various attributes. Of especial importance is 'product differentiation'. This is the degree to which buyers perceive products from alternative suppliers to be different, or as it is expressed by economic theory, the degree to which buyers perceive imperfections in product substitutability. The buyers of differentiated products may have to pay a price when satisfying their preference for something special, in return for greater added-value. The connection between the producer and buyers may be reinforced, at least to the level of customer loyalty, and perhaps to the point of establishing a partnership between them. Such a relationship imposes 'switching costs' on the buyer, because its internal processes become adapted to the beneficial peculiarities of the particular factor of production, and use of an alternative would force internal changes. Hence product differentiation also serves as an entry barrier. In addition, a continuous process of product differentiation may produce an additional cost advantage over competitors and potential entrants, through intellectual property protections, such as patents, and the cost of imitation. The activities performed by a particular enterprise can be analysed into primary activities, which directly add value to the enterprise's factors of production, which are together referred to as the 'value chain', and supporting activities. Figure 7.4 reproduces Porter's diagram. Figure 7.4
Copyright © Genetic Computer School, 2002
Page 86
The primary, value-adding activities include those involved in the production, marketing delivery, and servicing of the product. They are linked, generally in a chain. Support activities include those providing purchased inputs, technology, human resources, or overall infrastructure functions to support the primary activities. By co-ordinating linked activities, an enterprise should be able to reduce transaction costs, gather better information for control purposes, and substitute less costly operations in one activity for more costly ones elsewhere. Co-ordinating linked activities is also an important way to reduce the combined time required to perform them. Hence co-ordination is increasingly important to competitive advantage. Gaining competitive advantage requires that an enterprise's value chain be managed as a system rather than as a collection of separate parts. Reconfiguring the value chain, by re-locating, re-ordering, re-grouping, or even eliminating activities is often at the root of a major improvement in competitive position. An enterprise's value chain for competing in a particular industry is embedded in a larger stream of activities that Porter terms its 'value system', but which might be more usefully referred to as the 'industry value-chain'. This includes suppliers and
distribution
channels.
Figure
7.5
reproduces
Porter's
representation.
Competitive advantage is a function of how well a company can manage the entire industry value-chain. A corporation can create competitive advantage by co-ordinating its links in that chain. An enterprise's activities are subject to influence from: Figure 7.5 Porter’s Industry Value-Chain
Copyright © Genetic Computer School, 2002
Page 87 • New technologies. These may alter the path of the value chain, e.g. the invention of semiconductors forced many vacuum-tube producers out of business, and the printing and publishing industries are currently confronted by a major upheaval; • New or shifting buyer needs. Customers are demanding the convenience and consistency offered by fast-food chains. This in turn influences related market segments; • Change in industry segmentation. The disappearance of old intermediaries and the emergence o new ones create the potential to substantially reconfigure the value chain. Enterprises that fail t adjust will be forced out; • Shifts in the costs or availability of factors of production. Competitive advantage can be gained b optimizing based on current conditions. On the other hand, enterprises saddled with assets and approaches tailored to outdated modes of operation suffer; • Change
in
government
regulations.
Changes
in
product
standards,
environmental controls, restrictions on entry to the market, and trade barriers all affect an enterprise's performance. 7.6 Three key success factors common to effective business First is integrating the network into the organization's business strategy thus ensuring the flexibility that today's businesses require. New business opportunities require new application that need network support. An integrated planning process ensures that the network can meet new demands and support new directions. The second factor is the homogeneity of services the network provides. An enterprise network facilitates information exchange among all authorized subscribers. Homogeneity refers to the ease and commonality of use wherever a customer connects to the network, which is especially important to subscribers who travel, use devices at other locations, or connect via remote services
Copyright © Genetic Computer School, 2002
Page 88 Finally, an effective enterprise network is a strategic, resource used for competitive advantage. Telecommunication industry publications feature numerous examples of organizations using networks to gain competitive advantage. Competition in the last half of the 1990's and beyond requires that all business investment provide benefits greater than their cost, networking is no different. Planning any major business investment is challenging; however, planning the enterprise network is especially complex. Network planners often, face several constraints that limit their progress. For example, most organizations large enough to consider data networks have already committed to a computing strategy, which may confine network planning options. Many have already implemented data networks, and the financial inertia of existing systems limits planning mobility. Additionally, new computing architectures, machine platforms, and business reengineering make network planning little like building an ocean liner in mid-voyage. Network planners must consider the constraints as they develop their plans and evolve towards an effective enterprise network. Today's businesses put much greater demands on the network than those of just a few years These increasing demands are coupled with unprecedented technological advances, service offerings, and vendor selection options. Uniting the business and IT plans with an appropriate enterprise network is becoming increasingly difficult.
Copyright © Genetic Computer School, 2002
Page 89
CHAPTER 8: WOrKGROUP INFORMATION SYSTEMS 8.1 Introduction A major class of new information systems is emerging. These systems are group productivity systems designed to enable workers to collaborate on a series of information-based tasks. Workgroup computing offers the potential to dramatically improve the effectiveness and productivity of organizations by radically improving the coordination of complex tasks which require cross-functional teams to work together. These systems must balance the need to structure and coordinate work with the seemingly conflicting need to support flexible communication between knowledge workers. Workflow systems are the flagship information system within the class of group productivity software. Business processes are dynamic in nature. As the market changes, as the organization changes, as regulations change, the business process must change in response. Workflow systems are the essential tools which an organization must have to respond to in this changing environment. Workflow systems must have several attributes to enhance the effectiveness of workgroups: • Rapid Application Generation: Rapid application development strategies enable organizations to quickly define a new workflow, to simulate its operation, to visualize the new workflow graphically, and to generate new workflow process definitions through a series of iterations. • Implication Independence & Extensibility: Workflow products must support comprehensive line-of-business applications from such wideranging areas as insurance and bank loan processing to engineering change management and process safety management. To achieve this breadth of support, the workflow product must be completely independent of
Copyright © Genetic Computer School, 2002
Page 90 application requirements for data definition, process routing, and applications interfaces. • Robust Workflow Engine: The essence of a workflow product is the underlying, which supports process definition, scheduling, and routing. To completely support application requirements the workflow engine must support several key functions such as job definition, security, parallel and ad hoc routing, drill-down workflows, event monitoring, and simulation. • Modular Product Design: A modular product architecture with welldefined interfaces between subcomponents ensures that compliance with emerging workflow standards can be achieved at reasonable cost. • Environmental Independence: The underlying technology environment in which workflow products must operate continues to change rapidly. Welldesigned workflow products must operate independently of key technology infrastructures such as database, server, and network. Workflow systems operate not in isolation but rather within an information systems infrastructure. With the Internet and especially internal corporate Intranets increasing in popularity, the emerging battle between traditional Client/Server configurations and the Internet/Intranet is likely to last for a few years with no conclusive winners. Each technology infrastructure has its own significant advantages. Workflow products which work across these major groupware environments will become essential elements of the emerging information infrastructure required to. manage the global enterprise. Workflow products which work in both traditional client-server environments as well as with the Internet/Intranet will provide the best of both worlds, taking full advantage of each world while offering the buyer the opportunity to use a common workflow product across the corporation. The basic idea behind groupware is said to be moving information, not people by using software supported "intentional group processes" aimed at what one commentator defines as "proactive analysis, compression and automation of information based tasks
Copyright © Genetic Computer School, 2002
Page 91 and activities". Group Decision Support Systems (GDSS) are defined as interactive computer systems supporting a group's formulation and solution of "unstructured problems". Workflow is defined as a "process driven" way of managing a series of tasks defined by procedures related to the flow of documents through organizations. It is noted that one of the biggest problems in the workflow world is the lack of a "succinct definition", although a March 1996 report identifies four types of workflow: Production, Collaborative, Administrative and Ad hoc. At the beginning of 1994, Groupware is seen as being challenged by multifunctional bundles of integrated applications software. A distinction is made between workflow and groupware on the basis that the former are designed for "simple and repetitive activities", while the latter tried to address more complex and variable organisational processes, although comparisons in the field are felt to be problematic. Notes is portrayed as being good at dealing with "free form and semi-structured" information and is praised especially for its "extended cross platform support" and flexibility, but is said not to be "robust enough" for transaction processing or activity management. News of Notes add-on products, especially products enabling seamless connections between Notes and other applications is regularly reported, while Notes upgrades feature automatic document versioning, full text search, "security features" and an object store for version 3.x, and promises of "sleeker" user interface, outside company connections, agent technology and an object oriented scripting language for version 4.0. Collabra described as a "discussion database" and "an excellent, inexpensive approach to groupware" is reported being integrated with Novell's GroupWise suite. IBM and Oracle are said to be releasing workgroup applications based on enhancements to DB2 and Oracle relational databases, while BeyondMail and ClipBook are portrayed as extending messaging functions to encompass groupware applications. As workflow applications are characterised as becoming "increasingly sophisticated", GUI and OO-based, object-orientation is said to be enabling the creation of new generation of more easily customisable groupware. At the end of 1994, reports begin to appear on multimedia groupware --initially with voice, then video.
Copyright © Genetic Computer School, 2002
Page 92 8.2 The Evolution of Workgroup Computing Information systems initially evolved for the purpose of processing transactions. Transaction processing systems emphasized the efficient and reliable recording of business events such as customer order receipt and invoicing, inventory control, etc. Mainframe-based computer systems grew from this need, and even today they manage the majority of business transactions. The dramatic improvements in processing power and the emergence of the personal computer gave rise to systems that expand personal productivity. Significant software innovation began to be focused not just on transaction management, but also on task automation for individual workers. The use of desktop publishing packages by graphics designers and computer-aided design systems by product designers are merely two examples of personal productivity systems. Today, a third major class of information systems is emerging. These systems increase group productivity. They are designed to enable workers to collaborate on a series of information-based tasks. Workgroup computing offers the potential to dramatically improve the effectiveness and productivity of organizations by radically improving the coordination of complex tasks, which require cross-functional teams to work together. Such tasks as product design, customer support, and regulatory compliance require systems to enable cross-functional teams to communicate, collaborate, and coordinate their work. These systems must balance the need to structure and coordinate work with the seemingly conflicting need to support flexible communication between knowledge workers. A convergence of technologies is shining the spotlight on workgroup computing. Technology advances such as reliable networking, relational databases, powerful workstations, and internal corporate Intranets have provided a backbone to make enterprise-wide workgroup computing possible. At the same time, dramatic marketplace changes are flattening organizations. These changes are forcing rapid product development cycles, and accelerating the demand for global information networks. As the need for communication and coordination across organizational boundaries accelerates, so does the requirement for developing workgroup computing
Copyright © Genetic Computer School, 2002
Page 93 systems respond to these needs in dramatic fashion. Indeed, workgroup computing offers the potential to finally deliver the organizational productivity gains so long promised by the information age. 8.3 The Evolution of Workflow Market Early workgroup computing systems began to appear in the mid-1980s. The focus of these systems, not surprisingly, was on the automation of mission-critical business processes, which are repetitive in nature. These systems became known as "workflow systems", with FileNet Corporation becoming the clear market leader in automating repetitive business processes. In reality, these systems are simply transaction processing systems that automate paper processing in addition to data processing. Early applications have been bank loan processing, insurance claims processing, and records management archiving. These systems streamlined the process of managing paper-based transactions. However, they were not without their limitations: 1. Transaction-based workflow systems are focused on streamlining a single business transaction, which flows through an organization in a sequential and predictable order. They place great focus on automating routine tasks at the user workstation, but are not well suited to business processes which are less orderly. As an example, the processing of a loan application in a bank is reasonably predictable. On the other hand, the processing of an engineering change order in a manufacturing firm is much less predictable and may in fact need to be routed to different groups based on the type of product, the nature of the change, or the financial magnitude of the project. Thus, transaction-based workflow systems optimize the automation of repetitive tasks and "paper mills" characterized by clerical workers. They are not well-suited for the automation of collaborative businesses processes driven by knowledge workers. 2. Most transaction-based workflow systems provide little or no document management capability. This is of little consequence when the basic unit of work is a piece of paper being routed through an organization; however, most collaborative
Copyright © Genetic Computer School, 2002
Page 94 business processes are document-intensive rather than transaction-intensive. In reviewing engineering contract changes or large insurance policy changes, there is a constant need to check documents into and out of archives in a controlled manner. Thus, the workflow process must be fully integrated with a system which provides document management and revision control capabilities. This can be provided by integrating the workflow system with a document management system. However, ideally this should be accomplished through one system providing both document management and workflow, so that workers performing collaborative tasks can share one user interface in an easy-to-administer system- a system where the user organization does not have to maintain coordination between two systems when each one comes out with a new release. 3. Transaction-based workflow systems require significant customization, often through high-level programming (4GL or "C" type), to "add the application to the work flow". This effort was certainly productive when automating mission-critical, repetitive business processes, which rarely change from year to year. However, for less predictable, more collaborative business processes the process of system definition and "business process reengineering" must be more iterative. This results in an absolute requirement for a new class of "collaborative workflow" system, which supports the rapid prototyping of new business processes with little, or better yet, no programming. Of course, such rapid application development facilities would also be beneficial in repetitive application building as well. 4. Finally, transaction-based systems typically use server-based architectures, which maintain much of the workflow logic on the system server. These systems have proven difficult to migrate into the latest advances in client-server technology, and therefore have often lagged other systems in their support of advances in networking environments,
client
workstation
technologies,
and
application
development
environments. Newer generation workflow systems have placed much of the logic in the client environment, thus enabling the system to take advantage of the increasing processing power of the client workstation while using the SQL and ODBC standards to support a wide variety of relational databases.
Copyright © Genetic Computer School, 2002
Page 95 As the workflow market has evolved, systems have tended to specialize in one of three types of applications: • Transaction-based workflow systems for automation of repetitive, paper-based business transactions. • Collaborative workflow systems for the automation of the more fluid, missioncritical, knowledge-based business processes. • Ad hoc workflow systems which are enhanced e-mail systems to route documents "on the I fly", but typically lack the ability to structure, coordinate, and manage the business process. A critical weakness of Ad hoc systems is that they are completely user-driven rather than being driven by the organization as a whole. With a repetitive or collaborative workflow model, the corporation can establish a new process definition, modify the process as new knowledge is generated, and capture the benefits of process re-engineering. With an ad hoc model, all process knowledge is "user-dependent" and leaves the organization if the user departs. This market structure is depicted in Figure 8.1 below: Workflow Market Structure (Exhibit 1) Figure 8.1
Copyright © Genetic Computer School, 2002
Page 96 8.4 The Essential Attributes of Workflow as a Strategic Technology To enhance the effectiveness of workgroups, workflow systems must include several attributes which are often not apparent from a superficial analysis of product functions and features. These; attributes are crucial to ensuring that the workflow product can support the underlying business process which it is automating over the life cycle of the business process. 8.5 Rapid Application Generation Business processes are dynamic in nature. As the market changes, as the organization changes, as regulations change, the business process must change in response. The workflow system must be a tool to support the underlying results, which the organization is attempting to achieve, rather than the other way around. To support business processes, which are subject to rapid changes, workflow systems must support rapid application generation. The workflow product must provide an easy to use, visual, point and click interface for configuring screens and building workflow scripts. Screen building, script building, and workflow should be integrated into a single user interface. The workflow product must be application independent, that is, it must not impose a particular data or routing structure. The rules, which define the organization structure, document structure, and process routing must be completely customizable by authorized users. Furthermore, the user must be able to change any of these workflow definition structures through an intuitive interface, which does not require programmers to execute. The workflow product must support application frameworks, which provide building blocks that enable the user organization to quickly generate new applications as well as modifications to previously defined applications. The workflow product should be extensible from market-leading application development environments such as Visual Basic. Developers operating in these
Copyright © Genetic Computer School, 2002
Page 97 environments should be able to define workflow objects with all of the key capabilities of the underlying workflow product. Rapid application development strategies enable organizations to quickly define a new workflow, to simulate its operation, to visualize the new workflow graphically, and to generate new workflow process definitions through a series of iterations. 8.6 Application Independence & Extensibility Workflow products must support comprehensive line-of-business applications from such wide-ranging areas as insurance and bank loan processing to engineering change management and process safety management. Figure 8.2 depicts common lineof-business applications, which are greatly enhanced by collaborative workflow:
Common Line-of-Business Applications (Exhibit 2)
Engineering Change Management Commercial Loan Processing Legal Contract Management Process Safety Management Regulatory Product Approval Insurance Policy Underwriting ISO 9000 Compliance Construction Project Management Collaborative Workflow Systems Figure 8.2 To achieve this breadth of support, the workflow product must be completely independent of application requirements for data definition, process routing, and applications interfaces. The objective of the workflow product is to provide an underlying framework for screen building, document routing, process definition,
Copyright © Genetic Computer School, 2002
Page 98 and process measurement without imposing requirements on the application. The workflow product must contain a full set of Application Program Interfaces (APIs) to enable third-party applications to easily integrate with all of the functions of the workflow product. OLE automation is a key technology for providing integration with a variety of Windows based document processors. An innovative approach to APIs is to use OLE to enable the workflow vendor's product to work within OLE enabled applications. Most leading development environments, such as Visual Basic, allow developers to make calls directly from their scripts to any other application that provides an OLE API. 8.7 Robust Workflow Engine The essence of a workflow product is the underlying engine which supports process definition, scheduling, and routing. To completely support application requirements the workflow engine must support several key functions: • Job Definition: The ability to define the workflow job as a series of tasks and link each task to a set of task completion rules. • Security: Each workflow task must be linked to the organizational groups and individuals who can view, modify, or complete the task. The product should provide multiple user-definable levels of security at the database record level. The security rules should also be definable at the workflow step level so that security rules can be modified as the workflow proceeds. • Process Modification: Authorized users must be able to change workflow jobs in process without altering the baseline job definition. • Parallel routing: Collaborative workflow can be streamlined by routing the workflow task to multiple users at once, - taking advantage of the existence of electronic rather than paper documents. To achieve this the workflow engine
Copyright © Genetic Computer School, 2002
Page 99 must support parallel document routing and the ability to re-synchronize the workflow process based on managing the document approval cycle. • Ad hoc routing: In addition to parallel routing, authorized users must also be able to define flexible routing paths, as many workflow routings are "conditional" on business events and user decision-making. • Approval rules: Complex business processes often require the workflow engine to support sophisticated approval rules including approval delegation, limits `to approval authority, and various approval voting formulas. • Drill-down workflows: The organization must be able to define "workflows within a workflow" so that the organization can take full advantage of the process definition work already completed when establishing a new workflow. • Event Monitoring: The engine must monitor significant events such as when a user does not approve or reject a document within a specified time frame, and generate appropriate system actions based on such events. • Status Reporting: The workflow system must provide detailed status reporting on all jobs in process. Graphical displays should be provided to allow the user to quickly comprehend the status of the workflow job. • Audit logging: Audit logs and reports should be provided to allow managers and regulators to review historical actions and provide proof of compliance with mandated processes. • Simulation: There should be a method of simulating the newly defined workflow process, using event-based probabilities, prior to actually putting the new process into operation. In providing these functions the workflow engine is supporting both the need to structure the workflow process as well as to enable authorized users to generate ad hoc events within the workflow process. If developed properly, the workflow product will
Copyright © Genetic Computer School, 2002
Page100 support completely structured workflow processes, completely ad hoc functions, as well as combinations of the two approaches. 8.8 Modular Product Design As the workflow market matures, different vendors may specialize in different areas of the classical workflow product. The Workflow Management Coalition, in recognition of this trend, has begun to develop interface standards between the major subcomponents of a workflow product. The Coalition has been well accepted by both the vendor and user community of the workflow market. Mission-critical workflow products must be designed to conform to these standards. A modular product architecture with well-defined interfaces between subcomponents can ensure that compliance with the ongoing development of standards can be achieved at reasonable cost. 8.9 Environmental Independence The underlying technology environment in which workflow products must operate continues to change rapidly. Well designed workflow products must operate independently of key technology infrastructures such as database, server, and network. It is possible today to utilize standards such as SQL and ODBC to develop database-and network independent workflow products. The workflow product should support market-leading databases such as Oracle, Sybase, and Informix. For large Lotus Notes installations, the workflow product should also be able to use the Lotus Notes database as a back end data store as well as provide an API for integration with the Notes client interface. Even as the migration to client-server technologies continues, there are still many mainframe-based line-of-business applications. Products which can extract data from legacy systems, and utilize that information to define and drive the workflow process, will add significant value to the, business process. Client/server based workflow products which support relational databases, butt do not force organizations
Copyright © Genetic Computer School, 2002
Page101 to convert data from legacy systems into relational databases, will provide the best of both worlds. 8.10 Workflow and the Emerging Information Infrastructure Workflow systems operate not in isolation but rather within an information systems infrastructure (see Figure 8.3 below).
