The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
IJCST 15,1
16
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Viscoelastic properties of threads before and after sewing Isao Ajiki
Received February 2001 Revised October 2002 Accepted October 2002
Faculty of Education, Yamagata University, Japan
Ron Postle Department of Textile Technology, University of New South Wales, Sydney, Australia
Keywords Sewing, Threads, Physical properties Abstract The viscoelastic properties of the sewing thread before and after loading in the sewing process were investigated. Sewing threads are subjected to dynamic tension and friction in the sewing process. In order to compare polyester, cotton and silk sewing threads, the fineness of the threads were selected to be almost equal. There are some differences between the stress extension curves of the parent thread and the sewn thread except for the polyester sewing thread. The phenomenon of inverse relaxation occurs for high levels of retraction. The stress-inverse relaxation index for the polyester sewing thread is larger than for other threads and the inverse relaxation for silk thread is small. From the creep curves, the sewn threads show higher secondary creep and lower instantaneous recovery than the parent threads.
International Journal of Clothing Science and Technology Vol. 15 No. 1, 2003 pp. 16-27 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310461132
Introduction Sewing threads are subjected to dynamic tension and friction in the sewing process. It is important for clothing production to know the effect of the viscoelastic properties of a sewing thread and the dynamic tension in the stitch formation process. Vangheluwe (1992a, b, 1993) showed that warp yarns were subjected to dynamic loading during weaving before stress relaxation during a loom stop. For a relaxation test starting at a low force level, there is an inverse relaxation after a large number of load cycles. A small number of load cycles produced mixed relaxation. Manich and Castellar (1992) studied the recovery and inverse relaxation phenomena of polyester staple fibre rotor spun yarns. Better fibre orientation resulted in higher permanent deformation and lower delayed elastic recovery of the yarn. The inverse relaxation effect increased with the level of orientation of the fibres in the yarn. Gersak and Knez (1991) showed the analysis of the thread loading in the sewing process. They also showed (Gersak, 1995, 1997) that the viscoelastic properties of sewing threads directly influence the dynamic loads of threads in
the stitch formation process. The results of research work on viscoelastic properties of threads during dynamic loading showed a very similar behaviour of two sewing threads having similar tension and extension at the flow point. Sundaresan et al. (1997, 1998) discussed the mechanism of strength reduction of sewing threads during high speed sewing. The structural openness of the thread, namely, the pull-out of fibres and the displacement of the plies at the thread interlocking point in the stitch, were found to be the dominant factor influencing the strength reduction. Cotton threads exhibit higher strength loss than polyester threads. More recently, Webster and Laing (1998) investigated the effects of repeated extension and recovery on selected physical properties of lockstitch seams. When the yarns and threads are subjected to dynamic loading (or cyclic loading) during weaving or sewing, the mechanical properties depend on the previous load history. Further research is needed on the viscoelastic properties of threads and seams in response to the dynamic loads during the use of garments. It is important for clothing production to know the viscoelastic properties of a sewing thread and the dynamic tension in a stitch formation process. The present aims are to investigate the viscoelastic properties of threads and their dynamic loads in the sewing process and to compare the sewing thread before and after loading.
Viscoelastic properties of threads 17
Experimental method Sewing threads A commercial cotton spun sewing thread, a polyester spun sewing thread and a commercial silk filament sewing thread were selected for the present study and showed with details as given in Table I. In order to compare the three sewing threads, the thread fineness was selected to be almost equal. The tension before and after passing through the eye of the sewing needle (#11) ( Japanese Industrial Standard, 1987) was measured using the apparatus (Ajiki, 1989). It was shown that three threads could not distinguish the tensions at the eye of the needle. The domestic lockstitch machine was used for these trials. It was run at 1,000 r.p.m. with JUKI needle (#11) ( Japanese Industrial Standard, 1987) and stitch density 5 stitches/cm. The seams were made with two layers of fabric,W-W-9 (Table II). The dynamic tension was adjusted to obtain a balanced stitch and the seam balance ratio (Ajiki and Iwakami, 1981; Mahar et al., 1989) Code SP-80-1 SC-80-1 FS-50-1
100 per cent polyester spun 100 per cent cotton spun 100 per cent silk filament
Fineness (tex)
Twist ply/single (per m)
22.2 (7.4 £ 3) 22.2 (7.4 £ 3) 21.0 (7.0 £ 3)
1,020/1,568 (S) Z/1,100/1,380 (Z/S) 590/720
Table I. The three sewing threads tested
IJCST 15,1
18
was measured in all cases. The bobbin thread tension was 20 cN for all sewing threads. In order to investigate the change in the physical properties of the sewn threads, the needle thread was carefully taken out of the seam and then tested on each apparatus. Load-elongation The tensile properties of the thread until breaking-point were measured using the Instron constant-rate-of-extension instrument. Figure 1 shows stress-extension curves for the three kinds of thread with the parent thread and the sewn thread. The needle thread tension was almost as high as 2 N (about 0.09 N/Tex) for the three threads during the sewing process (Ajiki, 1999). This region is shown in Figure 1 by diagonal lines. The parent threads and the sewn threads were experimented over each 100 samples. Table III shows the properties of the parent thread with their CV percentage. There are some differences between the stress extension curves of the parent thread and the sewn threads except for polyester SP-80-1. The sewn thread of cotton SC-80-1 is much more extensible than the parent thread. Stress-relaxation and stress-inverse relaxation When the sewing thread is held stretched during extending, the stress in the sewing thread gradually decays. It may drop to a limiting value or it may disappear completely. This phenomenon is known as relaxation. When the sewing thread is held retracted during stress recovery, the stress in the sewing thread may increase with time under some conditions. This phenomenon is known as inverse relaxation. Figure 2 shows a typical curve demonstrating the load in a retracted specimen recorded on a time scale. The first part represents the rise in load resulting from the initial tension of the sewing thread to a predetermined load (1.98 N). The crosshead of the Instron is stopped at A (load W1), followed by relaxation in time. This is a typical relaxation phenomenon. At point A, the strain e1 is calculated as: e1 ¼ vt1 at a constant speed v for a time t1. At A (load W1) the crosshead of Instron is reversed and the specimen is allowed
Symbol (fibre, fabric) W-W-9 (wool, flannel)
Table II. Fabric data
LC
WC J/m2
RC per cent
T mm
W g/m2
K kPa
0.472 0.54 55.8 1.56 271 0.455 1.05 57.3 3.07 719 Notes: The upper section in the cell: one layer, the lower section in the cell: two layers, LC: linearity of compression-thickness curve, WC: energy in compressing fabric under 4.9 kPa, RC: compressional resilience, T: fabric thickness at 49 Pa pressure, K: compressional elasticity.