Figure 8.3 As workgroup computing has increased in importance, a new class of systems infrastructure called "groupware" has emerged. Groupware environments are a collection of software which facilitate document delivery and collaboration between multiple users. Groupware makes corporate information available to anyone on the network enabled with the groupware product. Groupware capabilities may include electronic mail, routing, foldering, and document replication. Groupware environments are the backbone upon which users will expect document management systems and workflow systems to operate. Major commercial groupware products include Lot Notes, Novell GroupWise, and Microsoft Exchange. Lotus Notes is the current groupware market leader with approximately 4.5 Million
Copyright © Genetic Computer School, 2002
Page102 installed seats. Groupware products include built-in electronic mail and database replication where data is distributed based on the user group. Each user group contains an object database which may include groupware forms as well as other document types. The rapidly growing popularity of the Internet and web (HT-FP) servers` threatens the market position of the major commercial groupware products. Many analysts in fact anticipate that the Internet as well as major corporate Intranets will become the dominant groupware environment of the future. The Internet provides several significant advantages over the currently installed technology infrastructure: • Real-time information access to massive data servers providing large volumes of information, including information external to the corporation. • Low incremental cost per user relative to current user workstations. The Internet relies only on a "thin client" to retrieve information from back-end servers, and may in fact open up new opportunities for lower cost client hardware. Low cost multi-platform development environment. Very simple, "point and click", standard user interface The potential of an easier distribution strategy for new software and software updates, where users can subscribe to software for periods of time, rather than purchasing software which must be installed on each client. Major corporations are establishing internal Intranets, in an effort to realize the advantages of the Internet but "firewalling" their internal environments from the array of possible intruders on the vast public Internet. According to a recent article in PC Week, over 60% or Fortune 1000 corporations either currently have, or plan to install, a corporate Intranet. The Internet has no built-in database structure, as one would find in a groupware product. However, major relational database vendors such as Oracle, Informix, and Copyright © Genetic Computer School, 2002
Page103 Sybase have announced their own Web servers so that Web pages can be dynamically generated "on-the-fly" to serve as a front-end to these relational databases. This potential convergence between the Internet front-end and the relational database backend threatens the position of many commercial groupware products. The emerging battle between traditional Client/Server configurations and the Internet/Intranet is likely to last for a few years with no conclusive winners. Each technology infrastructure has its own significant advantages. Traditional client/server has the advantage of being proven, mature technology with a large installed base. The Internet/Intranet provides a platform independent, information publishing environment at low incremental cost per user. While these infrastructures will provide a "groupware backbone", none of them provide the key capabilities of a robust workflow product: The ability to streamline a mission-critical business process through a structured workflow definition, task scheduling, task automation, and process metrics. The ability to establish and manage a complex process definition through an intuitive graphical interface. The ability to seamlessly integrate with key technologies such as relational databases, windowing environments, document management, and imaging. The ability to rapidly generate sophisticated and yet secure applications from application frameworks. Workflow products will support a number of integration strategies to groupware technologies. For integration with groupware products, the groupware database should be supported at the back end, while the Workflow client should be seamlessly integrated with the groupware client via OLE automation. For the internet and
Copyright © Genetic Computer School, 2002
Page104 intranets, one integration strategy will be to allow a web browser to serve as the client for the workflow engine. Workflow products, which work across the major groupware environments will become essential elements of the emerging information infrastructure required to manage the global enterprise. Workflow products which work in both traditional client-server environments as well as with the Internet/intranet will provide the best of both worlds, taking full advantage of each world while offering the buyer the opportunity to use a common workflow product across the corporation. 8.11 Teleprocessing systems This is a classic method of supporting a multi-user database system, which uses one computer and one CPU. Users operate dumb terminals that transmit transaction messages and data to the centralized computer. Since there is little intelligence at the users' end, all commands for formatting the screen must be generated by the CPU and transmitted over the communication lines. This means the users interface is generally character oriented and primitive. All inputs and outputs are communicated over a distance to the centralized computer for processing.
Copyright © Genetic Computer School, 2002
Page105
Chapter 9: groupware Introduction A workgroup support system (WSS) is a system that is designed specifically to improve the performance of teams by supporting the sharing and flow of information. The foundation of any WSS is groupware - the popular term for the software component that supports the collaborative efforts of a team. SSs primarily support the information-processing tasks of conveying, creating, and communicating. WSSs also work with all types of information - internal, external, objective, and subjective. Groupware is among the newest IT-based tools that you'll find in an organization. Below we've listed the three functions that groupware supports and the software that facilitates that support. 1. Team Dynamics ♦ Group Scheduling Software ♦ Electronic Meeting Software ♦ Videoconferencing Software ♦ Whiteboard Software 2. Document Management ♦ Group Document Databases 3. Applications Development ♦ Workflow Automation Software Comparing those design options across applications yields interesting new perspectives on well-known applications. Also, in many cases, these systems can be used together, and in fact, are intended to be used in conjunction. For example, group
Copyright © Genetic Computer School, 2002
Page106 calendars are used to schedule videoconferencing meetings, multi-player games use live video and chat to communicate, and newsgroup discussions spawn more highlyinvolved interactions in any of the other systems. Consider how these systems can be integrated in other ways. We are still quite far from developing the grand groupware system that encompasses every type of communication, and we will probably never get there since the possibilities are constantly evolving with changes in both our patterns of social interaction and the technology we have available. The features of each of the above technology tools will be considered: 9.1 Group scheduling software and technologies Group scheduling software ♦ is a component of groupware ♦ is a part of the electronic meeting support component of groupware ♦ provides facilities for maintaining the day-to-day electronic calendars of team members and evaluating those calendars to schedule optimal meeting times ♦ may even let you reserve a certain room for a meeting and any equipment that you may need Office Tracker ♦ Can schedule rooms and other facilities ♦ Supports both personal and group scheduling ♦ Finds available times for groups ♦ Displays side-by-side group schedules ♦ Automatically posts notifications and reminders
Copyright © Genetic Computer School, 2002
Page107 ResSched ♦ Can schedule by fraction of the hour ♦ Can schedule equipment, space, people, and other resources ♦ Supports schedule reviewing by participant, meeting type, etc. ♦ Provides alternatives for booking conflicts ♦ Restricts who can schedule meetings WallCHART ♦ Can schedule teachers and classrooms ♦ Used by United Nations for scheduling meetings ♦ Provides good reporting features (e.g., room utilization) ♦ Works on intranets and the Internet ♦ Large international presence Livelink OnTime ♦ Personal calendars can be dispersed across multiple servers ♦ Works on Windows '95, '98, and NT and Mac clients ♦ Scalable to tens of thousands of users ♦ Supports browser access COMSEC Scheduling Software ♦ Rooms, resources, activities, and office appointments ♦ Attach notes and service requests for schedules
Copyright © Genetic Computer School, 2002
Page108 ♦ Use FM WebView to check schedules via the Internet ♦ Works under Windows 3.1/Nt/95/98 and OS2/Warp Visto Briefcase ♦ Primarily a personal information manager with group-support functionality ♦ Enables relationship identification between workers on the road administration ♦ Supports group appointment setting and e-mailing ♦ Shows private appointments as blocked out times ScheduleSoft ♦ Rules-based schedule generation system ♦ Runs across all operating system platforms ♦ Supports use of SQL-Server ♦ Runs on LAN or WAN Meetingmaker ♦ Scalable to tens of thousands of users ♦ Works under DOS, Windows, and Mac ♦ Provides for laptop, palmtop, and remote access ♦ Provides instant access to free/busy time information for both users and resources ♦ Runs in a standard browser
Copyright © Genetic Computer School, 2002
Page109 Electronic meeting software: ♦ is a component of groupware ♦ lets a team have a "virtual" meeting through IT ♦ supports both synchronous and asynchronous meetings MeetingWorks for Windows ♦ Supports brainstorming, idea organization, ranking, and voting ♦ Generates high quality reports and graphs of meeting information ♦ Supports joint application development (JAD) ♦ Free Internet download and use for up to 8 participants ♦ Also available in MeetingWorkds Internet Edition and MeetingWorks Connect Groupsystems ♦ Supports brainstorming, list building, information gathering, voting, organizing, prioritizing, and consensus building ♦ Supports strategic planning activity-based costing, business process reengineering, and knowledge management ♦ Includes utility for on-line surveying ♦ Supports graphical business process analysis and redesign GroupSystems for Windows ♦ Supports synchronous and asynchronous meetings ♦ Supports anonymous input by participants ♦ Can be integrated with whiteboard software Copyright © Genetic Computer School, 2002
Page110 ♦ Facilitates ranking of issues by secret ballot Microsoft NetMeeting ♦ Allows people to share audio, video, and computer files ♦ Can be integrated with whiteboard software ♦ Supports Chat features which allows geographically dispersed participants to communicate messages quickly and easily ♦ Supports simultaneous application sharing Meeting Builder ♦ For organizing and delivering special events ♦ Triggers further investigation of details so nothing is left out ♦ Can be used in conjunction with Hotel personnel package Meeting Center 2000 and MyPlaceWare Professional ♦ Real time Web conferencing ♦ Powerful presentation features such as the ability to demonstrate a live application ♦ Combines real-time visual content through a Web browser Videpconferencing software Videoconferencing software: ♦ is a component of groupware ♦ is part of the electronic meeting support component of groupware
Copyright © Genetic Computer School, 2002
Page111 ♦ allows teams to have meetings and see each other when the team members are geographically dispersed NetMeeting ♦ Made my Microsoft ♦ Works on Windows 95, 98, and NT ♦ Works over the Internet or an organization's intranet Includes application sharing whiteboard capabilities EVX ♦ Made by Avistar Systems Corporation ♦ Allows full-window or mosaic view of team members ♦ Connects to NTSC (North American TV video standard) or PAL (TV standard in many other parts of the world) ♦ Has enterprise messaging capabilities Bright Light ♦ Made by Avalon Information Technologies ♦ Includes an information management system ♦ Allows multimedia-rich content delivery Omega Product Line ♦ Made by VSI Enterprises ♦ Various products for all levels of videoconferencing
Copyright © Genetic Computer School, 2002
Page112 ♦ Controls all aspects of conference via on-screen icons and mouse movements ♦ Includes
diagnostic
software
allowing
problem
diagnosis
or
reconfiguration from remote sites CineVideo/Direct ♦ Made by CINECOM Corp ♦ Works over phone lines using 28.8 KB or faster modems ♦ Supports one-on-one videoconferencing and live broadcasts ♦ Can be used with video capture devices PictureTel Live 200 ♦ Made by PictureTel ♦ Compatible with Windows 95 and Windows NT ♦ Allows the sharing of applications ♦ Allows transfer of files CU-SeeMe ♦ Videconferencing for the Internet ♦ Can be used on any TCP/IP network ♦ Has a CU-SeeMe Contact List to allow contact with just an e-mail address Intel PC Camera w/ Video Phone Software ♦ Enjoy special occasions with family and friends
Copyright © Genetic Computer School, 2002
Page113 ♦ With Pro pack video-in feature, share video from your camcorder or VCR while you talk ♦ Hold a point-to-point videoconference with clients of colleagues MyPlaceWare ♦ Present any type of information in a virtual conference ♦ Integrates all types of media into a presentation ♦ Select a time for the meeting and send e-mail addresses an invitation with the URL QVIX/CU30 ♦ High quality, low bandwidth, real-time full video ♦ Videoconferencing for LINUX Whiteboard software Whiteboard software: ♦ is a component of groupware ♦ lets team members interactively edit and share documents ♦ can capture hand-written notes on an electronic whiteboard SMART Board ♦ Captures notes written, on an electronic whiteboard ♦ Displays those notes on larger screens and individual computer screens ♦ Supports geographically dispersed video and data conferencing ♦ Even recognizes pen color and saves in that color Copyright © Genetic Computer School, 2002
Page114 InPerson Whiteboard ♦ Allows you to view, mark up, scale, and manipulate 2D and 3D models ♦ Allows you to share and manipulate information from Windows-based applications ♦ Can capture images from video streams ♦ Can import files and images GroupSystems Whiteboard ♦ Can display images for review and annotation by participants ♦ Works with text as well ♦ Facilitator can control who has update or view only privileges ♦ Saves whiteboard contents for later use TeamBoard ♦ Supports mouse point & click concept with your finger ♦ Captures notes and drawings electronically ♦ Can communicate via LAN or the Internet ♦ Can also distribute notes and drawings via e-mail Soft Board ♦ Supports remote collaboration ♦ Supports simultaneous access to spreadsheets, the Internet, and presentation tools ♦ Electronic pen acts as on-board mouse Copyright © Genetic Computer School, 2002
Page115 ♦ Available in plasma display technologies Group Logic Imagexpo ♦ Interoperable with Windows and Mac ♦ Adds color slides and documents to phone calls ♦ Lets you create softproofs from all common graphic arts formats ♦ Annotation can be made in different, password-protected layers with unique names TearnWave ♦ Use whiteboard for meetings, note taking, design, discussion, and more ♦ Share files, URL links, images, etc. ♦ Communication
through
text
and
audio-video
based
synchronous
communication MicroTouch Electronic Whiteboard Ibid 2.1 ♦ Simplifies whiteboard teleconferences by detecting use of NetMeeting ♦ E-mail, print, drag-and-drop, fax, or post meeting notes to a Web site Work Flow Automation Software? Workflow automatic software ♦ is a component of groupware ♦ supports application development ♦ is designed to automate the flow of business documents in a specific work process or procedure
Copyright © Genetic Computer School, 2002
Page116 Work Expeditor ♦ Designed for use with Microsoft Exchange ♦ Supports SQL server for data access and control ♦ Uses Microsoft Outlook as the CUI ♦ Supports security down to the document level ♦ Provides an O-O business repository MQSeries Workflow ♦ Provides business process reengineering tools ♦ Helps automate processes for the cyber-corp ♦ Uses process diagrams to implement work flows ♦ Allows you to assign staff to data and processes ♦ Works with Lotus Notes and Web browsers Staffware ♦ Supports a Java-based Web client ♦ Allows and organization to map, control, and improve business processes ♦ Supports flowcharting of business processes ♦ Includes Executive Information Service, allowing managers to track a graphical view of a business process PowerFlow ♦ Enterprise-wide workflow solution
Copyright © Genetic Computer School, 2002
Page117 ♦ Allows organizations to graphically design and control processes ♦ Supports SQL database server ♦ Integrates with a variety of third party tools METEOR Enterprise Application Suite ♦ Supports rapid application and deployment of process management applications ♦ Works simultaneously on different servers ♦ Supports Web, CORBA, and Java-based development ♦ Integrates with current legacy systems and information Ultimus Workflow ♦ Includes graphical workflow design tools ♦ Integrates easily with relational databases ♦ For both Internet and intranet applications ♦ Includes automated form designing tools Dolphin ♦ Doesn't require programming skills to model work flows ♦ Works with NT and Exchange ♦ Processes can be modeled in list or diagram form ♦ For small, medium, and large departments
Copyright © Genetic Computer School, 2002
Page118 Cabinet NG ♦ Written in C and compatible with most other languages ♦ Interfaces with most known databases ♦ Works with word processing, spreadsheet, CAD, etc. ♦ Can produce reports and write them to laser disc Action Workflow System ♦ Workflow software based on Action's workflow methodology ♦ Available for Lotus Notes and SQL databases ♦ Metro Application Center brings the workflow technology to the Internet/intranet ♦ DocRoute provides workflow and automated routing to the leading document management systems 9.2 GroupWare suites GroupWare suites are software packages designed to provide a full range of functionality for the complete workgroup environment. These packages often include facilities for group scheduling, group communication, group meetings, whiteboard work, group document management, and applications development. Lotus Note/Domino R5 ♦ Delivers messaging and collaborative solutions ♦ Supports Web applications and integration ♦ Integrates with DB2, Oracle, Informix, and Sybase databases ♦ Integrates with SAP's R/3 enterprise software
Copyright © Genetic Computer School, 2002
Page119 Microsoft Exchange ♦ For messaging and collaborative solutions ♦ Unlimited storage capacity, hosting thousands of users ♦ Supports a variety of security controls and encryption methods ♦ Incorporates built-in Internet standards ♦ Works with Lotus Notes and Domino GroupWase ♦ Includes calendaring, workflow, and much more ♦ Includes images support (e.g., scanning and archiving) ♦ Supports publishing documents on the Web ♦ Y2K compliant LinkWorks ♦ Document-based business oriented ♦ Supports imaging software such as PowerScan and StorageWorks ♦ Supports O-O architecture ♦ Runs under Windows, Mac, and Internet browsers ♦ Scalable for enterprise-wide deployment TeamWARE Office ♦ Supports enterprise-wide discussion forums ♦ Provides team calendaring functions
Copyright © Genetic Computer School, 2002
Page120 ♦ Accesses you mobile phone when an important e-mail arrives ♦ Supports multi-level access control Web-4M ♦ Transforms traditional chat rooms into multimedia forums ♦ Collaboration is integrated with mail and Web documents and a distributed file system ♦ Supports platform independence ♦ Supports full-duplex phoning Netscape - SuiteSpot ♦ Manages large-scale, business-critical intranets and extranets ♦ Web-based Express clients for mail and calendar ♦ Deploys extranets globally in many languages and character sets ♦ Integrated software set that forms the basis of the networked enterprise 9.3 Asynchronous GroupWare Email is by far the most common groupware application (besides of course, the traditional telephone). While the basic technology is designed to pass simple messages between 2 people, even relatively basic email systems today typically include interesting features for forwarding messages, filing messages, creating mailing groups, and attaching files with a message. Other features that have been explored include: automatic sorting and processing of messages, automatic routing, and structured communication (messages requiring certain information). Newsgroups and mailing lists are similar in spirit to email systems except that they are intended for messages among large groups of people instead of' 1-to-1
Copyright © Genetic Computer School, 2002
Page121 communication. In practice the main difference between newsgroups and mailing lists is that newsgroups only show messages to a user when they are explicitly requested (an "on-demand" service), while mailing lists deliver messages as they become available (an "interrupt-driven" interface). Workflow systems allow documents to be routed through organizations through a relatively-fixed process. A simple example of a workflow application is an expense report in an organization: an employee enters an expense report and submits it, a copy is archived then routed to the employee's manager for approval, the manager receives the document, electronically approves it and sends it on and the expense is registered to the group's account and forwarded to the accounting department for payment. Workflow systems may provide features such as routing, development of forms, and support for differing roles and privileges. Hypertext is a system for linking text documents to each other, with the Web being an obvious example. Whenever multiple people author and link documents, the system becomes group work, constantly evolving and responding to others' work. Some hypertext systems include capabilities for seeing who else has visited a certain page or link, or at least seeing how often a link has been followed, thus giving users a basic awareness of what other people are doing in the system -- page counters on the Web are a crude approximation of this function. Another common multi-user feature in hypertext (that is not found on the Web) is allowing any user to create links from any page, so that others can be informed when there are relevant links that the original author was unaware of. Group calendars allow scheduling, project management, and coordination among many people, and may provide support for scheduling equipment as well. Typical features detect when schedules conflict or find meeting times that will work for everyone. Group calendars also help to locate people. Typical concerns are privacy (users may feel that certain activities are not public matters), completeness and accuracy (users may feel that the time it takes to enter schedule information is not justified by the benefits of the calendar).
Copyright © Genetic Computer School, 2002
Page122 Collaborative writing systems may provide both realtime support and nonrealtime support. Word processors may provide asynchronous support by showing authorship and by allowing users to track changes and make annotations to documents. Authors collaborating on a document may also be given tools to help plan and coordinate the authoring process, such as methods for locking parts of the document or linking separately-authored documents. Synchronous support allows authors to see each other's changes as they make them, and usually needs to provide an additional communication channel to the authors as they work (via videophones or chat). 9.4 Synchronous or Realtime Groupware Shared whiteboards allow two or more people to view and draw on a shared drawing surface even from different locations. This can be used, for instance, during a phone call, where each person can jot down notes (e.g. a name, phone number, or map) or to work collaboratively on a visual problem. Most shared whiteboards are designed for informal conversation, but they may also serve structured communications or more sophisticated drawing tasks, such as collaborative graphic design, publishing, or engineering applications. Shared whiteboards can indicate where each person is drawing or pointing by showing telepointers, which are color-coded or labeled to identify each person. Video communications systems allow two-way or multi-way calling with live video, essentially a telephone system with an additional visual component. Cost and compatibility issues limited early use of video systems to scheduled videoconference meeting rooms. Video is advantageous when visual information is being discussed, but may not provide substantial benefit in most cases where conventional
audio
telephones
are
adequate.
In
addition
to
supporting
conversations, video may also be used in less direct collaborative situations, such as by providing a view of activities at a remote location. Chat systems permit many people to write messages in realtime in a public space. As each person submits a message, it appears at the bottom of a scrolling
Copyright © Genetic Computer School, 2002
Page123 screen. Chat groups are usually formed by having listing chat rooms by name, location, number of people, topic of discussion, etc. Many systems allow for rooms with controlled access or with moderators to lead the discussions, but most of the topics of interest to researchers involve issues related to unmoderated realtime communication including: anonymity, following the stream of conversation, scalability with number of users, and abusive users. While chat-like systems are possible using non-text media, the text version of chat has the rather interesting aspect of having a direct transcript of the conversation, which not only has long-term value, but allows for backward reference during conversation making it easier for people to drop into a conversation and still pick up on the ongoing discussion. Decision support systems are designed to facilitate groups in decisionmaking. They provide tools for brainstorming, critiquing ideas, putting weights and probabilities on events and alternatives, and voting. Such systems enable presumably more rational and even-handed decisions. Primarily designed to facilitate meetings, they encourage equal participation by, for instance, providing anonymity or enforcing turn-taking. Multi-player games have always been reasonably common in arcades, but are becoming quite common on the internet. Many of the earliest electronic arcade games were multi-user, for example, Pong, Space Wars, and car racing games. Games are the prototypical example of multi-user situations "non-cooperative", though even competitive games require players to cooperate in following the rules of the game. Games can be enhanced by other communication media, such as chat or video systems. 9.5 Groupware: Design Issues As with all user interface design, the method used for designing a groupware system is more significant than specific design suggestions. This introduction thus
Copyright © Genetic Computer School, 2002
Page124 begins with the groupware design process. The remaining sections address some of the most common` issues that face groupware designers. The Groupware Design Process It's best to start by gaining a solid understanding of your prospective users, what their goals are, and how they go about their work. For broadly-targeted groupware applications, such as videophones or email, understanding users can boil down to understanding how human beings communicate in the first place. A design is also best informed by conducting user studies on system prototypes. In these cases user testing is often significantly more difficult than with single-user systems for the following reasons: ♦ Organizing and scheduling for groups is more difficult than for individuals. ♦ Group interaction style is hard to select for beforehand, whereas individual characteristics are often possible to determine before a study is conducted. ♦ Pre-established groups vary in interaction style, and the length of time they've been a group affects their communication patterns. ♦ New groups change quickly during the group formation process. ♦ Groups are dynamic; roles change. ♦ Many studies need to be long-term, especially when studying asynchronous groupware. ♦ Modifying prototypes can be technically difficult because of the added complexity of groupware over single-user software. ♦ In software for large organizations, testing new prototypes can be difficult or impossible because of the disruption caused by introducing new versions into an organization.
Copyright © Genetic Computer School, 2002
Page125 When designing groupware, it is often best to begin with field studies. The goal is to understand a particular type of group or organization that will be using the groupware system. A number of different studies can be conducted: interviews, surveys, analysis of artifacts used in the work process, examination of processes and workflows, etc. In all cases, the object is to identify the users' tasks and goals, understand how the group communicates and determine the power structures and roles. One key challenge is to appear non-threatening and objective to the users in order to obtain accurate information and to insure that they will accept any design that results. Another challenge is translating the findings from one organization to others -- this is especially a concern when the groupware is intended for organizations which are truly unique or too large to effectively study. Adoption and Acceptance Many groupware systems simply cannot be successful unless a critical mass of users chooses to use the system. Having a videophone is useless if you're the only one who has it. Two of the most common reasons for failing to achieve critical mass are lack of interoperability and the lack of appropriate individual benefit. Interoperability In the early 90s, AT&T and MCI both introduced videophones commercially, but their
two
systems
couldn't
communicate
with
each
other.
This
lack
of
interoperability/compatibility meant that anyone who wanted to buy a videophone had to make sure that everyone they wanted to talk to would buy the same system. Compatibility issues lead to general wariness among customers, who want to wait until a clear standard has emerged. Perceived Benefit Even when everyone in the group may benefit, if the choice is made by individuals, the system may hot succeed. An example is with office calendar systems: if everyone enters all of their appointments, then everyone has the benefit of being able to
Copyright © Genetic Computer School, 2002
Page126 safely schedule around other people's appointments. However, if it's not easy to enter your appointments, then it may be perceived by users as more beneficial to leave their own appointments off, while viewing other people's appointments. This disparity of individual and group benefit is discussed in game theory as the prisoner's dilemma or the commons problem. To solve this problem, some groups can apply social pressure to enforce groupware use (as in having the boss insist that it's used), but otherwise it's a problem for the groupware designer who must find a way to make sure the application is perceived as useful for individuals even outside the context of full group adoption. Avoiding Abuse Most people are familiar with the problem of spamming with email. Some other common violations of social protocol include taking inappropriate advantage of anonymity, sabotaging group work, or violating privacy. The Commons Problem If a village has a "commons" area for grazing cattle then this area can be a strong benefit to the community as long as everyone uses it with restraint. However, individuals have the incentive to graze as many cattle as possible on the commons as opposed to their own private property. If too many people send too many cattle to the commons, the area will be destroyed, and the whole village is worse off as a result. There are a couple of straightforward solutions to the Commons Problem: an appropriate fee can be charged for each head of cattle or a limit can be imposed on the number of cattle any individual may bring. These solutions are an appropriate starting point for solving problems of abuse in GroupWare. CSCW (Computer-Supported Cooperative Work) refers to the field of study which examines the design, adoption, and use of GroupWare. Despite the name, this field of study is not restricted to issues of "cooperation" or "work" but also examines competition, socialization, and play. The field typically attracts those interested in software design and social and organizational behavior, including business people, Copyright © Genetic Computer School, 2002
Page127 computer scientists, organizational psychologists, communications researchers, and anthropologists, among other specialties. 9.6 How is GroupWare Design Different from Traditional User Interface Design? GroupWare design involves understanding groups and how people behave in groups. It also involves having a good understanding of networking technology and how aspects of that technology (for instance, delays in synchronizing views) affect a user's experience. All the issues related to traditional user interface design remain relevant, since the technology still involves people. However, many aspects of groups require special consideration. For instance, not only do million-person groups behave differently from 5-person groups, but the performance parameters of the technologies to support different groups vary. Ease-ofuse must be better for GroupWare than for single-user systems because the pace of use of an application is often driven by the pace of a conversation. System responsiveness and reliability become more significant issues. Designers must have an understanding of the degree of homogeneity of users, of the possible roles people play in cooperative work and of who key decision-makers are and what influences them. Why Is GroupWare design worth paying attention to in the first place? GroupWare offers significant advantages over single-user systems. These are some of the most common reasons people want to use GroupWare: ♦ to facilitate communication: make it faster, clearer, more persuasive ♦ to enable communication where it wouldn't otherwise be possible ♦ to enable telecommuting ♦ to cut down on travel costs ♦ to bring together multiple perspectives and expertise
Copyright © Genetic Computer School, 2002
Page128 ♦ to form groups with common interests where it wouldn't be possible to gather a sufficient number of people face-to-face ♦ to save time and cost in coordinating group work ♦ to facilitate group problem-solving ♦ to enable new modes of communication, such as anonymous interchanges or structured interactions In addition to the benefits of GroupWare, another good reason to study usability and design issues in GroupWare is to avoid a failed design. GroupWare is significantly more difficult to get right than traditional software. Typically, a GroupWare system can't succeed unless most or all of the target group is willing to adopt the system. In contrast, a single-user system can be successful even if only a fraction of the target market adopts it.
Copyright © Genetic Computer School, 2002
Page129
CHAPTER 10: PERSONAL INFORMATION SYSTEMS 10.1 Data Representation Data and Information Raw data, as it is initially collected, is of little value or use. The human brain organizes audio/visual (and other) inputs and the mind interprets these inputs and assigns meaning to them. Thus a pattern of vibrations in the air is interpreted as conversation or as music, and a pattern of retinal impulses-is a rose or perhaps a lover. One sees and hears, but there is much more to this than just a few organized sensory inputs, for intelligent organization is required to give meaning to the stimuli. Likewise, economic and scientific data consists only of raw numbers (symbols) until it has been organized and interpreted. Once such higher levels of meaning have been attached to data, it is termed information. This intelligent act of attaching meaning to raw data is clearly an abstraction process. Indeed, assigning meaning may to some extent be thought of as a synonym for the whole abstraction forming activity. At a somewhat more concrete level, it is often possible to automate certain repetitious calculating tasks that are part of the process of placing meaning on data collections. These tasks are ideally suited to modern computing machinery and are driven by sets of instructions that achieve the mechanical aspects of the data organization. One can go farther than this and say that certain standard meanings are collected and tabulated, then automatically assigned to the items in the data collection by the computer program. The latter is then just a fast and reliable extension of human intelligence bent to the abstraction task Encoding (Representing) Data However, there are practical issues to solve at a lower level than describing what data processing is. These centre on how data is communicated. Whenever one writes a symbol like "4" or “four," a potential for communication exists, based on the fact that
Copyright © Genetic Computer School, 2002
Page130 these symbols encode a certain idea. By convention, everyone encodes the same idea with these symbols, so communication is possible. Computation devices must also encode data in some consistent way. There are two categories of such codes: 1. External codes. These are usually human-readable characters that are input into a computer or output from it using a keyboard, screen, printer, or other device. The most common form for this data is a character such as "4," "a," "%," and so on. 2. Internal codes. Because computer storage is based on electronic circuits that can be thought of as ON/OFF switches, internal storage is in a different format than that used for human interface with the machine. In this form, it is not directly accessible by a human user. The person using a computing machine does not usually need to know what kind of internal representation is employed for data, because there are input/output programs that translate the data between internal and external formats. There are some issues relating to data representation that do make a difference to programmers, however. Bits and Bytes The modern digital computer can only understand binary representations of data. Binary means "two" -- that is to say, only two "states" can be represented. Over the half century since the computer has been in existance, many ways have been used to electrically or mecahically record binary representations. In the beginning, punched tape and cards were used-- hole or no-hole indicated the binary state. With today's magnetic media, the polarity (positive or negative) is used to indicate these two binary states. Many other methods have been used in the past and presently. For humans (not machines), we use the convenient notation of zero [0] or one [1] to write these two binary states. We call these elemental states bits--the "zero bit" and the "one bit." The term "bit" is derived from "BInary digiT".
Copyright © Genetic Computer School, 2002
Page131 Quite obviously, if zero and one were all we could represent in the computer, we wouldn't have much data. Only rarely is business data represented by a single bit. Bits, however, are similar to atoms. They serve to create more complicated representations-to carry the analogy further, the "molecules" of data that humans can use to encode business information. Bits are assembled into eight-bit patterns called bytes. The basic storage unit in the computer's main memory and secondary storage is the byte. Bits are not stored separately. This technique permits up to 256 possible combinations, which can be assigned any meanings humans desire. Most frequently, they are used to represent binary numbers or alphanumeric codes needed to represent business data. However, they can also be used to represent colors, positions on computer screens, or computer instructions that the CPU can interpret and act upon, such as moving data to a disk or printing a line. 10.2 Database Processing One of the major functions of an organization is the processing of data to generate the information needed to control and manage the organization. Historically, this processing was done manually, but the advent of the computer allowed large companies and government agencies, that could afford a computer and a supporting data processing department, to computerize the processing of data. If a new report was needed, a request would be submitted to the data processing department and a programmer would be assigned to write the program to generate the report. The advent of the microcomputer was the beginning of radical changes that are still evolving today. New software was developed in the area of data base management that allowed the user to manipulate data directly without going through the data processing department. This database management software gave the average computer user a great deal of power in working with a database. Data can now be handled in a way that was formally reserved for the data processing department: the end user can create files, maintain files, select information from the files, sort the files and print reports based on the information in
Copyright © Genetic Computer School, 2002
Page132 the files. More sophisticated application are still programmed frequently using the programming language available with the data base management system. A database is a single collection of data. It can hold the information about many different entities related to the company (an entity being simply a person, place or thing). Databases not only contain the information about these entities (known as attributes), they also contain information about relationships. A database is all the raw data needed to create and provide information that will satisfy the varying needs of users in the organization. In some cases the database may be a mixture of data held on computer and manual filing systems but for the purposes of future sessions we will assume it is all held on computer. Inside the database are all of the items of data that belong to the organisation but this data may have been stored on different systems in different locations (a distributed database). In order to be able to locate where the information is stored, the database requires a management system which effectively provides a map of where data is stored and a programme to locate and retrieve the data specified. In order to overcome the problem of data stored on different systems, written in different 'computer languages', programming staff have had to develop bespoke management systems specifically for the purpose. The Data Base Management System is, in effect, the intermediary between raw information and those who need the information, giving users the freedom to use the data in their own specific way. This may not be the ideal solution - disparate databases often require a high level of maintenance to maintain their effectiveness and there is a substantial cost associated with this. A preferred solution may be to develop a corporate database with a corporate Information Systems structure and framework. The database is the primary means of integration and dissemination of data (refer to figure 10.1)
Copyright © Genetic Computer School, 2002
Page133 The data that is stored or updated by one system is then available to the others. It can also be used to produce specific management information.