Viscoelastic properties of threads 19
Figure 1. Typical stress-extension curves for the parent and sewn thread for the three kinds of thread
to retract up to the point B where the tension reaches a low value W2. At this instant of time t2, the crosshead is stopped. The load now starts rising along BC to level off to a value W3. The zone AB corresponds to the second stage of the test during which the thread undergoes a recovery of e2 ¼ v(t22t1) at the same speed v from t1 to t2. In this work, the inverse relaxation index (R) was calculated for the time after 300 s from point B. For practical purposes, the load W1 was 1.98 N and the load W2 was 1.49, 0.99, 0.5 and 0.25 N. The tensile and reverse speed (v) was 5 mm/min.
IJCST 15,1
20
Description of the creep apparatus A schematic diagram of the principal components of the creep apparatus is given in Figure 3. The apparatus consists essentially of three elements arranged in vertical alignment: an adjustment device at the top, a linear variable differential transformer at the bottom and the thread specimen in the centre. Results and discussion Theory The mechanism of stress recovery can be explained in the following way. A load applied to a viscoelastic material produces immediate elastic and delayed extensions; the former is generally ascribed to the extension of lateral molecular bonds and the latter to molecular slippage resulting from the breakage of some of these bonds. The slippage is time-dependent and goes on until the stress distribution becomes uniform throughout the specimen. When a stretched specimen is allowed to retract, this process is reversed, because Breaking stress (N/Tex) Code
Table III. Properties of the parent thread
Figure 2. Typical curve of relaxation and inverse relaxation. (a) A typical relaxation, (b) a typical inverse relaxation curves and R : the inverse relaxation index (Manich and Castellar, 1992)
SP-80-1 SC-80-1 FS-50-1
Polyester Cotton Silk
Breaking elongation (per cent)
Toughness (cN/Tex)
Mean
CV (per cent)
Mean
CV (per cent)
Mean
CV (per cent)
0.365 0.253 0.503
10.2 6.3 2.4
22.3 8.4 27.5
5.7 7.9 5.5
3.53 0.98 9.14
15.3 10.3 8.4
Viscoelastic properties of threads 21
Figure 3. A schematic diagram of creep apparatus
the stress now tends to get reduced equally throughout the material (Nachane et al., 1982). Thus, during the retraction the molecular slippage occurs in the reverse direction. The same mechanism may do inter fibers slippage and extension of crimp on the thread. During the dynamic loading, there are viscoelastic effects, which will influence relaxation and creep behaviour. Figure 4 shows that the maximum of needle-thread-tension per stitch was about 1.8 N. The domestic lockstitch machine showed that the distance between the thread take-up lever and the needle eye is about 20 cm. The needle thread may undergo more than 60-80 time tension cycles by the take-up lever in the process of sewing because the length per stitch is 0.25-0.35 cm. Typical curves of the relaxation and inverse relaxation of the polyester thread (SP-80-1) are given in Figure 5 for the parent thread. Three kinds of relaxation curves (Vangheluwe, 1993) appear:
IJCST 15,1
22 Figure 4. Tension waveform of sewing needle thread during one revolution of arm shaft
Figure 5. Typical curve of the relaxation and the inverse relaxation for the polyester sewing thread SP-80-1
(1) ordinary relaxation at the low level of retraction; (2) inverse relaxation at the high level; and (3) mixed relaxation at the middle level (first there is an increase of force, then after some time, the load decreases). The load retraction percentage was calculated as follows (Figure 2): Load retraction ðper centÞ ¼
W1 2 W2 £ 100 W1
ð1Þ
The recovery was calculated as follows: Recovery ðper centÞ ¼
t2 2 t1 £ 100 t1
ð2Þ
The existence of inverse relaxation and of mixed relaxation seems to be due to viscoelastic effects during the previous dynamic loading of the thread.