Figure 10.2 The higher up the organisation, the less structured information tends to become (refer to figure 10.2) In a previous session, a figure similar to the one above was used to explain the role of different levels of management within an organization. If we consider the types of decision with, which these levels will be involved then there will be a tendency to move from structured decisions (programmed) at the operational level, to unstructured (non-programmed) decisions at the strategic level. Underpinning the pyramid is a representation of the operational and functional systems of the business, i.e. accounting, stock control, debtors/creditors, providing information for the operational level of management. In the past, these systems have also provided the management information for the tactical and strategic levels of management. In effect, the next layer up has received information contextualized by the level below. This has three major disadvantages: i. The information coming from these systems is looking backwards (historical), whereas the higher levels of the management structure are focused on increasingly longer forward projections. ii. Information being passed upwards is a reflection of what that level thinks the next level up requires (but is often not the case). iii. By the time information has passed up the structure, been acted upon and decisions passed back down again, the situation will have altered in any case!
Copyright © Genetic Computer School, 2002
Page134 In summary, a single management information system, which attempts to satisfy the needs of all levels of management, often satisfies none. 10.3 Personal Strategic Planning What are some of the benefits of Personal Strategic Planning? First of all, strategic planning helps provide a personal vision for the future. It is so easy to get wrapped up in the present that we lose sight of where the future might be leading us. If we intend to grow in any significant aspect of our lives, we need to periodically focus on what the future will look like. Depending on where you are in your career and your life, this future vision may be as little as two or three years out or could be as much as twenty or thirty years in the future. The ironic thing about focusing on a period in the future is that what you project for that time is probably not what's going to happen. Specific circumstances, opportunities, threats, and personal preferences may lead you in a distinctly different direction from what you established in your initial planning effort. That does not invalidate the planning process. By focusing on the future, we are able to determine when it is appropriate to change a course of direction. Career direction is a strong concern for most professionals, particularly during the early stages of their careers. It is interesting to note that only a small fraction of professionals end up in the career 'for which they initially prepared. Our interests change; different opportunities present themselves, family obligations impact the direction in which we should be moving. By looking forward, we have a better chance of identifying some of the road blocks that may interfere with where we want to go and identify some of the options that may become available to us. Sometimes, by playing the "what if' game, it will open up new vistas that might not otherwise become apparent to us. While this will not necessarily prevent us from stumbling along the way, there is a greater likelihood that we will be able to respond more effectively to things that can significantly impact where we are going.
Copyright © Genetic Computer School, 2002
Page135 Creating and taking advantage of opportunities is generally much more productive if we keep our peripheral vision open. These "opportunities" are not limited to our profession. They can include such things as financial investments, real and personal property, family and friends, or anything that has a significant meaning in our lives. While there is always a risk in looking at opportunities, that risk tends to be reduced when we have anticipated it and planned how to deal with it. As a counterpart to the risk involved in opportunity, strategic planning can also significantly enhance our personal and business security and safety. This includes, of course, financial security, health and well-being, family concerns, and any factors that might cause significant future trauma. The subject of retirement can arouse a variety of different emotions, depending on where we are in our chronological and professional lives. For many professionals, total retirement may never be a reality. Many of us feel that, as long as we have our physical and mental faculties, we will continue to practice our professions indefinitely. The base and focus of our efforts may change, but we plan to continue practicing our professions to some degree as long as we are able. On the other hand, there are many professionals who look forward to retirement as an opportunity for a significant change in lifestyle that will create new experiences that their careers may have interfered with in the past. Regardless of your own perspective on retirement, when you come to the point where it is appropriate for you to back off from what you have been doing, it will be much more satisfying and fulfilling if it comes as a result of planned effort rather than the closing of a door. The building and maintaining of a balanced life represents another significant benefit of personal strategic planning. Oftentimes, practicing professionals develop "tunnel vision" which keeps them narrowly focused on their business and careers. During certain periods of our lives, this may be justified. However, life is much more than getting ahead professionally or in business. We need to achieve a balance that includes family and friends, health and wellness, personal fulfillment that may not directly relate to our profession, financial planning, spiritual
Copyright © Genetic Computer School, 2002
Page136 development, and service to others. Just as a chair will not function properly if one of its legs is longer or shorter than the other, neither will our lives function effectively without some sort of holistic balance. Giving significant attention to another important aspect of our lives does not necessarily detract from our professional focus. In fact, it is possible to achieve a true synergy wherein the "whole" person can be even more productive professionally as well as personally. A final significant benefit is the opportunity to involve others in making our futures come alive. These could include our life partners, parents and siblings, children, professional colleagues, employers, and friends who have either a vested interest in our success or a genuine concern about our future well-being. Since there are very few things in our lives that we can accomplish without the help of others, it is much more meaningful to get them involved as early as possible in the planning process. Not only are they able to offer significant contributions that may help make our plans more effective, their active support in the pursuit of these efforts can go a long way toward assuring their achievement. Why do Professionals tend to resist Planning? Thinking versus action frequently provides an interesting dilemma for many professionals. We tend to be action-oriented. When we are doing something, we have a feeling of being productive. When we take the time to think, we sometimes feel guilty of being unproductive. While there is no question but that whatever we do should lead to some sort of productive action, in reality very little such action comes without some kind of realistic thinking. Therefore, it is not "either-or"; rather, it is "both-and." In fact, without investing the time and effort required to think about where we are going, there is a very strong likelihood that we will never get there. And, of course, we have to decide where "there" is. It is difficult to do that without putting significant effort into thinking. Time required frequently is a barrier to effective planning. How often have we said "I just haven't got the time to do that right now". While that may be a legitimate reaction occasionally, in many cases it turns out to be just an excuse. It's
Copyright © Genetic Computer School, 2002
Page137 interesting to note that we frequently don't have time to plan the job appropriately ahead of time. However, when the job does not get done right, somehow or other we always find the time to do it over again. Time is one of those exhaustible commodities with which we have to deal as professionals. Consequently, the planning process, while it does take time, could lead us to much more productive use of what limited time we do have available. Changing circumstances frequently is given as a reason for not putting significant effort into planning. "How can we plan when things are changing so fast!" is a frequently-heard expression. As we identified earlier, changing or modifying plans is not only inevitable but actually desirable in many situations. The key, of course, is to recognize when changes are taking place. There is nothing wrong with changing direction, provided we know we are changing and there is a valid reason for doing so. The purpose of a plan is to provide us with a foundation from which we can make appropriate digressions which may be more productive at the time. When things are going well is a time when it is easy to fall into the trap of thinking that they are going to continue to go well "forever." When we are on a highly successful path is probably the time when strategic planning is most critical. Despite the euphoria we may feel at such times, it is absolutely certain that it will not go on "forever." Something will happen that will be other than what we had anticipated, thus moving us, whether we like it or not, in a different direction. Being better prepared for dealing with those situations that may run counter to our desired direction is one of the single most important reasons for doing strategic planning in the first place. While we may not be able to anticipate everything that will have an impact on where we are going, we stand a much better chance of dealing effectively with that if we have looked ahead and anticipated some of the things we might be facing. When things are not going well is probably a more legitimate reason for postponing strategic planning efforts. When your house is on fire is not the time to
Copyright © Genetic Computer School, 2002
Page138 think about installing a sprinkler system. Really, when we are faced with survival, that has to take precedence over where we are going in the long run. For, as one wag said, "In the long run we are all dead." The problem that frequently faces us, however, is that we tend to think we are always in a crisis situation. Unfortunately, this frequently comes as a result of the way we function personally. If you find yourself in a situation where you are moving from one crisis to another on a continuing basis, perhaps that is when you need to take some time off, sit back and really think about where it is that you want to go. We frequently find that these crises come as a result of a lack of effective planning in the first place. At some point, we may need to break that pattern in a way that is going to be more productive for us. Finally, not recognizing the difference between strategic or long-term planning and operational planning may inhibit our efforts. Operational planning, which is commonly what we have to apply in dealing with crisis situations or in achieving our short term results, tends to be quite specific and very much actionoriented. Strategic planning, on the other hand, is much more conceptual and visionary. It is far more directional than specific. If you plan to commit yourself to some form of strategic planning in your own professional life, it needs to be addressed at a time when that is the principal focus. You may find it helpful to have someone else assist you in the process who can keep you focused on the strategic or long-term concerns. Think of strategic planning as helping you determine where it is you are going and operational planning as how you will get there. 10.4 Survey of Personal Information Systems Text Processing A.M. Turing said that if a machine could impersonate a human being, then the machine was thinking. A computer may be said to have artificial intelligence when it exhibits intelligence ordinarily associated with human behaviour (reasoning, learning, use of language, and so on). In the field of AI, there are opposing view points as to whether computers should be programmed to imitate the way the
Copyright © Genetic Computer School, 2002
Page139 human mind works, or whether computers should simply simulate human behaviour in any way practical, whether or not that is the way people think. Computer scientists generally agree that an AI programme is designed to produce knowledge and that such processing includes inferencing (the ability to derive additional knowledge from the original knowledge base). Inferencing capabilities are required, for instance, to fulfil the task of determining phonemes from variable sound or allophones. If "intelligent computers" are ever developed, if artificial intelligence ever approaches human intelligence, it is clear that linguistics will play a major role in these developments, because human linguistic ability is the single most prolific manifestation of intelligence. Computers are masters of the statistical analysis of language. They can be programmed to reveal such properties of language as the distribution of sounds, allowable word orders, permitted combinations of morphemes, relative frequencies of words and morphemes, etc. Existing texts, such as the Bible or Shakespeare's works, can be analyzed, and so can a collection of utterances gathered from spoken or written sources. Such a collection is called a corpus. At Brown University (U.S.A.), a corpus of written American English (consisting of newspaper passages, articles from magazines, and other literary material) was compiled. A corpus of spoken American English, similar in size to the Brown corpus, was also collected. The Brown corpus and the spoken American English corpus were analyzed and compared. This comparison provided a contrast between written and spoken American English. The pronoun "I" occurs ten times more frequently in spoken language. Profane and taboo words are, as expected, more frequent in spoken language. Prepositions occur more frequently in written than in spoken American English, so different syntactic structures are used in written and spoken English.
Copyright © Genetic Computer School, 2002
Page140 Computers can also be used to produce a concordance of a literary text, which gives the frequency of every word in a text and the line and page number of each occurrence. Nowadays, concordances are used to help ascribe the authorship of a text to a certain author. If the concordance of a text -- of which the author is unknown -- is compared to the concordances of known works by two authors, it is possible to say which of the two authors wrote the text. A concordance of "sounds" by computer may reveal patterns in poetry that would be nearly impossible for humans to detect. Poetic and prosaic features such as assonance, alliteration, meter, and rhythm have always been studied by literary scholars. These days, computers can do this mind-numbing work. Electronic Spreadsheets Information Systems oral history and some published newspaper interviews report that in 1978 Dan Bricklin, an M.I.T. alum and then student at Harvard Business School, invented the electronic spreadsheet. Bricklin supposedly was preparing a spreadsheet for an HBS "case study" and had two alternatives: 1) do it by hand; or 2) use a clumsy time-sharing mainframe program. Bricklin thought there must be a better way. He wanted a program where people could visualize the spreadsheet as they created it. His metaphor was "an electronic blackboard and electronic chalk in a classroom." By the summer of 1978, Bricklin had programmed the first working version of his concept. The program would let users input a matrix of five columns and 20 rows. The first version was not very "user friendly" so Bricklin recruited Bob Frankston to improve and expand the program. Frankston expanded the program and "packed the code into a mere 20k of machine memory, making it both powerful and practical enough to be run on a microcomputer". During the summer of 1978, Daniel Flystra joined Bricklin and Frankston. Flystra was also an M.I.T./HBS grad. Flystra was marketing oriented and suggested that the product would be viable if it could run on an Apple computer.
Copyright © Genetic Computer School, 2002
Page141 The three formed Software Arts corporation in January 1979. In April 1979, the company began marketing "VisiCalc", a compression of the phrase "visible calculator". VisiCalc became an almost instant success and provided many business people with an incentive to purchase a personal computer. About I million copies of the spreadsheet were sold during Visicalc's product lifetime. What is a spreadsheet? In the accounting world a spreadsheet was and is a large sheet of paper that lays everything out for a businessperson. It spreads or shows all of the costs, income, taxes, etc. on a single sheet of paper for a manager to look at when making a decision. An electronic spreadsheet organizes information into columns and rows. The data can then be "added up" by a formula to give a total or sum. The, spreadsheet summarizes information from many sources in one place and presents the information in a format to help a decision maker see the financial "big picture" for the company. What came after VisiCalc? The market for electronic spread sheets was growing rapidly in the early 1980s and VisiCale was slow to respond to the introduction of the IBM PC that used an Intel computer chip. Mitch Kapor developed Lotus and his spreadsheet program quickly became the new industry standard. It not only made spreadsheet formula easier by using the shorter, more intuitive Al referencing system (as opposed to Visicalc's R1C1 system) but also added graphics and set spreadsheets on the road to become major data presentation packages, as well as complex calculation tools. Lotus was also the first spreadsheet vendor to introduce naming cells, cell ranges and spreadsheet macros. While at Visicorp, Kapor wrote Visiplot/Visitrend which he sold to Visicorp for $1 million. Part of that money was used to start Lotus Development Corp. Ironically, Kapor offered to sell Visicorp his Lotus 1-2-3 program. Supposedly VisiCorp executives declined the offer because Lotus 1-2-3's functionality was "too limited". To date Lotus 1-2-3 is still the all-time best selling application software in the world. Copyright © Genetic Computer School, 2002
Page142 In 1985 Lotus acquired Software Arts and discontinued VisiCalc. A Lotus spokeperson indicated at that time that "1-2-3 and Symphony are much better products so Visicalc is no longer 'necessary." The next milestone was Excel. Originally written for the 512K Apple Macintosh in 1985 Excel was one of the first spreadsheets to use a graphical interface with true pull down menus and a mouse point and click. The spreadsheet instantly became easier to use than the "archaic interface" of PC-DOS products and many people bought Apple Macintoshes simply so that they could use Excel. Excel never did come out in a DOS version. When Microsoft unveiled the original Windows in 1987, Excel was one of the first products to be written for it, and even now many people still use Excel 2.1 which was written to run under Windows version 2. When Windows finally took off with Version 3.0 in late 1989 Excel was its flagship product. It remained the only Windows spreadsheet for nearly 3 years and has only received any real competition from other products since summer 1992. By the mid 1980s many companies had introduced spreadsheet products. Lotus had acquired Software Arts and the rights to VisiCalc. Also, Microsoft had-joined the fray with the Excel spreadsheet. By the mid-1990's IBM had acquired Lotus and Microsoft Excel was the spreadsheet market leader.