Stress-relaxation and stress-inverse relaxation Table IV shows the inverse relaxation index and its standard deviation with the recovery at load 1.49, 0.99, 0.50 and 0.25 N. Mean and its standard deviation were calculated over 30 experiments. The stress-inverse relaxation index for the polyester thread SP-80-1 is larger than for the other threads. The inverse relaxation for silk thread FS-50-1 is small, but the difference of the index between the silk parent thread and the silk sewn thread is significant at all levels of load retraction (Table V). In the present work, the levels of recovery for the parent thread were up to 36.1 per cent for the polyester thread SP-80-1, up to 32.5 per cent for the cotton thread SC-80-1 and up to 51.3 per cent for the silk thread FS-50-1. The thread parent Load retraction (per cent)
Recov. (per cent)
Mean
Polyester SP-80-1 0 25 50 75 87.5
0 5.8 13.8 26.7 36.1
Cotton SC-80-1 0 25 50 75 87.5 Silk FS-50-1 0 25 50 75 87.5
Load retraction (per cent)
23
The thread sewn
R SD
Recov. (per cent)
Mean
SD
223.9 2 5.1 6.3 11.8 12.3
0.96 0.41 0.20 0.11 0.33
0 5.1 13.3 25.2 35.4
224.9 26.8 6.0 11.9 12.8
0.88 0.66 0.34 0.29 0.37
0 7.9 15.2 24.3 32.5
223.9 2 8.6 2.1 7.1 7.8
0.90 0.81 0.36 0.30 0.55
0 6.8 9.8 17.1 23.7
230.6 212.3 0.8 6.8 7.8
1.70 2.53 0.54 0.34 0.89
0 15.2 24.1 40.2 51.3
225.5 2 9.0 2 0.8 3.2 3.7
2.46 1.08 0.22 0.26 0.57
0 8.5 16.9 29.6 34.1
229.6 213.6 21.8 4.0 4.8
3.46 1.07 0.99 0.26 0.74
SP-80-1 (Polyester)
Viscoelastic properties of threads
SC-80-1 (Cotton)
R
Table IV. The inverse relaxation index (R) and its standard deviation
FS-50-1 (Silk)
0 O O O 25.0 O O O 50 £ O O 75 £ £ O 87.5 O £ O Notes: “O” sign indicates that at the significant level 0.05 the difference is significant, “ £ ” sign indicates that the difference is not significant at the same level.
Table V. Inverse relaxation
IJCST 15,1
The phenomenon of inverse relaxation occurred for the high levels of retraction. This is important for the seam structure in the sewing process. The sewing thread may recover after the seam formation in the sewing and may exhibit the inverse relaxation for the high retraction. Another tightening tension may cause on the seam structure after sewing.
24 Creep and creep recovery On the application of a load to a sewing thread, it will, after an instantaneous extension, continue to extend as time goes on; on removal of the load, the recovery will not be limited to the instantaneous recovery, but will continue to take place. This behaviour is shown in Figure 6 and is known as creep, and creep recovery (Morton and Hearle, 1962). Figure 6 indicates the typical curve for the application of a constant load (1.98 N) to a sewing thread for a given time (5 min) and then removing the load. The instantaneous extension was followed by creep. The removal of the load gave rise to an instantaneous recovery, followed by a further partial recovery with time and still left some unrecovered extension. The total extension may be divided into three parts: (1) the immediate elastic deformation, which is instantaneous and recoverable; (2) the primary creep, which is recoverable in time; and (3) the secondary creep, which is non-recoverable. Figure 7 shows the instantaneous load and the total creep, which were calculated on 40 experiments. Figure 8 shows the instantaneous recovery,
Figure 6. Creep under constant load and recovery under zero load, showing instantaneous extension, a-b; total creep, b-c; instantaneous recovery, d-e; primary creep, e-f; and secondary creep, g-h (Morton and Hearle, 1962)
Viscoelastic properties of threads 25
Figure 7. Parameters from creep curves under constant load (1.98 N) for the parent and sewn threads
Figure 8. Parameters from creep recovery curves under zero load for the parent and sewn threads
the primary creep and the secondary creep by comparing the parent thread and the sewn thread. From the creep curves, the parent thread showed higher instantaneous deformation and lower total creep than the sewn thread in tensile deformation. The silk sewing thread FS-50-1 showed less instantaneous deformation than the other two sewing threads. In recovery, the sewn threads showed higher secondary creep and lower instantaneous recovery than the parent threads. The recovery of the parent thread was higher than that of the sewn thread. Figure 9 shows the relationship between the instantaneous recovery and the recovery at the load retraction 87.5 per cent in the inverse relaxation
IJCST 15,1
26 Figure 9. The relationship between the instantaneous recovery in the creep and the recovery at the load retraction 87.5 per cent in the inverse relaxation experiment
experiment. It is reasonable to correlate between them because the recovery process in the inverse relaxation experiment are nearly instant recovery. We found that the sewing threads had viscoelastic behaviour due to dynamic loading in sewing process. The behaviour could be analysed by a viscoelastic model which would study the effect of inverse-relaxation on the sewing thread tension in the stitch structure. The sewing machine will desire to be low dynamic loading or no loading for sewing threads. Conclusion We found that the maximum needle-thread-tension per stitch was about 1.8 N and that the distance between the thread take-up lever and the needle eye was about 20 cm. The needle thread may have more than 60-80 times the tension applied by the take-up lever in the process of sewing because the length per stitch is 0.25-0.35 cm (Ajiki, 1999). The viscoelastic properties of the sewing thread before and after loading in the sewing process were investigated. In order to compare the different sewing threads, the fineness of the threads was selected to be almost equal. The results are as follows. (1) The phenomenon of inverse relaxation occurred for high levels of retraction. The stress-inverse relaxation index for the polyester thread SP-80-1 is larger than for the other threads and the inverse relaxation for silk thread FS-50-1 is small. (2) From the creep curves, the sewn threads show higher secondary creep and lower instantaneous recovery than the parent threads. The recovery of the parent thread is higher than that of the sewn thread.