Copyright © Genetic Computer School, 2002
Page143
Chapter 11: executive information system 11.1 Introduction At the strategic level of the organization, a reliance upon operational, historical information is like 'trying to drive a car forward by only looking in the rear view mirror'. Of course, historical information can be useful in identifying patterns and possible future trends but these will be affected by external factors, particularly over long time spans. Executive Information Systems enable data and information to be drawn out from any number of systems, including those external to the business, for example, Stock Market Share information. Models using this information can be constructed and 'what if?' questions posed to help evaluate future positions. For example, a 'model' might be constructed of the potential profitability of a new product over a five years life cycle. The question, "what if a similar product is brought onto the market in year 3 at 5% below our selling price?", can be fed into the model to help provide an answer to that particular scenario. Executive Support System are graphics-oriented systems designed for senior management athat provide generalized computing and telecommunication facilities for monitoring and controlling a business. Executive Information Support System software and communication package integrates virtually all of your information sources into a cohesive network. Functions such as manufacturing, purchasing, customer service, accounting, which generate production schedules, inventory usage, income statements, balance sheets, are connected to provide immediate access and dissemination of critical information. Executive support systems are very important for information technology to support the management of business. Executive support systems (ESS) help managers make unstructured and semistructured decisions. They focus on the information needs of senior management and combine data from both internal and external sources. A system is a generalized
Copyright © Genetic Computer School, 2002
Page144 computing and communications environment that can be focused on a changing array of problems. ESS include tools for modeling and analysis. ESS can and do change the workings of organizations. Executives are better able to monitor activities below them, allowing them to push decision making further down in the organization while expanding the executive's span of control. ESS flexibility allows executives to shape the problems, using the system as an extension of their own thinking. ESS offer executives the ability to analyze quickly and to compare and highlight trends, freeing up executives -- and even more so their staffs -- for more creative analysis and decision making. An executive support system (ESS) is a decision support system especially made for senior-level executives, including all the necessary hardware, software, data, procedures, and people. Several commercial software packages are available for specific modeling purposes. An ESS must take into consideration: ♦ The overall vision or broad view of company goals ♦ Strategic, long-term planning and objectives ♦ Organizational structure ♦ Staffing and labor relations ♦ Crisis management ♦ Strategic control and monitoring of overall operations Basic qualities of an EIS are: ♦ capability to look over the situation in your company “in half an hour" ♦ punctual taking steps in eventual deviation from planning activities
Copyright © Genetic Computer School, 2002
Page145 ♦ reduced and merged data, interesting for an executive ♦ ability to communicate with employees by local Email ♦ it doesn't require any keyboard skills from user, executives can use just mouse ♦ very adaptable in concrete user environment and satisfaction of his demands ♦ graphical presentation of data ♦ many kind of printed reports and reviews due the different criteria Executive Information Systems (EIS) have grown in interest and use over the past 15 years. They are a response to inadequacies in Management Information Systems (MIS) which, although capable of manipulating vast quantities of data, are frequently difficult to use and not able to respond to managers' needs with any degree of flexibility. An executive decision maker requires a precise understanding of the current organisational situation and it has been observed (Dreyfus & Dreyfus 1986) that businessmen prefer concrete information, even gossip, speculation and hearsay to the abstracted summary information contained in routine MIS reports. EIS are relatively new tools which purport to provide executives with computerbased information support for decision making. Not surprisingly, one of the specified characteristics of an EIS is that it be ‘user-friendly and require minimal or no training to use' (Watson et al, 1991). Due to the reluctance of executives to adopt information technology for their own work, developers of EIS have been particularly concerned with the user-friendliness of their products, with the result that most EIS have attractive and ease to use interfaces. Despite this, or maybe because of this, the majority of EIS projects are not successful. This concentration on a generic interface between a user and the machine arises from work in HCI (Human Computer Interaction) based on cognitive psychology which, although adequate for lower level information systems, may account for a lack of EIS success when applied to these more complex systems Copyright © Genetic Computer School, 2002
Page146 This leads us to seek an optimal interface design for users who are making executive decisions' and to question whether the current models of HCI are adequate. Questions raised relate to the characteristics of this user population and the way executives may be expected to approach and make use of such a system. Do executives model things differently to others? Do they focus on different information? Do they deal with information in a different way? Are executives a homogeneous group in this regard? Have studies been conducted which examine these questions? In this paper we look at the background to this problem and suggest ways of discovering solutions to this most fundamental problem in EIS interface design. 11.2 Components of an Executive Information System Examining the components of an EIS is dependent upon the definition used. Opinions differ on how much peripheral support, as opposed to amalgamated information, should be provided. Taking the distinction made by Watson et al (1991) between an EIS and an Executive Support System (ESS) we would exclude consideration of support for electronic communication (e-mail, word processing, etc) and organising tools (electronic calculator, automated rolodex, etc). In Watson's view an EIS would be limited to providing information. Rockart & de Long (1988) identify three broad ESS capabilities - Communication, Status Access, and Query & Analysis. These are listed in ascending order of complexity and skill required by executives. Watson excludes data analysis from the definition of EIS but it is included here as an extra or adjunct function which an executive can call from the status access function. Some executives will not want this. Some will ignore this initially but take advantage of the capability as they gain familiarity and develop confidence with the system. The interface for the communication tools can possibly be ignored since these are specified by the offthe-shelf packages adopted generally in an organization. Status Access largely requires an interface to choose the displays which will be made available, the information that will be included and how it will be formatted. Data Analysis will require a more sophisticated interface with provision for input of various types as
Copyright © Genetic Computer School, 2002
Page147 well as more complex choices of what analysis should be done and which data will be used. Status Access and Data Analysis may shade into each other depending on how much control is given to the user in status access to vary displays. If a standard data access tool is used then the system design may be constrained by the tool specified. Excluding Communication/Organizing tools from a system leaves Status Access/Query & Analysis as a single module, which can be looked at in terms of a consistent interface. Another aspect is the scope of the system. A common misconception of EIS is that they are merely an easy to use graphical interface onto a small database containing summarized information on the organization's progress. If the EIS concept is limited to this, it is almost certainly doomed to failure. To implement a successful EIS the system must be seen to include the whole organization and its environment, from the executives who will use the information, to those responsible for ensuring the integrity of the data in the main company databases from which EIS database is fed. In most reports on EIS, developers of these systems have emphasized the notion that "ease-of use" is their prime usability constraint. However, in a study of executive use of EIS, Hasan & Gould (1994), have found that no matter how `friendly' the interface, executives will only use the system if they have confidence in the reliability of the information. Executives in this study stated that it was far more important that information systems for executive use, should have a consistent interface style, allow verification of the accuracy of the information, allow annotation to explain exceptional cases and permit sensible cross-functional comparisons of data from different systems. 11.3 Executive Cognitive There are formal and abstract models of human cognition which deal with ‘language, inference and consciousness' such as Johnson-Laird (1983) but these are not procedural or practical enough to be useful. Rockart & de Long (1988) outline a number of models of the way executives operate but are founded on a rather informal theoretical base lacking a solid psychological foundation. Allen (1994)
Copyright © Genetic Computer School, 2002
Page148 has studied the concept of usability which he defines as the non uniform effects of system characteristics on user performance. He also explored the offering of choice to users in the interface as a means of improving performance. This is important because it may indicate that EIS need to be individually tailored to particular cognitive styles. Er (1988) has classified the cognitive style of managers along two dimensions; the preferred way of getting data and the preferred way of processing it. In the former, an individual may be classified as a 'sensing' (S) type who prefers hard data that deals with specific problems or an 'intuitive' (N) type preferring holistic information that describes possibilities. On the processing dimension individuals may be 'thinking' (T) using logic or other formal means for reasoning or 'feeling' (F) where preference is given to personal terms in decision making. Combining these dimensions gives four possible decision styles: ♦ Systematic (ST) This decision maker prefers hard data and logic such as cost benefit analysis and evaluation research. ♦ Speculative (NT) This type prefers to speculate future possibilities with dicision trees and sensitivity analysis. ♦ Judicial (SF) This type prefers quantitative measures in conjunction with feedback from decision making groups. ♦ Heuristic (NF) Heuristic decision makers place great emphasis on current possibilities and prefer reaching decisions through mutual adjustment. The implications of this are that a decision making process needs to be developed which takes into account the different cognitive styles of managers
Copyright © Genetic Computer School, 2002
Page149 making the decisions. For example 'what if analysis suits the speculative decision maker and cost benefit analysis of hard data suits the systematic type. There is some evidence (Phillips, 1984) that decision makers do not make decisions based on the probability of success but rather on the basis of their aversion to failure due to the more drastic penalties. Phillips goes on to suggest that decision technology should be centred on problem solvers (with experience, intuition and knowledge) supplemented by information technology (computers, software, databases, networks and modelling) and preference technology (value judgements, time and risk preferences and trade offs). 11.4 Psychological Profiling There seems to be a role for psychological profiling of executives as a first step in the design of an EIS. These profiles are currently used extensively in organisations for providing management with information about the future potential of employees. The most common form of administering this profiling is by a video taped interview with a professional psychologist. The tape is then viewed and responses by the subject transferred to profile data which is input to a statistical package for analysis. The design of the structured interview questions is usually determined by close consideration of company policy and desired characteristics of future executives. This technique seems readily adaptable to the analysis of executive requirements for EIS. For example the determination of the class of executive along the lines suggested by Er (1988) would be an important first step. Other factors to consider in this profile would be to determine whether the executive focussed on key issues and preferred to skip details, whether they were able to make decisions in the absence of real data and what are their reactions to being forced to make a decision under pressure. 11.5 The Psychological Basis of EIS To obtain user requirements and draw up user profiles for the development of any computer system it is necessary to have all understanding of its prospective users. EIS are no different in this respect except that the users are high powered
Copyright © Genetic Computer School, 2002
Page150 company executives. With conventional computer-based information systems, cognitive science has been used with some success in defining user characteristics as part of an "information processing loop" (Kaptelinin 1994). It is a form of natural-scientific materialism that has evolved in the wake of the scientific and technological revolution associated with the building and use of computers. The human mind was seen as a special kind of information processing unit and, when viewed in this light, the similarity to the newly emerging computer technology was compelling. But, typically, it fails to distinguish between the psychic in the human and the information processes of the machine. When cognitive psychology is applied to defining user characteristics, the HCI system is seen as being composed of two information processing units, the human beings and the computers, so that the output of one unit is the input to the other. In this light, the role of the computer is seen by many as "replacing" the human in intellectual spheres, so that the two loops both human and computer are compatible and in fact identical so that they can be considered as one extended loop. However, as Kaptelinin (1994) points out, this combined information processing loop, made up of two other loops, is itself closed, making it difficult to take into consideration any phenomena that exist outside the loop. Dynamic, interactive systems such as EIS highlight the severe limitations of cognitive science as a theoretical basis to explain the interaction of people with the computer. It does not provide an appropriate conceptual basis for computer use in its social and organisational context in relation to the goals, plans and values of the user (Kaptelinin, 1994). A more holistic approach such as that provided by Activity Theory may be more helpful. Activity Theory is able to conceptualise humans in their context and positions tools as the means of mediating work activities and developmental processes. Taking on an Activity Theory approach means that activity is the core point of departure for any kind of human research. It provides a "bottom line" from which to start analysing any type of human interaction be it with a computer or any other tool. This centralised approach in research is in stark contrast to cognitive science which by virtue of its unsound basic premise of the human as an information processor, has been forced into a piecemeal approach. It has also been argued that one of the main Copyright © Genetic Computer School, 2002
Page151 weaknesses in traditional Western psychology is the establishment of so many individual schools of psychology that none of them will ever be able to investigate any single problem in enough detail to arrive at a satisfactory answer (Wertsch 1981). Activity Theory, however, is not a monolithic "theory", but more a frame to be filled out and a set of insights to be utilised - a set of basic principles evolved from a dialectic materialistic approach to understand human life. All the principles that have been delimited in Activity Theory are no more than premises that determine the general direction of development of contemporary psychology (Asmolov 1987). An analysis of human-computer activity would permit more complete use of the possibilities of the development of human activities, which are revealed thanks to computers. The main applied psychological problems revolve around how to make sure that people using the computer are able to further improve their thinking. An EIS itself does not have a clear cut goal as do most conventional computer based information systems. It is rather a tool which has the flexibility to be used as a source of information when needed. On the other hand executives are often very goal oriented in their work on a specific task and rely on sets of intuitive procedures which are known to have worked previously. This relates well to an Activity Theory approach as outlined in Boedker (1991) where she maintains that all activity is bound to a goal and/or an object and the characteristics of the goal or object partially determine and structure the activity. The goals or objects of activity, undertaken by executives in their day-to-day tasks, vary and cannot be anticipated when an EIS is being developed. This concept is rarely appreciated by EIS developers. The relationship of goals, objects and activities, as found in Activity Theory, provides a framework on which new development. methodologies can be created for the building of flexible EIS where the goals of the user can be determined as the system is used. 11.6 Conclusion Mediated mental processes are the central thesis of Activity Theory. Just as the use of language represented a new stage in the development of human higher mental processes can we, as Tikhomirov (1981) poses, "distinguish a new stage in the
Copyright © Genetic Computer School, 2002
Page152 development of higher human processes" as a result of using the computer. If computers were able to "think" we need to be able to distinguish how mediation of mental processes by the computer differs from mediation by signs such as language. He explains why it is "not a new stage in the path to internal mediation" but "is the further development of external mediation". The use of computers in storing and searching for information, in solving pre-programmed problems and other computer specialities, he says, does not mean the "disappearance of human thought, but the reorganisation of human activity". This leads to the appearance of new forms of mediation in which the computer as a mental activity tool ultimately transforms the very mental activity which it was devised to assist.
Nowhere is the concept of mediation more evident than in the use of EIS for organisational strategic decision making. EIS is used as a flexible and exploratory tool by the executive to identify problem areas and windows of opportunity thus transforming the very activity it was designed to assist. In this way by continued EIS use the executive's internal model of the organization is modified and new demands are made of the EIS. The EIS must be flexible enough to adapt to these changing requirements and to evolve along with the executive if it is to take its place as a useful tool in a long line of other artefacts adopted by humans EIS over the centuries. Much work has still to be done in this area in the adaptation of currently available psychological profiles so that they provide information about executives, which can be used as input to HO design of EIS interfaces. But the topic seems to be promising as a future research area.
Copyright © Genetic Computer School, 2002
Page153
CHAPTER 12: DECISION SUPPORT SYSTEMS 12.1 Decision Support Systems Introduction A Decision Support System is used by Upper Management to make decisions for the future of the company. A DSS is a set of very special computer programs that are connected to all internal networks but use mostly external Wide Area Networks, the information from which is used to make corporate decisions. A DSS uses mostly external data and is more hardware complex than the other forms of MIS. At the operational level of a business, decisions are often required to be taken by relatively junior staff but with the minimum of delay. Decision Support Systems are programmes, which allow staff to input a relatively small amount of specific information, which dependent upon whether or not that data falls inside acceptable parameters, provides a decision. This avoids the need to 'refer upwards' in all but the most exceptional circumstances. A Decision Support Systems (DSS) is a computer-based information system that provides a flexible tool for analysis and helps managers with nonroutine decisionmaking tasks. Some features of DSS are: ♦ Input and output: Inputs consists of some summarized reports, some processed transaction data, and other internal data. They also include data that is external to that produced by the organization. The outputs are flexible, on-demand reports with which a top manager can make decisions about unstructured problems. ♦ Mainly for top managers: A DSS is intended principally to assist top managers, although it is now mainly used by other managers, too. Its purpose is to help
Copyright © Genetic Computer School, 2002
Page154 them make strategic decisions - decisions about unstructured problems, often unexpected and nonrecurring. These problems may involve the effect of events and trends outside the organization. ♦ Produces analytic models: The key attribute of a DSS is that it uses models. A model is a mathematical representation of a real system. The model allows the manager to do a simulation-play a "what-if' game - to reach decisions. The four capabilities of all DSS are: ♦ Representations: Conceptualization of information used in decision making, such as graphs, charts, reports and symbols to control operations. ♦ Operations: Logical and mathematical manipulations of data. ♦ Memory aids: Databases, views of data, libraries and other capabilities to update memory. ♦ Control Aids: Capabilities that allow the user to control the activities of the DSS. Three basic components of a DSS: ♦ DSS database: a collection of current or historical data from a number of applications or groups, organized for each access by a range of applications. The DSS database is separated from active organizational data but contains data extracted from corporate databases. ♦ Model base: a collection of mathematical and analytical models that can be made easily accessible to the DSS user. Various kinds of models may be in the model base, including libraries of statistical, optimization; sensitivity analysis ("what if”) and forecasting models. ♦ DSS software: permits easy interaction between users and the DSS database and model base and delivers the end-user interface.
Copyright © Genetic Computer School, 2002
Page155 A DSS assists management decision making by combining data, sophisticated analytical models and user-friendly software into a single powerful system that can support semi-structured or unstructured decision making. These systems help end users utilize data and models to discuss and decide semi-structured and unstructured problems, but they do not solve the problems for the user. Unstructured problems are novel, non-routine and have no predefined algorithms or solutions. DSS combine data with models to produce various alternative scenarios that can be used for making choices. In large organizations decision making is inherently a group process, and DSS can be designed to facilitate group decision making. DSS generally use smaller amounts of data than MIS, and do not need on-line transaction data. DSS have a smaller number of important users, and tend to employ more sophisticated analytic models than MIS. DSS are customized to specific users and so require even more user participation than building MIS. Moreover, DSS are continually evolving and changing, so they must use a flexible, iterative method of development, usually prototyping. Successful management depends on one's ability to make good decisions. Good decision making relies on the availability of information. The Decision Support System (DSS) can provide that information. Today DSS uses operational files from Purdue's human resources and financial information systems to create DSS tables. Business offices and academic units need, this information to manage their operations. DSS allows access to an Oracle server containing the administrative database. Query tools such as BRIO 3.5 and other software that support Open Database Connection (ODBC) such as Microsoft Excel 5.0 are used to retrieve the data. Information can then be processed in customized and ad-hoc reports, graphics and tables for management decision-making. Decision support systems help the decision making process by providing structured sets of spatial information. Anything from a photograph to a map can be used to support
Copyright © Genetic Computer School, 2002
Page156 decision making and in developing a decision support system, experts are consulted to ensure accuracy and understanding of the information being used. A decision support systems is based on the requirements for making a specific range of decisions and may be quite unsuited to a purpose other than that for which is has been structured. When placed in context with environmental management and geographic information systems, these tools may be described as highly specialised computer-based systems which use expert knowledge and have a specific range of uses. Geographic information systems may be used as decision support systems but not necessarily vice versa. That is, we clan form late decision support systems from various sources of spatial information using a geographic information overlay system which may have no use other than for the range of questions it was designed to help answer. Of course a decision support system may, have some dynamic modelling capabilities not available in a geographic information system, particularly where non-spatial data sources need to be incorporated in a model. Challenge of Decision Support System The challenge/purpose of DSS is to help people avoid mistakes in the various different phases of decision making. The primary mistakes are: 1. Intelligence: Missing Important problems or pieces of information 2. Decision design: ♦ Focusing on the wrong decision ♦ Identify the wrong set of alternatives ♦ Not correctly identifying the outcome of the alternatives 3. Choice: ♦ Misapplying the criteria
Copyright © Genetic Computer School, 2002
Page157 ♦ Using incomplete criteria (Mismatch between criteria and goals) 4. Implementation: Many (A whole, very different problem) Decision Support Systems and Executive Support Systems focus on supporting the fast 3 phases. Components of a DSS 1. Data (Usually drawn from the organizations databases) 2. Models: A collection of models that can be used to determine the outcome of various alternatives 3. Sensitivity analysis features - How do the results change if the values change (not if the model changes) 4. DSS User Interface & Integration Software: The part of the system that lets the users user the data in the model Uses of DSS They started off aimed as helping senior executives, but they have ended up being used extensively by middle managers and other professionals (e.g. engineers). [This is because the types of decisions faced by the two groups differ in terms of their frequency, the kind of information that is needed, the amount of structure in the problems] General examples Pricing decisions, route planning, budget planning, production planning, inventory planning, site selection
Copyright © Genetic Computer School, 2002
Page158 Specific examples • MRP - Manufacturing Resources Planning systems (and the DSS that go with them) Problem is this: •
Orders -> (Some regular, some not regular)
•
Production facilities (How things are produced is different for
different orders) •
Materials -> (Can take time to get there, buy lots - but not too
much, inventory costs money) It is often hard to answer questions like: Can you ship 10 batches of product X by next week? How long will it take to produce a batch of Y? Requires a complete model of the production process - DSS can help by providing this model and detailed information about the current status. • Police Crime map - An Intelligence System Geographic information systems - Providing a useful representation of large data sets (Useful for site planning, ATM placements, etc) • Egyptian cabinet •
Complex decision
• Many alternatives • Relatively unique decision In this case the value of the system was probably the process of building it. To build a DSS that was acceptable they needed to consider in detail the situation - quite possible more detail and with more consistency than they otherwise would have. Copyright © Genetic Computer School, 2002
Page159 12.2 Group Decision Support System (GDSS) Introduction People meet to create better, more efficient and stronger organizations in Innovation Suites designed and built by Group Decision Support Systems. With an ideal mix of technology and personal environment, furniture and open space, the Innovation Suites foster learning, decision making and the formation of corporate visions. These designs have been used by a variety of clients in a number of applications, including strategic planning, procurement evaluation, business process innovation, and large scale change management. The design principles are simplicity, productivity, and an appropriate level of balance between technology and human interaction. A group decision support system (GDSS) enables teams of co-workers to use networks of microcomputers to share information and cooperate on projects. A GDSS is also called workgroup computing. In a GDSS environment, there is usually a big room with something like 40 seats, which means that 40 people can be at the meeting at any one time. There are not only 40 seats but also 40 microcomputers. This enables every participant to have the use of one microcomputer during the course of the meeting. The reason why each participant needs a microcomputer depends on how GDSS works. In the GDSS, with special computer software, the facilitator of each meeting will first make the agenda of the meeting, which will be projected onto a big screen that everyone can see. Then the participants will type simultaneously in their ideas of the topic of discussion on the individual microcomputers next to them. Then the computer will sort the ideas, and then the participants will then vote or comment on which ideas they like or they dislike. In the course of the whole meeting, GDSS stores, categorizes and prints out all the ideas, comments and vote tallies, so that each of the meeting participants will get a summary of the meeting when it ends.