(3) The results depend on the viscoelastic properties of threads and fibres, especially the relationship between the dynamic cyclic loading and the stress-inverse relaxation. The behaviour could be analysed by a viscoelastic model which would study the effect of inverse-relaxation on the sewing thread tension in the stitch structure. References Ajiki, I. (1989), “Needle thread tension and seam structure on lock-stitch sewing machine for home use”, Bulletin of the Yamagata University (Educational Science), Vol. 9, pp. 597-603. Ajiki, I. (1999), “Mechanical analysis of lockstitch seams”, Doctoral thesis, University of New South Wales. Ajiki, I. and Iwakami, N. (1981), “A study of stitch structure and thread length in lock-stitch seams”, Bulletin of the Yamagata University (Educational Science), Vol. 7, pp. 605-17. Gersak, J. (1995), “Rheological properties of threads – their influence on dynamic loads in the sewing process”, International Journal of Clothing Science and Technology, Vol. 7, pp. 71-80. Gersak, J. (1997), “Evalution of rheological properties of a thread using numerical methods”, International Journal of Clothing Science and Technology, Vol. 9, pp. 236-40. Gersak, J. and Knez, B. (1991), “Reduction in thread strength as a cause of loading in the sewing process”, International Journal of Clothing Science and Technology, Vol. 3, pp. 6-12. Japanese Industrial Standard ( JIS) B9012 (1987), Needles for Household Sewing Machines. Mahar, T.J., Ajiki, I. and Postle, R. (1989), “Fabric mechanical and physical properties relevant to clothing manufacture – part 2”, International Journal of Clothing Science and Technology, Vol. 1. Manich, A.N. and Castellar, M.D.E. (1992), “Elastic recovery and inverse relaxation of polyester staple fiber rotor spun yarns”, Textile Research Journal, Vol. 62, pp. 196-9. Morton, W.E. and Hearle, J.W.S. (1962), Physical Properties of Textile Fibres, Butterworths and The Textile Institute, Manchester and London. Nachane, R.P., Hussain, G.F.S. and Krishna Iyer, K.R. (1982), “Inverse relaxation stress recovery in cotton fibers and yarn”, Textile Research Journal, Vol. 52, pp. 483-4. Sundaresan, G., Salhotra, K.R. and Hari, P.K. (1997), “Strength reduction in sewing threads during high speed sewing in industrial lockstitch machine – Part 1. Mechanism of thread strength reduction”, International Journal of Clothing Science and Technology, Vol. 9, pp. 334-45. Sundaresan, G., Salhotra, K.R. and Hari, P.K. (1998), “Strength reduction in sewing threads during high speed sewing in industrial lockstitch machine – Part 2. Effect of thread and fabric properties”, International Journal of Clothing Science and Technology, Vol. 10, pp. 64-79. Vangheluwe, L. (1992a), “Stress relaxation and tensile modulus of polymeric fibers”, Textile Research Journal, Vol. 62, pp. 306-8. Vangheluwe, L. (1992b), “Influence of strain rate and yarn number on tensile test results”, Textile Research Journal, Vol. 62, pp. 586-9. Vangheluwe, L. (1993), “Relaxation and inverse relaxation of yarns after dynamic loading”, Textile Research Journal, Vol. 63, pp. 552-6. Webster, J. and Laing, R.M. (1998), “Effects of repeated extension and recovery on selected physical properties of ISO-301 stitched seams”, Textile Research Journal, Vol. 68, pp. 854-64.
Viscoelastic properties of threads 27
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
IJCST 15,1
28
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Further studies on fabric objective measurement (concentrated loading method) A. Alamdar-Yazdi
Received February 2002 Accepted September 2002
Department of Textile, The University of Yazd, Iran Keywords Woven fabrics, Mechanical properties Abstract This paper considers the ability of the concentrated loading method for measuring the basic low stress mechanical properties of heavy weight woven fabrics. Fifty-three randomly selected woven fabrics were tested and the results show the ability of the method. It also introduces the parameters, which indicate the behavior of the woven fabrics.
Introduction One of the major concerns of the textile and clothing industries is the low stress behavior of fabrics, while the cost and time consumption of the testing is also vital. Bishop (1996) gives comprehensive surveys about different methods of evaluation together with discussions. Recently a new system called “concentrated loading method” has been introduced, and the method was tested, using 36 cases of lightweight woven fabrics (Alamdar-Yazdi, 1998; Amirbayat and Alagha, 1995; Anderson, 1994; Bishop, 1996). This paper considers the ability of the concentrated loading method for measuring the basic mechanical properties of woven fabric for heavy weight woven fabrics. Fifty-three randomly selected woven fabrics were tested by two methods. First, the samples were tested by KES system (Stylios, 1991) and the results were based on actual values for the evaluation and analyses of the concentrated loading method. Second, the specimens were also tested by the loading-unloading for 200 g forces. The results show the ability of the method and introduce the parameters, which indicate the behavior of the woven fabrics.
International Journal of Clothing Science and Technology Vol. 15 No. 1, 2003 pp. 28-46 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310461141
Theoretical background Principles of the method are based on two behaviors of orthotropic sheets: the inter relations of in-plane properties; and buckling of flexible sheets under tension, when the external forces are not uniformly distributed (Amirbayat, 1991; Amirbayat and Alagha, 1995; Amirbayat and Bowman, 1991). The method has been explained and well described in few publications (AlamdarYazdi, 1998; Alamdar-Yazdi and Amirbayat, 2000).
Experimental work Studies on fabric Materials objective Fifty-three pieces of heavy weight woven fabrics (in the weight range of measurement 201-500 g/m2) were randomly selected. Fabrics included different structures, with wide range of materials, such as cotton, wool, jute, synthetics and blend of different fibers. Appendix 1 (Table AI) shows the specifications of the fabrics 29 used in this work including thickness and weight. Test specification Size of the specimen and length/width ratio The length and the width of the samples were 10 and 5 cm, respectively. The length/width ratio was taken to be two. This selection was due to the fact that the transverse stresses are compressive throughout the middle portions of the sample, when the length/width ratio is greater than one and when the ratio is equal or greater than two, the distribution of stress is close to the condition of the uniform loading i.e. s ¼ F=W , where s is stress, F is concentrated force inserted, and W is width of the specimen (Figure 1). Preparation of the samples Three rectangular specimens, each 24 cm long and 5 cm wide were cut from every sample fabric, one at an angle of 22.58 to the warp direction, (which is 67.58 to the weft direction), one at 458 to the warp (which makes the same angle to the weft direction), and one at 67.58 to the warp (which is 22.58 to the weft direction), using a special template. Since bias samples develop shear strain under tensile stress, the strips were folded in half to form a double ply of face-to-face fabrics of 12 cm long.