Copyright © Genetic Computer School, 2002
Page160 What so special about GDSS is that it enables meeting participants to simultaneously "talk", when the computer sorts and sends ideas to each of the terminal, all at the same time. That saves a tremendous amount of time, because all these are done electronically instead of manually, and the time saved will enable participants to spend more time manipulating and expressing their ideas. This can consequently increase the productivity and efficiency of the group. The time-consuming benefit also has an added bonus: when productivity and efficiency in meetings increase, it is likely that the team spirit can be consolidated, resulting in an increase of the strength of binding among team members. Besides, under this GDSS, no one can dominate the meeting. This is because of another feature of GDSS. GDSS provides an anonymous scheme, so that whatever you type in the terminal (i.e. your opinion) will be protected. Under this circumstance, no one really knows who is typing what. Because of this, not a single person can dominate the meetings. In the worst case, we might say "some ideas" are dominating the meeting, but this is perfectly fine because this is as a matter of fact an aim of the GDSS: to help meeting participants voice their opinions from an idea-oriented mindset. For example, simply because you have a prejudice against person A does not mean that you are going to reject the idea being proposed in the meeting, because you do not know who is proposing that idea!! Besides, this anonymity scheme will also help those team members who are shy to voice opinions. And with the anonymity, people are likely to be more honest, just as you'll say more, and more honestly on the professor's evaluation form if you know whatever you write will not affect your final grade on the course. This, of course„ is because you know you don't have to worry about the consequences. However, whether this anonymity is good or not can be very controversial. The success of meetings supported by GDSS depend largely on the conduct of the participants. If people are taking advantage of the anonymity system by typing obscene words or foul languages, this system may be banned for the good of the organization.
Copyright © Genetic Computer School, 2002
Page161 The three underlying problems in group decision making that led to the development of GDSS are: ♦ The explosion of the number of decision-maker meetings (managers spending 35 to 70 percent of their time in meetings). ♦ The growing length of decision-maker meetings. ♦ The increased number of attendees at these meetings--optimal size of traditional meetings is only three to five participants. Three elements of a GDSS: ♦ Hardware: including the conference facility itself, the room, tables, chairs, the layout of the facility, electronic hardware, audiovisual equipment and computer hardware; ♦ Software tools: including tools for organizing ideas, gathering information, and ranking and prioritizing; ♦ People: including participants, a trained facilitator and hardware and software support staff. GDSS software tools: ♦ Electronic questionnaires: aid the organizers in pre-meeting planning by identifying issues of concern and by helping to insure that key planning information is not overlooked. ♦ Electronic brainstorming tools: allow individuals contribute ideas on the topics of the meeting. ♦ Idea organizers: facilitate the organized integration brainstorming. ♦ Questionnaire tools: support the facilitators and group leaders as they gather information before and during the prioritization process.
Copyright © Genetic Computer School, 2002
Page162 ♦ Voting or prioritizing tools: make available a range of methods of prioritizing and decision making, from simple voting, to ranking in order, to a range of weighted prioritizing and voting techniques. ♦ Stakeholder identification and analysis tools: use structured approaches to evaluate the impact of an emerging proposal upon the organization, and to identify stakeholders and evaluate the potential impact of those stakeholders upon the proposed project. ♦ Policy formation tools: provide structured support for developing agreement on the wording of policy statements. ♦ Group dictionaries: document group agreement on definitions of words and. terms central to the project. ♦ Additional tools might include: group outlining and writing tools, software that stores and reads project files and software that allows the attendees to view internal operational data stored by the organization's production computer systems. The Commander
Figure 12.1 The Commander (refer to figure 12.1) If the corporate boardroom is both a showcase for beauty and the "heart" of an organization's decision making process, then the Commander Innovation Suite is the ideal platform. Originally designed for the Department of Defense's Air Staff Innovation Center, the Commander is built with the
Copyright © Genetic Computer School, 2002
Page163 highest quality materials and exceptional attention to detail. For your organization, the Commander's classic "c" shape provides executives with a unique sense of team, while the 40" of elbow room will deliver ample personal space and comfort. The Executive
Figure 12.2 As the signature table of the IS line, the Executive is a high quality platform that serves both as an exquisite center-piece and a powerful collaborative tool(refer to figure 12.2). Utilized by Learning Organizations across the country, including GDSS, the Executive addresses the demands of an organization that needs to unleash the creativity and innovation of its employees while maintaining an atmosphere of elegance. The Diplomat
Figure 12.3 Simplicity often produces the greatest results. The Diplomat was forged from this ideology. This table incorporates the basics of group collaboration into a simple, elegant design that is perfect for teams' ongoing process work. (Refer to figure 12.3) Available with or without the incorporation of collaborative technology, the Diplomat's compact Copyright © Genetic Computer School, 2002
Page164 design is ideal for small conference rooms or break out areas. As a stand alone table, or as a compliment to a larger Innovation Suite, the Diplomat will provide your organization with a tool that empowers teams to produce quality results in less time. The Two Top
Figure 12.4 The Two-Top design was born out of a demand for a flexible collaborative tool that could meet a variety of needs in a taxing environment (refer to figure 12.4). Built with the same attention to detail and quality as the full size Innovation Suites, the Two-Top table takes advantage of sliding-covered laptop computers, self-contained networking hardware, and unique "daisy-chaining" capabilities to produce a multitude of configurations. This configurable platform increases the flexibility of boardroom environments, serving as a traditional conference table, a roll-around work station, or a powerful collaborative tool. Advantages and Disadvantages of GDSS Advantages Group Decision Support Systems give groups several advantages over many traditional, non-automated group meetings (Nunamaker, et al., 1991): • More participation Because a GDSS allows anonymity, group members may be encouraged to participate because they do not feel as vulnerable to group censure for asking what may be perceived as "foolish" questions or making unpopular comments. Similarly, the Copyright © Genetic Computer School, 2002
Page165 participants will not be as subject to group think or conformance pressure (the reluctance to criticize the comments of others due to politeness or fear of reprisals). In addition, each group member will have more "air time" or time to contribute ideas. In non-automated meetings, people must listen to others speak and pausing to reflect can cost a turn at comment or response; a GDSS allows everyone to "speak" in parallel. In a typical meeting, group members have only few minutes to express their ideas rather than the entire meeting time. In some non-automated meetings, a few group members may exert undue influence or monopolize the group's time; a GDSS makes every participant equal, eliminating member status incongruities. Finally, more information will be presented to the group as more participate. • Group synergy Other group members will be able to use an idea in a manner that the originator did not because participants have different information skills. Also, the group as a whole will be able to catch errors in a comment better than the individual who proposed the idea. Reading a comment often gives creative stimulus to others in the group. Also, groups may be more likely to consider an idea as the group's idea rather than an individual's because ideas have been merged together. • Automated record keeping A GDSS can record all comments generated during the meeting, and consequently, the group participants may not need to take notes. In a non-automated setting, group members have to remember comments (rather than thinking of new ones) until they have a chance to speak. Participants may also forget what has been said before. In vocal meetings, some participants may not understand what was said or they may not be able to process the information quickly enough. This automated log of the discussion supports the development of an organizational memory from meeting to meeting. • More structure A GDSS also provides a certain amount of structure to the meeting. With this structure in place, it is more difficult to deviate from the problem-solving cycle and make
Copyright © Genetic Computer School, 2002
Page166 incomplete or premature decisions. The group has a more concentrated discussion, and they stay focused on the issues throughout the meeting. Lower levels of non-task interactions (gossiping, for example) in such groups have been observed as compared with traditional meetings. • Other benefits As a result of more participation, group synergy, record keeping, and structure, many groups have been able to accomplish more in significantly less time necessary for traditional, non-automated meetings. Also, these factors have contributed to higher group satisfaction with the meeting process. Finally, the new technology has enabled larger groups to meet, resulting in more information, knowledge, and skills that are brought to bear to the task at hand. • Disadvantages There are some disadvantages to the technology, however, and they include: • Slow Communication Most people speak much faster than they type, and thus would usually prefer a verbal environment (other things being equal). However, a GDSS allows participants to review recorded comments (people may read and scan faster than they can hear and process). Other advantages, including anonymity and parallel communication, may override the slow typing speed. The breakeven point, where it is more efficient to type in parallel rather than speak and listen in sequence, occurs at a group size of approximately eight members (depending upon typing speed). • Not all Tasks are Amenable to GDSSs Group meetings, which involve "one-to-many" communication (for example, a leader lecturing to the group) would not benefit from a GDSS. Only those tasks, which require group members to exchange ideas or preferences efficiently ("many-to-many") would benefit. • Conclusion A Group Decision Support System (GDSS) presents an efficient and effective method for large groups to conduct meetings in which comments or preferences must be exchanged.
Copyright © Genetic Computer School, 2002
Page167
Chapter 13: expert systems 13.1 Introduction Expert systems are computer programs designed to review a set of facts (market conditions) and apply a set of rules (knowledge base) to arrive at the same conclusion that a team of experts would make if presented with the same set of facts. There are two primary components to an expert system: knowledge base, and inference engine. An expert system begins by capturing human expertise in the form of IF/THEN rules. For example, the rule "IF the traffic light is red THEN step on the brake," captures one small aspect of driving expertise. Often dozens (or hundreds) of such rules are required to arrive at solutions to complicated, financial market problems. The next step in the process involves applying the rules against a set of facts to determine which rules apply. Embedded within the expert system is an inference engine, which determines when and how to apply the rules. The unique order in which the rules are applied to a given set of facts allows the expert system to be dynamic. As one might expect given the features and capacity of these technologies, the number of potential good solutions to a problem can be vast. The inference engine helps to quantify the solution sets to provide our portfolio managers with consistent, unbiased, unemotional problem analysis. Recently, expert systems based decision models have been applied to various areas within the auditing domain (McCarthy et al. [1992]). These models function adequately when compared to human judgements. However, expert systems based techniques may not be the only approach to decision problems within the auditing domain. Another approach that is often available consists of conventionally coded decision support computer models. It may be the case that these techniques provide a more efficient and at least equally effective means of addressing problems within the auditing domain. Decision support systems may have a potential for greater efficiency when applied to some audit decision problems. This paper assesses whether decision support systems
Copyright © Genetic Computer School, 2002
Page168 are a match for the effectiveness of expert systems when applied to one particular audit decision problem, namely, the assessment of inherent and control risk in the purchases transaction cycle and the revision of the hours allocated to an audit program based on those assessments. Two computer models were constructed, one based on an expert systems shell (Clips Version 4.2, developed by the NASA Space Labs), and the other a decision support model constructed using a conventional compiler (Microsoft QuickBasic Version 4.5). Both models used the same data set as the basis of their knowledge base. This data set was derived from a case study questionnaire administered to a sample of 80 auditors. The objective of constructing these two models was to compare the output they generated to determine if they were comparably effective. This comparison is primarily concerned with the consistency of the output, but it also addresses other factors such as the format in which the output is presented and the degree to which the computer models were accessible to the user. Before identifying the data used by decision makers and the significance they attach to them, it is necessary to develop a framework within which to ft these data items. The audit risk model provides a decision framework that is prescribed for use by auditors for audit planning decisions. The way that auditor's risk assessments impact on audit planning provides a readily accessible field for the development of computer models. The application of the audit risk model by practicing auditors has attracted considerable research attention (Peters, Lewis and Dhar [1989]) Mock and Wright [1990] examined the link between audit planning judgement and observed levels of inherent and control risk. They argued that risk assessments have a stronger impact on the extent of audit work carried out than on the nature of work carried out and that risk assessments may be subject to change over time. The models examined in this paper use a fixed set of audit procedures; the models adjust the hours allocated for performing those procedures, not the list of procedures. 13.2 Expert System Architecture The user interacts with the system through a user interface, which may use menus, natural language or any other style of interaction). Then an inference engine is used to
Copyright © Genetic Computer School, 2002
Page169 reason with both the expert knowledge (extracted from our friendly expert) and data specific to the particular problem being solved. The expert knowledge will typically be in the form of a set of IF-THEN rules. The case specific data includes both data provided by the user and partial conclusions (along with certainty measures) based on this data. In a simple forward chaining rule-based system the case specific data will be the elements in working memory. Almost all expert systems also have an explanation subsystem, which allows the program to explain its reasoning to the user. Some systems also have a knowledge base editor which help the expert or knowledge engineer to easily update and check the knowledge base. One important feature of expert systems is the way they (usually) separate domain specific knowledge from more general purpose reasoning and representation techniques. The general purpose bit (in the dotted box in the figure) is referred to as an expert system shell. As we see in the figure, the shell will provide the inference engine (and knowledge representation scheme), a user interface, an explanation system and sometimes a knowledge base editor. Given a new kind of problem to solve (say, car design), we can usually find a shell that provides the right sort of support for that problem, so all we need to do is provide the expert knowledge. There are numerous commercial expert system shells, each one appropriate for a slightly different range of problems. (Expert systems work in industry includes both writing expert system shells and writing expert systems using shells.) Using shells to write expert systems generally greatly reduces the cost and time of development (compared with writing the expert system from scratch). 13.3 Components of an Expert Systems Expert systems operate in a fundamentally different manner from conventional "procedural" programs. That is, rather than run through a set of commands in the same way every time, an ES uses "declarative" knowledge and an inference engine to drive the reasoning process. The basic components of an expert system are described below in figure 13.1:
Copyright © Genetic Computer School, 2002
Page170
Figure 13.1 • knowledge base: knowledge from "domain experts" typically represented as IFTHEN rules, • user interface: provides a means for the non-expert to interact with the knowledge contained in the knowledge base, • inference engine: the core of the expert system--determines how the rules in the knowledge base are processed, • working memory: an area of memory containing (a) observed facts, and (b) new facts deduced from observed facts. 13.4 An overview of deductive inference Expert systems work by making deductive inferences. The main advantage of deductive inference is that it allows specific facts to be derived from general information (inductive inference, on the other hand, uses specific facts to generate general hypotheses) To illustrate the basic concepts of deductive inference, consider the wellknown syllogism:
Copyright © Genetic Computer School, 2002
Page171 All men are mortal (general information) Socrates is a man (observation) Socrates is mortal (specific information)
From an ES point of view, the syllogism consists of three parts: 1. an observed fact: Socrates is a man 2. a rule: All men are mortal 3. a new fact deduced from the observed fact and the rule: Socrates is mortal The only difference between this syllogism and the inference mechanism of an expert system is that in expert systems, the rules are represented differently. Specifically, expert systems cannot operate on universal quantifiers such as "all" and thus variables have to be used instead. For example, the rule. All men are mortal would have to be re-written as follows for inclusion in an ES
knowledge base: RULE 1 IF x is a man THEN x is mortal
This so-called "IF-THEN rule" has two parts: an antecedent (IF x is a man) and a consequent (THEN x is mortal). A rule is said to fire whenever its antecedent is satisfied. When a rule fires, its consequent is instantiated and added to working memory. To illustrate this, consider the following "inference trace": Step 0 working memory = observations: Socrates is a man Step 1 RULE 1 is satisfied:
IF x is a man THEN x is mortal
Socrates is a man working memory = working memory + Socrates is mortal
Copyright © Genetic Computer School, 2002
Page172 Note that there is nothing magic here. The expert system simply has the ability to match patterns. For instance, it considers the antecedent of the rule and identifies the following pattern: is a man. Then it searches working memory for a fact of the form is a man. Since a match is found, the variable x is bound to the concrete object "Socrates" and the consequent Socrates is a man is added to working memory. This new fact may be used to satisfy some other rule, and thus the chain of inference continues. 13.5 Inference Strategies The deductive process itself is relatively straightforward. However, there are two fundamentally different approaches to deciding how to navigate the knowledge base: forward chaining and backward chaining. In either case, the problem to be solved is posed as a question, for instance: "Is Socrates mortal?" In forward chaining, deduction is data driven. In other words, the goal is not used to drive the inference process. Instead, the inference engine simply checks the contents of working memory from time to time to see if the question has been answered. The basic flow of the forward chaining process is shown below in figure 13.2.
Figure 13.2
Copyright © Genetic Computer School, 2002
Page173 In backward chaining, deduction is goal driven. The inference starts with the original question and seeks to answer it by matching it with the consequent of a rule. The basic flow of the backward chaining process is shown below in figure 13.3
Put the original question on the back burner and make the unresolved variable the new question
Figure 13.3 There are two important features o± backward chaining. The first is that the goal (or question to be answered) always drives the search strategy. The second is that the current goal changes as new unresolved variables are encountered. The result is that the backward chaining procedure is called recursively until the original question in answered or a sub-goal cannot be resolved.
Copyright © Genetic Computer School, 2002
Page174 13.6 The Problem with Expert Systems The dependence people place on cases poses a problem to those who treat human cognition as being primarily rule-based. Much work in artificial intelligence, for example, is done in "expert systems." These systems are based on the notion that expert knowledge consists of a collection of rules. By determining the rules an expert in a domain uses, the idea goes, we may then simulate expert behavior in that domain. Not surprisingly, expert systems have run into a significant problem: they are brittle. When faced with a problem which bends the rules, they are unable to cope. They fail because they are not grounded in cases. They are unable to fall back on the details of their experience, find a similar case, and apply it. Likewise, they are unable to use similarities between tough problems and previous experiences to update their rules. Their failure to retain cases cripples their ability to learn from their experiences. 13.7 Choosing a Problem Writing an expert system generally involves a great deal of time and money. To avoid costly and embarrasing failures, people have developed a set of guidelines to determine whether a problem is suitable for an expert system solution: 1. The need for a solution must justify the costs involved in development. There must be a realistic assessment of the costs and benefits involved. 2. Human expertise is not available in all situations where it is needed. If the "expert" knowledge is widely available it is unlikely that it will be worth developing an expert system. However, in areas like oil exploration and medicine there may be rare specialized knowledge which could be cheaply provided by an expert system, as and when required, without having to fly in your friendly (but very highly paid) expert. 3. The problem may be solved using symbolic reasoning techniques. It shouldn't require manual dexterity or physical skill. 4. The problem is well structured and does not require (much) common sense knowledge. Common sense knowledge is notoriously hard to capture and represent. It
Copyright © Genetic Computer School, 2002
Page175 turns out that highly technical fields are easier to deal with, and tend to involve relatively small amounts of well formalised knowledge. 5. The problem cannot be easily solved using more traditional computing methods. If there's a good algorithmic solution to a problem, you don't want to use an expert system. 6. Cooperative and articulate experts exist. For an expert system project to be successful it is essential that the experts are willing to help, and don't feel that their job is threatened! You also need any management and potential users to be involved and have positive attitudes to the whole thing. 7. The problem is of proper size and scope. Typically you need problems that require highly specialized expertise, but would only take a human expert a short time to solve (say an hour, max). It should be clear that only a small range of problems are appropriate for expert system technology. However, given a suitable problem, expert systems can bring enormous benefits. Systems have been developed, for example, to help analyze samples collected in oil exploration, and to help configure computer systems. Both these systems are (or were) in active use, saving large amounts of money. 13.8 Rules and Expert Systems Rule-based systems can be either goal driven using backward chaining to test whether some hypothesis is true, or data driven, using forward chaining to draw new conclusions from existing data. Expert systems may use either or both strategies, but the most common is probably the goal driven/backward chaining strategy. One reason for this is that normally an expert system will have to collect information about the problem from the user by asking them questions - by using a goal driven strategy we can just ask questions that are relevant to a hypothesized solution. Anyway, in a simple goal-driven rule-based expert system there are often a set of possible solutions to the problem - maybe these are a set of illnesses that the patient
Copyright © Genetic Computer School, 2002
Page176 might have. The expert system will consider each hypothesized solution (e.g., has_Cold (fred)) and try to prove whether or not it might be the case. Sometimes it won't be able to prove or disprove something from the data initially supplied by the user, so it will ask the user some questions (e.g., "have you got a headache?"). Using any initial data plus answers to these questions it should be able to conclude which of the possible solutions to the problem is the right one. The Distinction Between Expert Systems and Decision Support Systems Expert systems, or knowledge based modelling, represent an application of the artificial intelligence field of research. The objective of this field is to imitate human mental activity. While computers can more than match human capabilities in activities that do not actually require "intelligence" or reasoning ability, they are not a clear match where reasoning abilities are required. The artificial intelligence field can be seen as an attempt to address this deficiency. One difficulty faced by researchers in the artificial intelligence field is to develop a suitable test, and its attendant definition, to identify what constitutes intelligence. Such a test is provided by the Turing Test (Turing [1950]). The test is based on an operational definition where a device is accepted as possessing artificial intelligence provided it performs as a reasoning human would when presented with the same problem. The test is administered as a simple question and answer process. The device to be tested and a human volunteer are both interviewed by an examiner in an attempt to identify which is the artificial device. If the examiner is unable to do this after a series of tests the device is deemed to have passed the test. The essential point of this test is that it places no restriction on what techniques are used to attain a successful result. Thus, if we accept this test, and its implications, artificial intelligence need not be the same form of intelligence, or function in the same way as human intelligence, so long as they provide results that are comparable with those produced by a human expert analysing the same problem. An extension of this position is to say that expert systems based models, and procedural models may
Copyright © Genetic Computer School, 2002
Page177 both be equally valid if they produce comparable results without necessarily duplicating each other's analytical techniques. The central component of an expert system is the inference engine. This is software that combines the contents of the system's knowledge base with case specific input obtained from the user in order to perform the problem solving tasks for which the system was developed. The knowledge base is composed of domain specific knowledge pertaining to the chosen problem domain. An essential characteristic of expert systems is to have a specifically, and often narrowly, defined problem domain. A detailed and specific knowledge base is the true source of the problem solving power of an expert system [Waterman 1986]. The manner in which the knowledge base is used will be determined by the system's meta-knowledge. Meta-knowledge can be seen as a set of rules governing how the knowledge base itself is applied. Meta-knowledge determines what case specific knowledge is obtained from the user and what knowledge from the knowledge base is incorporated in the problem solving process. Meta-knowledge and knowledge complement one another. In the absence of meta-knowledge, knowledge cannot be applied. However, the extent to which we rely on meta-knowledge may also offset the extent to which we need to rely on the knowledge base. An advanced expert systems shell may facilitate the construction of expert systems based models that embody a high degree of meta-knowledge. Such a model will be able to solve problems while making a relatively "economical" use of the systems knowledge base. It will be able to limit its use of the knowledge base to those pieces of knowledge that are most pertinent to the problem being addressed. A model with a less advanced meta knowledge structure may produce the same conclusion, but it would be less discriminating in its usage of the knowledge base. A procedural coded computer model can approach this type of structure. Such a model can access a knowledge base composed of heuristic, or numerically based rules. The programme code, which determines how that knowledge base is accessed and applied' can act in the role of meta-knowledge. However, conventional programming
Copyright © Genetic Computer School, 2002
Page178 languages are not adapted specifically for the construction of meta-knowledge structures while expert systems shells are. For this reason, the scenario outlined above may tend to arise. While it is possible that procedural and expert systems based models may reach comparable conclusions, the procedural model may utilise a considerably larger portion of the knowledge base than is strictly necessary to do so. This need not be a significant shortcoming. The actual cost of performing analyses on an ongoing basis, using what may be a less efficient model, is relatively small when compared to the potential cost of developing a more efficient model that involves much more advanced developmental techniques. Abdolmohammadi [1987] draws a distinction between expert systems, which seek data in order to make a judgement or diagnosis, and decision support systems, which perform analyses that assist in making a decision. Expert systems can be constructed out of heuristic decision rules obtained from experts in the problem domain. The rules, while relevant to the problem domain, will not all be relevant to each problem within the domain. The expert system operates by selecting those rules that are relevant (and thus the information that it should seek from the user) according to the problem outlined to it by the user's responses. This flexible approach inherent in expert systems, which enables them to selectively apply their knowledge base to a problem, renders them particularly suitable for application to unstructured tasks. Since a procedural model embodies a less advanced meta-knowledge structure, its inherently less flexible approach must make a higher level of assumptions about the form of the problem that is to be solved. Such a model applies an algorithm, which aims to produce an optimal solution to that particular problem, and only to that particular problem. To the extent that real world problems diverge from the assumptions of the system's developers, so the solution may be less than optimal (although in practice users of the system often may not recognise this). The more advanced meta-knowledge structure of expert systems models has the potential to reduce the level of assumptions that are made about the problem under analysis. It is our contention that the auditing
Copyright © Genetic Computer School, 2002
Page179 problem we examine here is sufficiently structured to enable it to be addressed by the less flexible meta-knowledge structure of a procedural based model. If problems exist which are capable of being addressed adequately by procedural techniques, as well as by expert systems, procedural techniques have the potential to offer a solution that is developmentally less complex and therefore a more efficient approach to the problem. Waterman (1986) presents three sets of criteria for identifying those situations in which the development of expert systems is appropriate. These are summarised below. 13.9 Criteria for Expert Systems development Necessary requirements for expert systems development: ♦ Task does not require common sense. ♦ Task requires only cognitive skills. ♦ Experts can articulate their methods. ♦ Genuine experts exist ♦ Experts agree on solutions. ♦ Task is not too difficult. ♦ Task is poorly understood. Justification for expert systems development: ♦ Task solution has a high payoff. ♦ Human expertise being lost. ♦ Human expertise scarce. ♦ Expertise needed in many locations.
Copyright © Genetic Computer School, 2002
Page180 ♦ Expertise needed in a hostile environment. Characteristics that make the use of expert systems appropriate: ♦ Task requires symbol manipulation. ♦ Task requires heuristic solutions ♦ Task is not too easy. ♦ Task has practical value. ♦ Task is of manageable size. The first of these, described as "necessary conditions" for the development of expert systems, acre presumably met by many problems within the auditing domain since satisfactory auditing based expert systems have been developed. Since this first set of criteria is presented as a set of necessary conditions that must exist before expert systems development can take place, the author presumably intends that these criteria must be met. Similarly, it would appear that Waterman's third class of criteria, the appropriacy of expert systems, must be met completely. In the absence of these criteria, the development of expert systems will be possible, however, the violation of these conditions would present a situation where the development of an expert system would not be justified, since the problem concerned could clearly be addressed by simpler modelling techniques, or was not worthy of modelling at all. Once again, it must be assumed that the presence of pre-existing systems denotes that problems from within the auditing domain meet these criteria. On the other hand, Waterman's second category, which seeks to codify those situations where the use of expert systems is justified, as opposed to possible, would not appear to need to be met completely in each case. Satisfaction of any one of these criteria would appear to provide a justification for the use of expert systems to address that particular problem, provided no simpler technique were available. It seems that at
Copyright © Genetic Computer School, 2002
Page181 least two of these criteria, being the (relative) scarcity of human experts, and high solution payoffs, are met by problems from within the auditing domain. A further condition should be added to Waterman's second category, "Justification for Expert Systems Development". That condition is that expert systems should provide the most efficient technique available to address the problem being examined. Further, we suggest that this would not be the case where the decision problem is sufficiently structured that a simpler degree of meta-knowledge, as embodied in procedural code is sufficient to address it, and that it will not be the case where a problem in the real world is capable of being characterized by an extensive set of starting assumptions. A further condition should be added to Waterman's second category, "Justification for Expert Systems Development". That condition is that expert systems should provide the most efficient technique available to address the problem being examined. Further, we suggest that this would not be the case where the decision problem is sufficiently structured that a simpler degree of meta-knowledge, as embodied in procedural code is sufficient to address it, and that it will not be the case where a problem in the real world is capable of being characterized by an extensive set of starting assumptions.
Copyright © Genetic Computer School, 2002
Page182
CHAPTER 14: DISTRIBUTED INFORMATION SYSTEMS 14.1 Introduction The litikage of minicomputers in each District (State) office and the Bureau mainframe computer foTrns the, basis for a Distributed Information System (Posson, 1985). The Distributed Information System (DIS) provides a high level of local computing and data-processing capability. Computerized files are transferred between sites (nodes) within the network and work is done at distant locations. The DIS provides interactive and batch processing in support of Water Resources Divisions's State and National water data files, and provides for the diverse computational needs of the Division. These needs include data management, hydrologic modeling and statistics, and administrative programs. The DIS computers are connected to each other via a TCP/IP (Transmission Control Protocol/Internet Protocol) over an ethernet Local Area Network (LAN). The LANs are connected together using a frame-relay Wide Area Network (WAN) called DOINET. The WAN uses Stratacom switches and Cisco routers for the network service backbone. As organizations and users increasingly rely upon networked applications for accessing information and making critical business decisions, securing distributed applications is becoming extremely significant. With the phenomenal growth in networks such as the Internet, information acid security issues are becoming the focus of intense concern on the part of users and businesses. 14.2 Automated Data Processing System The Automated Data Processing System (ADAPS) consists of a collection of computer programs (software) and data files that form a system of standardized water data processing procedures. In each District, there is a designated administrator or manager who is responsible for installation and maintenance of system programs and
Copyright © Genetic Computer School, 2002
Page183 hydrologic data files. In addition, the administrator performs other tasks that are necessary for efficient operation of the system. Most of the data needed by ADAPS programs are supplied interactively: However, some programs accept input from system files. Therefore, the programs require the pathname (data set name) of the input file. Occasionally the user is prompted to supply a pathname (either for input or output), but usually the name need not be known. Most pathnames are stored within a program or in a file containing pathnames, thereby relieving the user of having to keep track of the pathnames. A security system is used in ADAPS to restrict access to the data files and to limit the ability of some users to perform certain operations in the system. The security system is a multilevel system. User classes of System Administrator (SYST), Data Base Administrator (ADBA), User (USER), and Cooperator (COOP) have been established. In addition, Ingres Access Control Lists' (Seybold, 1985, p. 3-1 to 3-28) are used. Security measures are implemented by the local administrator or manager in consultation with District supervisory personnel. Water data stored in ADAPS, results from processing of data collected or recorded at field installations operated by each District. Data collection and processing is discussed in the ADAPS User's Manual. A vast majority of the field data is recorded in digital form on punched paper tapes or recorded by a data collection platform (DCP) and sent to the computer via satellite telemetry. The telemetry input subsystem is discussed later in this manual. The field data is processed by District personnel following established procedures. The data are reviewed for validity and correctness, and subsequently are published in State basic data reports. 14.3 Programs The programs (software) for ADAPS are developed by personnel of the Water Resources Division. The master copy resides at Headquarters in Reston Virginia, and is electronically transferred over the network or distributed by magnetic tape from Headquarters to each of the District offices. These offices are located nationwide and
Copyright © Genetic Computer School, 2002
Page184 include Alaska, Hawaii, Guam, and Puerto Rico. Some Districts have more than one office, and a copy of the software is located at the subdistrict offices, if they have a minicomputer. Most ADAPS programs are written using the Fortran77 programming language, and several different categories of software make up the water data-processing system. The major categories are: • System utility programs. • General-purpose programs. • Command Procedure Language (CPL) programs. • General application programs. • General graphics programs. • Vendor-supplied programs (subroutines). The programs m each of the above categories is used for a specific function or purpose. For example, the utility programs are used to initialize, create, update, and maintain the numerous s u p p o r t , processing, and data (time-series) files that make up the District data bank. The general-purpose programs are used to process many different types of water data along with the subsidiary calculations and computations that go into computing and producing a water data record. Most programs use insert and common blocks to share and communicate data between programs, to provide software flexibility, and to ease software maintenance. The operating system software is called UNIX(Seybold, 1985, p. 1-2). The Command Procedure Language (CPL) is a PRIMOS command level language that provides a programming capability (Landy, 1982). Operating system commands or directives are passed to PRIMOS for execution after they are stored in a CPL file (suffix .CPL). The application programs are used primarily to compute statistical information about the hydrologic data. The graphics programs are used for preliminary viewing of the data, for comparison purposes, and for report
Copyright © Genetic Computer School, 2002
Page185 purposes. The system contains both user-written and vendor-supplied programs such as DISSPLA (Integrated Software Systems Corporation, 1984). The Automated Data Processing System (ADAPS) programs are used to compute water-data records on an electronic computer. The machine operations generally parallel the manual operations. The sequence of processing the data is well established; however, the particulars of each step by electronic means may change in response to continued improvement in storage and access procedures, new or expanded needs, and search for additional efficiencies. The general functions of the programs are to provide input and output to and from the computer in a logical sequence; this sequence includes the computational steps necessary to efficiently compute water data records. Once the records are computed, they must be saved and archived for general use. Therefore, programs argil available to provide for a broad scope of functionality including initialization, maintenance, security, backup, recovery, restart, and other overall dataprocessing requirements. 14.4 Fibs The ADAPS files are the repositories that contain the information and data necessary to use ADAPS for computing and processing water data records as previously mentioned. The files consist of program source code and associated executable code, CPL files, and operation files. Some of these operational files are shared by the various NWIS systems and some are specific to ADAPS. The shared and ADAPS-specific files are briefly described in Section 3 of this manual. Most files used and/or maintained by ADAPS are structured as MIDAS files and managed by a utility and user-written software. The MIDAS files allow records to be retrieved rapidly and efficiently on the basis of selected data elements defined as key indexes (elements). MIDASPLUS utilities, user-written programs, and scripts are used to create the file templates, create input files, and subsequently populate the files. Other utilities are used to dump the files, delete files, cleanup files, and monitor files.