Figure 1.
IJCST 15,1
30
An eyelet was then punched 1 cm from the ply ends opposite to the fold. The second eyelet was inserted 10 cm far from the first one after any possible slack was removed. Preparation was done in the testing lab after 24 h of conditioning. Test procedure The samples were subjected to a single loading-unloading cycle at a rate of 10 mm/min with 200 g maximum force using a simple attachment to the jaws of tensile tester (Testometric-micro 350 made in UK by Shirly developments with 10 kg load cell). The loading-unloading curves obtained by the new method are shown as a sample in Figure 2. Features extracted from the tests under concentrated load (load-extension curves) Thirty-five quantities (called Features), which are categorized in four groups (strains, stresses, area, and slopes), were measured on each chart (Appendix 2, Table AII). Statistical analysis helped to find the correlation coefficients between the measured mechanical parameters (tested by KES method) and the characteristic factors ( features) obtained by the concentrated loading method (Appendix 3). Effective features, resulting in each mechanical parameters of the KES, were selected and utilized for multiple linear regression. (All correlation between the estimated and measured values of the properties are significant at 0.1 percent level.)
Figure 2.
Estimation of properties from measured qualities Studies on fabric Figures 3-11 give plots of actual shearing tensile and bending properties objective measured by KES against the estimated values from the present set of measurement tests. Charts belonging to all different fabric properties contain 106 points, which is double the actual number of the fabrics tested. This is done by: 31 . considering each fabric in its conventional sense and relating its warp properties to the parameters along 22.58, 458, and 67.58 from the warp direction, and . considering each fabric rotated through 908 and relating its weft properties to the parameters along 67.58, 458, and 22.58 from direction of the warp. As a result each property along a principal direction is related to the parameters 22.58, 458, and 67.58 from its direction. Discussions and conclusions Shearing properties For all the fabrics the curve corresponding to 458 shows the highest final extension (EML). This shows that ratio between Poisson’s ratio and tensile rigidity is maximum at 458 and is in agreement with the work of Alsawaf (1985) and the explanation of Amirbayat and Alagha (1995) (Figure 12). The measured shearing parameters (G-2HG.5-2HG5) showed a very high correlation with the estimated factors showing that the method is able to evaluate the shearing properties perfectly. Comparison between three sets of correlation results, shows that calculated features along the 458 bias sample (in contrast to the other two directions) had highest correlation with the shearing parameters, specially shearing modulus, proving the fact that shearing modulus can be estimated from tensile properties along 458 bias. The correlation coefficients between features and the KES shearing parameters (Appendix 3, Table AIII) show that post buckling slope (PBS) had the highest value of correlation with the shearing properties, specially shearing modulus. It should also be noticed that highest correlation among the three sets (PBS22.58, PBS458 and PBS67.58) belongs to 458 bias samples indicating results similar to previous works (Alamdar-Yazdi, 1998; Alamdar-Yazdi and Amirbayat, 2000; Amirbayat and Alagha, 1995). Pick slope is another important part of the curve related to shearing parameters specially hysteresis of shear force at 58 due to its high correlation. It shows that for heavy weight fabrics, especially those with tight structure, the shearing movement continued during the whole process of loading. The table of the correlation also indicates the importance of the ending slope and its relation to 2HG.5.
IJCST 15,1
32
Figure 3. The plot of measured shearing modulus versus the estimated values
Studies on fabric objective measurement 33
Figure 4. The plot of measured shearing hysteresis at 58 versus the estimated values
IJCST 15,1
34
Figure 5. The plot of measured shearing hysteresis at 0.58 versus the estimated values
Figure 6. The plot of measured fabric extension versus the estimated values
Studies on fabric objective measurement 35
Figure 7. The plot of measured linearity of the tensile curve versus the estimated values
Figure 8. The plot of measured tensile resilience versus the estimated values
IJCST 15,1
36
Figure 9. The plot of measured tensile energy versus the estimated values
Figure 10. The plot of measured bending rigidity versus the estimated values
Studies on fabric objective measurement 37
Figure 11. The plot of measured bending hysteresis versus the estimated values
Figure 12.
The shearing regression models contain features extracted from the 458 bias samples indicating the ability of the method to evaluate shearing properties based on testing 458 bias sample only. Figures 3-5 show the estimated equations and the plot of actual values of each shearing parameter versus its estimated values. The gap area between the loading and unloading curves (AGA), which is the sum of the final yarn and fibers movements, is shown to be the most related feature as Figure 3(b) shows. This is obvious due to the fact that during shearing movement, forces to be overcome are yarn friction at intersections and in some cases yarn bending and twisting. Figure 13 shows the important zones related to the shearing properties.
IJCST 15,1
38
Figure 13. Features related to shearing behavior of the woven fabrics
Tensile properties Among the features measured slope up to the critical point (SUC), PBS, ending slope (ES), SCP, ALF, AGL, and ATH, showed highest correlation with tensile properties (Appendix 3, Table AV). This relation highlights the important zones of the curve related to tensile properties (Figure 14). Initial part of the curve is due to yarn compression and extension (causing yarn straightening
Figure 14. Features and zones related to tensile properties of the woven fabrics
inside the plain of fabrics) and the top part is due to fiber movement inside Studies on fabric the yarns. Figures 6-9 show the estimated equations and the plot of actual objective values of each KES parameter versus its estimated values. measurement Bending properties Initial slope (IS), SUC, PBS and ES are the features which gave highest correlation to bending properties (Appendix 3, Table AIV). Figure 15 shows the important zones of the curve related to bending properties. Note that the forces to be overcome during the bending deformation are intersection contact force and bending the yarn (friction between fibers in the yarn). These forces are related to shearing properties as well. The selected features i.e. IS, SUC, PBS and ES, are in one hand highly correlated to shearing properties and on the other hand related to bending properties, indicating the above-mentioned fact (Table I).
39
Figure 15. Features and zones related to bending properties of the woven fabrics
IS SUC PBS ES
G
Shearing parameters 2HG.58
2HG58
0.738 0.805 0.896 0.775
0.675 0.827 0.845 0.872
0.751 0.896 0.906 0.854
Bending parameters B 2HB 0.589 0.553 0.512 0.475
0.589 0.573 0.563 0.525
Table I. The correlation between KES shearing and bending properties with some features extracted from the curves
IJCST 15,1
40
Correlation between KES bending parameters and the features showed importance of the slopes. It also lightened up the weight of the initial and buckling zones of the loading curve plus the significance of the ending part of the unloading curve. Moreover, regression analyses revealed the importance of the buckling zone by selecting the PBS as the most effective feature among 35 measured features. Having a good correlation between the bending parameters and IS means that for tight structure fabrics bending movement starts from the beginning of the loading. Figures 10 and 11 show the plots of the measured bending parameters versus the estimated values. References Alamdar-Yazdi, A. (1998), “A new method of evaluation of the low stress mechanical properties of woven fabrics”, Doctorate thesis, Department of Textile, University of Manchester, Institute of Science and Technology. Alamdar-Yazdi, A. and Amirbayat, J. (2000), “Evaluation of the basic stress mechanical properties (bending, shearing and tensile)”, International Journal of Clothing Science and Technology, Vol. 12 No. 5, p. 303. Alsawaf, F. (1985), “Area change as a measure of fabric performance”, Doctorate thesis, Department of Textile, University of Manchester, Institute of Science and Technology. Amirbayat, J. (1991), “The buckling of flexible sheets under tension”. Part I, Journal of Textile Institute, Vol. 82 No. 1, p. 61. Amirbayat, J. and Alagha, M.J. (1995), “A new approach to fabric assessment”, International Journal of Clothing Science and Technology, Vol. 7 No. 1, p. 46. Amirbayat, J. and Bowman, S. (1991), “The buckling of flexible sheets under tension. Part II. Experimental studies”, Journal of Textile Institute, Vol. 82 No. 1, p. 71. Anderson, V.J. (1994), “An alternative method of evaluation of fabric properties”, Department of Textile, University of Manchester, Institute of Science and Technology. Bishop, D.P. (1996), “Fabrics sensory and mechanical properties”, Textile Progress, Vol. 26 No. 3. Stylios, G. (1991), Textile Objective Measurement and Automation, Ellis Horwood, Chichester.
Appendix 1 F. no
Ends/ cm
Warp count Nm
Warp cri. percent
Picks/ cm
Weft count Nm
Weft cri. percent
Fabric structu.
Materials
Fabric thick mm
Fab.w g/m2
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53
20.0 23.0 28.0 28.0 25.0 34.0 33.8 28.0 28.0 28.0 22.0 30.0 40.5 28.0 31.0 15.0 30.0 39.0 56.0 26.0 28.0 31.0 28.0 03.33 28.0 27.5 33.4 37.4 28.0 33.0 32.0 33.5 35.2 46.0 12.5 12.5 12.5 45.0 19.6 14.0 1.25 12.5 14.0 18.0 30.3 18.0 28.7 33.5 44.0 20.0 07.5 20.0 29.0
44/2 20.0 50/2 48/2 24.0 60/2 38.0 48/2 48/2 48/2 19.0 48/2 64/2 46/2 48/2 24/2 48/2 34.0 50.0 46/2 22.0 48/2 44/2 02.7 44/2 17.0 22.0 034 20/2 22.0 23.5 22.0 25.0 33.0 14.0 14.0 14.0 26.8 12.0 11.0 14.0 14.0 09.5 11.2 32.0 10.0 25.5 40/2 40/2 07/2 03.5 07.9 09.5
20.0 03.0 04.0 08.0 10.0 05.0 09.0 08.0 08.0 07.0 04.0 04.0 06.0 05.0 05.0 08.0 06.0 03.0 07.0 10.0 06.0 04.0 06.0 0.5 07.0 03.0 13.0 02.0 08.0 10.0 08.0 14.0 06.0 07.5 20.0 15.0 15.0 10.0 06.0 04.0 20.0 20.0 05.0 04.0 11.0 06.0 12.0 10.0 07.0 06.0 08.0 05.0 11.0
22.0 16.5 21.0 18.0 22.3 28.0 22.0 20.0 22.0 22.0 19.5 22.0 28.3 22.0 23.0 12.0 24.0 24.0 19.3 26.0 21.0 26.0 21.0 3.7 25.0 23.0 21.2 16.5 23.0 22.0 28.0 20.5 15.0 24.0 10.5 10.5 10.5 16.0 14.0 13.0 10.5 10.5 14.0 16.0 14.2 14.0 21.6 32.0 30.0 10.0 07.0 25.0 18.0