Copyright © Genetic Computer School, 2002
Page186 Some information regarding the file creation was provided in informal instructions (Trapanese, S.M., U.S. Geological Survey, written common., 1987) to install ADAPS on the local minicomputer and to convert existing minicomputer and mainframe data files for use in ADAPS. Additional information concerning initialization (creation) of files is presented later in this manual. ADAPS menu (program) options are selected to perform the various functions necessary to create new MIDASPLUS files. Information is also presented later in the manual to perform incremental backups of the unit-values data files, and to restore the unit-values files using the incremental backups. 14.5 Need for File Maintenance In virtually every data-processing system, the possibility exists that errors may occur that accidently alter or destroy data stored on disks in data files. This may occur because of unexpected hardware failures, natural disasters such as power outages, etc., or through improper processing of the data. It is essential, therefore, to provide a means to ensure that any lost data can be recovered. The most common method used is backup files. A backup file is merely a copy of a file stored on magnetic tape or disk. If the file is destroyed or becomes unusable, the backup can be used to recreate or restore the file. In transaction-oriented (online) systems, backups are critical because updates to a file can occur at any time. Therefore, the system should provide for creating backups on a regular basis (sometimes every hour or day) and saving the transactions that occur to the file after the backup file is made (incremental backups). If the file is lost, it can be re-created from the backup file and then the transactions which have been saved can be processed against the file to bring it back to the status it was before it was lost. Another aspect in a transaction-oriented system is the ability to restart the system if it goes down. For example, if a transaction (update or modification) has been sent and has partially updated a file when power is lost, it is necessary to both restore the file and let the user know that the transaction was not processed. This problem may be solved by
Copyright © Genetic Computer School, 2002
Page187 using the backup files and messages sent to terminals asking the users to re-send those transactions not processed. 14.6 TalkMine: Knowledge Mining in Distributed Information Systems TalkMine, an algorithm developed as a model of human cognitive categorization, improves human-human and human-machine interaction by establishing a powerful tool for collaborative knowledge organization in large distributed information systems (DIS). The adaptive nature of TalkMine entails an open-ended human-machine symbiosis, which can be used in the automatic, adaptive, organization of knowledge in DIS such as the Internet, facilitating the rapid dissemination of relevant information, the discovery of new knowledge, as well as the ability to identify fraudulent behavior in DIS. TalkMine is based on several theories of uncertainty and belief, as well as connectionist ideas. Even though knowledge is ultimately grounded in the brains of the users of DIS, we can devise mechanisms that infer how tokens of knowledge in DIS relate to one another, from the way these tokens are used by a community of users. Tokens of knowledge are all the different types of information stored in the nodes of a network, e.g. text files, databases, links between documents, etc. The meaning of these tokens is ultimately established by the users of DIS, both individuals and societies. Whether artificial information networks can ever produce autonomous creation of meaning is an open question. What they can do exceptionally well, however, is to monitor how their users manipulate and relate the tokens of knowledge stored. TalkMine does not pursue a passive analysis of relationships of knowledge tokens, but rather an active categorization of these tokens by changing the structure of a DIS to adapt it to the changing interests of its users. We refer to this active organization of distributed information as knowledge mining. The kind of knowledge management proposed with TalkMine will result in the evolution of a "Second Generation Internet" or DIS. Theoretically, TalkMine is based on the idea that there cannot be creation and open-ended evolution of new knowledge in artificial systems until a fundamental coupling between structure and semantics is established, that is, until some artificial embodiment is implemented. TalkMine empowers DIS with such a coupling as a result of the continual feedback between the level of
Copyright © Genetic Computer School, 2002
Page188 information utilization by users (semantics and pragmatics), and the relations and links between tokens of knowledge in networks which define the structure of DIS. The ability to gather and combine information from different neighborhoods (e.g different networked databases or WWW sub-regions) of DIS as a result of the continual integration of user (interactive) queries, is shown to provide this ability for open-ended construction of knowledge categories. In practice, the kind of automatic, adaptive, self-organization of knowledge implemented by TalkMine dramatically changes the management of information in DIS from a centralized to a distributed new vision. Instead of the current reliance on central databases (in Intranets) and Search Engines (on the Internet), any portion of a DIS managed by TalkMine is capable of pointing users to relevant information in its neighborhood. In other words, the TalkMine managed DIS is itself a distributed database or search-engine which is continually adapting to and learning from its users and their patterns of information usage. Each node of the network learns to relate its information to other nodes from repeated interaction with users. With this permanent feedback loop between user and network, the structure of the DIS reflects the knowledge of its community of users. DIS with TalkMine function both as collaborative information networks and as ever-present, distributed, search engines.
Copyright © Genetic Computer School, 2002
Page189
MANAGEMENT INFORMATION SYSTEMS First Edition Work Book
Copyright © Genetic Computer School, Singapore. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, without the permission of Genetic Computer School. First Edition 2002 Genetic Computer School. 257 Selegie Road, #03-287B, Selegie Complex, Singapore 188350.
Copyright © Genetic Computer School, 2002
Page190
Chapter 1 Question 1 a) Give an example of a professional job, and describe one or more information systems that would provide each of the five types of benefits listed below: i. Support for Strategic Planning ii. Support for Management Control iii. Support for Operation Control iv. Improved Product Quality v. Improved Product Delivery List the five components of an information system and describe four roles people play in information environment. Question 2 What are the main reasons behind information management? Question 3 What are the main characteristics of information? Question 4 Explain in detail what is an information system. Give examples. Question 5 a) Define CBIS. Explain why information has to be managed.
Copyright © Genetic Computer School, 2002
Page191 b) Information is a valuable asset in most organisations. So managers have to be clearly focused on what information is important to them and what in not. Explain with proper example. c) What are the steps to be taken by managers to kept proper way valuable information for in most organisations? d) Information is one of the most important in very sector, either commercial or noncommercial. Explain how and which ways it will be provide to keep proper information. Question 6 Support for Strategic Planning Support for Management Control Support for Operational Control Improved Product Quality Improved Product Delivery What is the main purpose of computer-based information systems?
Copyright © Genetic Computer School, 2002
Page192
Chapter 2 Question 1 Write short notes on the following: a) Transaction processing b) Decision support system c) Office automation system d) Management Information e) Systems Exception Reporting f) Data Warehouses g) On-line Analytical Processing Question 2 Describe the five categories into which the applications of Management Information Systems fall. Question 3 Describe the applications into which Management Information Systems fall. Question 4 What is business network design? What are the four types of network redesign? Question 5 What are the main uses of Transaction Processing Systems?
Copyright © Genetic Computer School, 2002
Page193 Question 6 Explain how on-line analytical processing tools assist in maintaining data archives? Question 7 a) Discuss the features of Transaction Processing Systems, which are important in creating systems and solutions. b) Explain the difference(s) in building Executive Support Systems and Traditional Management Information Systems?
Copyright © Genetic Computer School, 2002
Page194
Chapter 3 Question l Distinguish between homogeneous and heterogeneous workgroups. Question 2 Describe all the phases of the system development diagrams. Question 3 Explain the terms a) Analytical support b) Communication support c) Tracking and monitoring support in the context of personal information systems. Question 4 What do you understand by the term organisation? Outline the three that organisations share in common. Question 5 Differentiate between personal and workgroup information systems. Question 6 With the help of a diagram, illustrate the system development process. Question 7 What is the system development life cycle? Briefly explain the different stages of the development life cycle.
Copyright © Genetic Computer School, 2002
Page195 Question 8 Write short notes on the following: a) Alternative Quality Process b) System c) Functional Specification
Copyright © Genetic Computer School, 2002
Page196
Chapter 4 Question 1 With respect to the evolution of IT applications, explain what do you understand by the automation of work. Question 2 What are the possible benefits arising from information management? Illustrate your answer with examples. Question 3 Define the term business transformation. What is its impact on the enterprise? Question 4 Briefly describe the stages through which IT has evolved. Question 5 What has been the impact of the emergence of Internet and virtual banking on the financial industry? Question 6 Explain why management thinking has failed to understand the implications of the evolving role of IT in business system. Question 7 Current management practice fails to adequately address the impact and resource implications of four critical dimensions of complexity. Explain why.
Copyright © Genetic Computer School, 2002
Page197 Question 8 a) Organisations have had to make changes in their provision of Information Service to support End User Computing (EUC). i. Describe one way in which support of EUC has changed the structure of an organisation's information service provision. ii. Briefly describe the general changes in the nature of work undertaken by information services personnel that you would expect with the introduction of EUC into an organisation. b) A Data Centre may be centralized or de-centralized. Describe the issues you would consider when deciding between a centralized Data Centre provision and a decentralized Data Centre provision.
Copyright © Genetic Computer School, 2002
Page198
Chapter 5 Question 1 Differentiate between the terms management and leadership. What are the main functions of management? Question 2 The main managerial roles consist of interpersonal, informational and decisional roles respectively. Under each of the above categories of roles define the roles that a manager plays thus explaining the function of each sub-role in turn. Question 3 With the help of a diagram explain the managerial structure. Explain the role of the manager at each level. Question 4 How is the decision-making process handled at each managerial level. Question 5 Different employees have different needs. The most effective managers know this and have learned how to adapt their style based on the individual's needs. In this respect, describe the following terms: a) Surviving perspective b) Learning perspective c) Competing perspective d) Relating perspective e) Teaching perspective
Copyright © Genetic Computer School, 2002
Page199 Question 6 a) What four common functions do all managers perform? Briefly describe them. b) What are the similarities and differences between managers at different organisational levels? Question 7 a) Define the team of `data and information'. To having good information what are the steps to be taken, list and explain briefly. b)
List and describe the Characteristics of good information.
Copyright © Genetic Computer School, 2002
Page200
Chapter 6 Question 1 Write short notes on the following: a) Knowledge representation b) Artificial Intelligence c) Knowledge Engineering d) Knowledge Base e) Expert Systems Question 2 a) List and briefly describe the disciplines that constitute the subject of artificial intelligence (AI). b) List the three ways in which knowledge is represented in expert systems. c) Do most expert systems in use today replicate the abilities of a human expert? What do they do? d) Explain the problems with the term artificial intelligence. Question 3 What are the main components of human intelligence? Question 4 Discuss whether Artificial Intelligence is possible.
Copyright © Genetic Computer School, 2002
Page201 Question 5 Define the term meta knowledge and what is its importance in the field of artificial intelligence. Question 6 Knowledge must be represented efficiently, and in a meaningful way. Discuss this statement. Question 7 What are the components of expert reasoning?
Copyright © Genetic Computer School, 2002
Page202
Chapter 7 Question 1 What does framework refer to? Question 2 Describe the term strategic planning? Question 3 What are the components of an IS strategy? Question 4 What are the contents of the IS strategic plan? Question 5 Write short notes on the following: a) Product differentiation b) Competitive scope c) Differentiation bases Question 6 What does the term ‘hybrid manager' refer to? Question 7 Describe the strategic planning process. Question 8 Discuss how has the role of Information systems changed over the years? Question 9 In Porter's view, the performance of individual corporations is determined by the extent to which they cope with, and manipulate, the five key 'forces' which make up the industry structure. What are those five key forces? Question 10 Discuss how differentiation can help enterprises achieve above average performance.
Copyright © Genetic Computer School, 2002
Page203
Chapter 8 Question 1 What are the criteria required to enhance the effectiveness of workgroups? Question 2 What are the main drawbacks of workflow market structure? Question 3 What are the key functions that the workflow engine must support to completely support application requirements? Question 4 What are the advantages that Internet provides over the workflow computing infrastructure? Question 5 Explain how a workgroup is connected to horizontal and vertical organizations. Give an example of a workgroup, one horizontal and one vertical organization connected to that workgroup, and one information flow for each connection. Question 6 What are the advantages of workgroup computing? Question 7 What are the main limitations of transaction-based workflow systems? Question 8 Distinguish between transaction-based, collaborative and adhoc workflow systems? Question 9 Explain with the aid of examples what is rapid application generation. Question 10 Explain the concept of environmental independence with respect to workflow systems.
Copyright © Genetic Computer School, 2002
Page204
Chapter 9 Question 1 How does a workgroup support system relate to a groupware? Question 2 Distinguish between asynchronous groupware and realtime groupware. Illustrate your answer with examples of each. Question 3 What are the functions that groupware supports and the give some examples of software that facilitate that support. Question 4 Write short notes on the following: a) Group scheduling software b) Electronic meeting software c) Videoconferencing software d) Whiteboard software e) Workflow Automatic Software. Illustrate your answers with examples. Question 5 Explain the term groupware and why is it being widely used? Question 6 What does the term CSCW (Computer-Supported Cooperative Work) refer to? Copyright © Genetic Computer School, 2002
Page205 Question 7 Why is user testing considered more difficult in groupware systems than in multi-user systems? Question 8 What are the applications of groupware? Question 9 What are the main factors that need to be taken into consideration while designing a groupware?
Copyright © Genetic Computer School, 2002
Page206
Chapter 10 Question 1 Computation devices encode data using basically two types of codes: external and internal code. Distinguish between these two types of codes. Question 2 Write short notes on the following: a) Data processing b) Database c) Database Management System Question 3 In what ways is personal strategic planning beneficial? Question 4 Why do professionals resist planning? Question 5 Describe four categories of information systems personnel. What are the functions of each? Which of these categories are combined in a personal information system? Question 6 Distinguish between data and information. Question 7 a) What tasks are involved in designing the people component of a personal information system? b) What is parallel installation? c) What is pilot installation? Question 8 Summarize how the systems development effort changes when developing information systems at the personal and workgroup levels.
Copyright © Genetic Computer School, 2002
Page207
Chapter 11 Question 1 What is an Executive Information System? What are the features of such an Information System? Question 2 Describe the components of an executive Information System? Question 3 Describe the four decision styles to which an EIS needs to be individually tailored to. Question 4 Explain the difference(s) in building Executive Support Systems and Traditional Management Information Systems? Question 5 How useful are Executive Support Systems? Question 6 What are the main issues that an Executive Support System need to consider? Question 7 Explain the evolution of Executive Information Systems. Question 8 An EIS itself does not have a clear-cut goal as do most conventional computer based information systems. Discuss this statement. Question 9 What is the role that psychological profiling plays in Executive Information Systems?
Copyright © Genetic Computer School, 2002
Page208
Chapter 12 Question 1 List some of the features of a decision support system. Question 2 List some Group Decision Support Tools. Question 3 Discuss the advantages and disadvantages of Group Decision Support Systems. Question 4 a) Explain the basic structure and process of the model management component of a DSS program. b) Explain the basic structure and process of the data management component of a DSS program. c) Explain the relationship of the data interface program in the DSS and extraction programs on an organizational computer. Question 5 How can a DSS be developed in-house using commercial program components? Question 6 a) What is a decision support system (DSS)? How does it differ from a management information system (MIS)? b) In what ways is building decision support systems different from building traditional MIS systems?
Copyright © Genetic Computer School, 2002
Page209 c) Explain the two ways in which the term management information system is used. Give the broad definition of MIS. Give the narrow definition of MIS. d) Define the following terms: i) Management information system and Decision support system. Question 7 What is the main purpose of Decision Support Systems? uestion 8 Describe the three basic components of a Decision Support System.
Copyright © Genetic Computer School, 2002
Page210
Chapter 13 Question 1 What are the components of expert systems? Question 2 Distinguish between forward chaining and backward chaining. Question 3 What are the main guidelines that need to be considered to determine whether a problem is suitable for an expert solution? Question 4 What are the five stages in the development of an expert system? In which two ways does this process differ from the process used to develop other personal, workgroup, and organisational information systems? Question 5 a) List and briefly describe the disciplines that constitute the subject of artificial intelligence (AI). b) List the three ways in which knowledge is represented in expert systems. c) Do most expert systems in use today replicate the abilities of a human expert? What do they do? d) Explain the problems in defining the term Artificial Intelligence. Question 6 What does the term deductive inference refer to? What are its advantages?
Copyright © Genetic Computer School, 2002
Page211 Question 7 What are the limitations of expert systems? Question 8 Distinguish between rule-based systems and expert systems. Question 9 What are the necessary criteria for expert systems development?
Copyright © Genetic Computer School, 2002
Page212
Chapter 14 Question 1 Write short notes on the following: a. Automatic Data Processing System b. Distributed Information System c. Knowledge Mining Question 2 How does TalkMine influence the management of information in Distributed Information Systems? Question 3 What are the major categories of ADAPS programs? Outline the purpose of each type of program. Question 4 Why do files need to be maintained? How are they maintained? Question 5 Why is a transaction-oriented system considered more secure than a data processing system? Question 6 How does Talkmine ensure the evolution of a distributed Information System? Question 7 How is security maintained in Automatic Data Processing Systems? Question 8 What are the main objectives of Distributed Information Systems?
Copyright © Genetic Computer School, 2002
Page213
Copyright © Genetic Computer School, 2002
Page214 REFERENCES - Management Information Systems ActiveBook, 7/e Kenneth C. Laudon, New York University Jane P. Laudon, Azimuth Information Systems, Prentice Hall, 2002 - Essentials of Management Information Systems, 4/e Jane P. Laudon, Azimuth Information Systems Kenneth C. Laudon, New York University, Prentice Hall, 2001 - Developing Management Skills, 5/e David A. Whetten, Brigham Young University Kim S. Cameron, University of Michigan, Prentice Hall, 2002 - Management, 7/e Stephen P. Robbins, San Diego State University Mary Coulter, Southwest Missouri State University, Prentice Hall, 2002 - Essence of Information Systems, 2/e Chris Edwards, Cranfield School of Mgt., Cranfield, UK John Ward, Cranfield School of Mgt., Cranfield, UK' Andy Byetheway, Cranfield School of Mgt., Cranfield, UK, Prentice Hall PTR,1998 - Managing Technological Change: Strategies For College And University Leaders By: Bates, Anthony W.; Bates, Tony; Other Bates, Published: November 1999 Jossey-Bass -
Information
Systems
Analysis
and
Modeling:
An
Information
MacRodynamics Approach (Kluwer International Series in Engineering and Computer Science, 2"d Edition by Vladimir S. Lerner Hardcover (October 1999), Kluwer Academic Publishers; ISBN: 0792386833 -
Information Modeling in the New Millennium
by Matti Rossi (Editor) Idea Group Publishing; ISBN: 1878289772, Hardcover - 528 pages (April 2001) - Workplace Studies : Recovering Work Practice and Informing System Design by Paul Luff (Editor), Jon Hindmarsh, Christian Heath Cambridge University Press, Hardcover (August 2000)
Copyright © Genetic Computer School, 2002
Page215
Copyright © Genetic Computer School, 2002