52/2 20.0 50/2 48/2 24.0 30.0 17.0 46/2 48/2 46/2 19.0 48/2 30.0 44/2 48/2 24/2 46/2 20.0 17.0 23.2 20.0 48/2 20.0 02.7 44/2 26.0 24.0 10.5 24.0 22.0 23.5 22.0 12.0 19.5 06.0 06.0 06.0 12.3 11.0 8.25 06.0 06.0 09.5 11.2 18.0 10.0 17.0 40/2 20.0 09/2 3.25 15.2 09.0
08 06 06 10 08 08 10 10 08 08 07 08 08 07 07 09 08 05 10 12 10 08 09 01 08 11 11 03 12 09 08 09 03 06 12 10 10 05 06 06 10 10 07 06 14 04 11 12 09 11 10 10 08
Fancy Plain Twill Plain Twill Twill Twill Twill Twill Twill Plain Twill Twill Twill Twill Plain Twill Twill Plain Twill Twill Twill Twill Plain Twill Twill Twill Plain Fancy Twill Twill Twill Fancy Twill Plain 1/1 Bask 2/2 Twill 2/2 Twill Fancy Fancy Twill 3/1 Twill 1/3 Plain Fancy Velvet Fancy Velvet Twill Twill Plain Twill Plain Twill
P&V P P&C P&W P&W P&W C P&W P&W P&W C&P P&W P&W P&W P&W P&W P&W C C P&W P&W P&W P&W JUTE P&W P&W P&C AC P&W P&C P&W P&C AC C C & AC C & AC C & AC C&P AC AC C & AC C & AC ACET AC V&C AC V&C P&W P&W C P&W C C
0.300 0.340 0.435 0.460 0.535 0.425 0.440 0.440 0.480 0.460 0.420 0.460 0.545 0.510 0.410 0.500 0.510 0.580 0.560 0.800 0.464 0.450 0.466 1.300 0.460 0.580 0.460 0.620 0.470 0.470 0.430 0.470 0.645 0.460 0.960 1.050 1.070 0.630 0.950 0.930 1.100 1.100 0.560 0.670 0.745 0.740 0.750 0.750 1.150 0.950 1.500 1.070 0.930
201 207 208 209 215 222 226 228 228 229 230 231 235 235 239 244 245 246 250 250 251 251 252 258 259 262 270 270 272 274 275 275 278 280 294 295 295 298 300 300 300 300 310 320 330 338 340 365 400 406 440 450 500
Notes: AC¼Acrylic, ACET ¼ Acetate, C ¼ Cotton, J ¼ Jute, P ¼ Polyester, W ¼ Wool and V= viscose.
Studies on fabric objective measurement 41
Table AI. Specification of the fabrics used in this experimental work
IJCST 15,1
42
Appendix 2 Sym.
Parameters
Units
AAC ABC AF AGA AGL AH AHF ALF ATH AUD CPD EBC EBCC ECD
Area under the loading curve, above the critical point Area under the loading curve, below critical point Area under the loading curve, for 50 g load Gap area between two curves, above the critical point Gap area between two curves, top 50 g load Area under the loading curve, for 100 g load Area under the loading curve, for 150 g load Area under the loading curve, final 50 g load Area under the loading curve, for 200 g load Area under the unloading curve Strain at critical point on the unloading curve Difference in strain between max. load and end point Strain between two curves at the critical point Difference in strain between max. load point and the point with the same load as the critical point on the unloading curve Strain at critical point Strain at end point Strain at 50 g load Strain at 100 g load Strain at 150 g load Strain at the location of MG on the unloading curve Strain at the location of MG on the loading curve Strain at maximum load (200 g) Ending slope (final 20 g of the unloading curve) Difference in strain between the critical point on unloading curve and end point Area under the unloading curve, above critical point Area under the unloading curve, below the critical point Load at which maximum gap occurs Maximum distance (strain) between two curves Post buckling slope (20 g load after buckling point) Peak. slope (first 50 g of the unloading curve) Peak slope for final 50 g loading Load at critical point Initial slope (first 10 g loading) of the loading curve Slope up to critical point Gap area between two curves
gf. gf. gf. gf. gf. gf. gf. gf. gf. gf. percent percent percent
ECP EE EF EH EHF EMGD EMGL EML ES ET
Table AII. Features extracted from the curves
LAAC LABC LAMG MG PBS PSD PSL SCP SI SUC TGA
percent percent percent percent percent percent percent percent percent gf. percent gf. gf. gf. percent gf. gf. gf. gf. gf. gf. gf.
Appendix 3 Shearing modulus Par.
G1
G2
Shearing hysteresis at 0.58 G3
AAC
20.555
20.644
20.555
ABC
0.042
0.191
0.042
2HG.51
Shearing hysteresis at 58
2HG.52
2HG.53
2HG51
2HG52
2HG53
20.470 20.536
20.467
20.598 20.681
20.599
0.178
0.337
0.166
0.112
0.256
AF
20.423
20.469
20.423
20.406 20.462
20.403
20.489 20.534
20.487
20.535
20.630
20.535
20.460 20.542
20.456
20.577 20.665
20.575
AGL
20.542
20.614
20.542
20.384 20.423
20.379
20.545 20.610
20.545
AH
20.497
20.574
20.497
20.452 20.526
20.449
20.555 20.630
20.555
AHF
20.539
20.622
20.539
20.468 20.535
20.466
20.588 20.666
20.589
ALF
20.568
20.631
20.568
20.436 20.461
20.432
20.588 20.640
20.590
ATH
20.558
20.641
20.558
20.466 20.526
20.463
20.598 20.676
20.599
AUD
20.448
20.501
0.448
20.438 20.493
20.435
20.523 20.583
20.523
CPD
20.498
20.546
20.497
20.364 20.390
20.363
20.500 20.533
20.502
EBC
20.438
20.484
20.438
20.436 20.477
20.433
20.515 20.557
20.515
EBCC
20.494
20.557
20.494
20.372 20.416
20.369
20.502 20.552
20.501
ECD
20.390
20.442
20.390
20.400 20.450
20.397
20.467 20.519
20.466
+0.012
ECP
20.232
20.044
20.232
20.083
EE
20.413
20.446
20.413
20.286 20.311
20.284
20.407 20.427
20.406
EF
20.405
20.434
20.405
20.392
0.429
20.389
20.471 20.498
20.470
20.099 20.185
20.210
EH
20.466
20.519
20.466
20.433 20.489
20.430
20.527 20.579
20.527
EHF
20.500
20.561
20.500
20.450 20.509
20.447
20.556 20.616
20.556
EMGD 20.538
20.616
20.538
20.435 20.496
20.433
20.568 20.634
20.568
EMGL 20.368
20.375
20.368
20.374 20.387
20.373
20.437 20.437
20.439
EML
20.586
20.519
20.455 20.516
20.453
20.571 20.636
20.571
20.519
ES
0.778
0.775
0.778
ET
20.431
20.360
20.431
LAAC
20.432
20.489
20.432
LABC
20.059
20.051
20.059
LAMG MG
0.377
0.488
0.377
20.502
20.579
20.502
0.731
0.817
0.854
20.378 20.283
20.380 20.473 20.374
20.478
20.431 20.493
20.428
0.872*
0.741
0.030
0.168
0.023
0.402
0.527
0.407
20.380 20.436
20.377
0.820
20.508 20.572
20.508
20.030
0.087
20.042
0.401
0.480
0.417
20.513 20.578
20.512
PBS
0.803
0.896*
0.803
0.694
0.845*
0.726
0.798
0.906*
0.878
PSD
0.664
0.648
0.664
0.585
0.644
0.603
0.703
0.724
0.765
PSL
0.777
0.833*
0.777
0.605
0.678
0.629
0.765
0.864*
0.841
SCP
0.466
0.603
0.466
0.566
0.690
0.544
0.564
0.643
0.539
SI
0.694
0.738
0.694
+0.468
0.675
0.590
0.547
0.751
0.774
SUC
0.805*
0.788
0.805*
+0.636
0.827*
0.722
0.733
0.823
0.896*
TGA
20.529
20.611
20.526
20.389 20.443
20.384
20.534 20.605
43
0.094
AGA
0.088
Studies on fabric objective measurement
20.532
Note: Bold values are used to show the highest correlation coefficient among the three sets of data and the sign * is used to show the three highest correlation values among all the features.
Table AIII. The correlation coefficients between features and the KES shearing parameters (106 cases of heavy weight fabrics)
IJCST 15,1
Bending rigidity Par.
44
Table AIV. The correlation coefficients between features and the KES bending parameters (106 cases of heavy fabrics)
AAC ABC AF AGA AGL AH AHF ALF ATH AUD CPD EBC EBCC ECD ECP EE EF EH EHF EMGD EMGL EML ES ET LAAC LABC LAMG MG P PBS PSL SCP SI SUC TGA
Bending hysteresis
B1
B2
B3
2HB1
2HB2
2HB3
20.196 0.053 20.132 20.141 20.192 20.142 20.170 20.236 20.195 20.294 0.004 20.305 0.005 20.273 20.004 0.082 20.150 20.149 20.163 0.100 20.300 20.178 0.439 20.297 20.279 20.079 0.071 20.013 0.357 0.393 0.401 0.260 0.266 0.348 20.079
20.232 0.150 20.188 20.154 20.191 20.207 20.219 20.227 20.228 20.313 2 0.011 20.339 2 0.016 20.298 0.027 0.112 20.187 20.201 20.211 0.138 20.339 20.219 0.457 2 0.292 20.298 2 0.034 0.082 20.042 0.326 0.512* 0.483 0.320 0.470 0.425 20.110
20.175 0.016 20.123 20.114 20.164 20.129 20.154 20.212 20.175 20.292 0.033 20.304 0.037 20.272 20.018 0.110 20.142 20.138 20.150 0.077 20.308 20.162 0.475 2 0.299 20.276 2 0.100 0.040 0.016 0.421 0.483 0.484 0.193 0.589* 0.553* 20.053
2 0.224 0.174 2 0.203 2 0.183 20.147 2 0.206 2 0.214 20.213 2 0.218 2 0.351 0.003 2 0.366 2 0.006 2 0.339 0.071 0.081 2 0.218 2 0.215 2 0.219 0.124 2 0.350 2 0.222 0.475 20.305 2 0.344 0.006 0.193 2 0.025 0.397 0.421 0.387 0.385 0.285 0.372 2 0.070
20.249 0.293 20.274 20.196 2 0.130 20.272 20.254 2 0.180 20.238 20.381 0.011 20.401 2 0.013 20.374 0.149 0.122 20.268 0.274 20.269 20.152 20.389 0.264 0.561 2 0.271 20.376 0.085 0.190 20.041 0.403 0.563* 0.470 0.444 0.528 0.508 20.079
2 0.189 0.138 2 0.200 2 0.146 2 0.104 2 0.185 2 0.185 2 0.173 2 0.184 2 0.346 0.046 2 0.363 0.038 2 0.336 0.076 0.125 2 0.216 2 0.201 2 0.198 2 0.089 2 0.359 2 0.198 0.525 2 0.306 2 0.336 2 0.025 0.110 0.018 0.468 0.500 0.472 0.301 0.589* 0.573* 2 0.022
Note: Bold values are used to show the highest correlation coefficient among the three sets of data and the sign * is used to show the three highest correlation values among all the features.
Par. SCP ABC AF AH AHF ATH AAC ALF LAAC LABC AUD TGA AGA AGL EMGL LAMG ET SI SUC PBS PSL P ES ECP EF EH EHF EML ECD EBC CPD EMGD EE EBCC MG
LT1
LT2
LT3
RT1
RT2
RT3
0.491 0.101 2 0.492 2 0.538 20.567 20.581 20.580 20.578 2 0.587 2 0.064 20.607 20.443 20.516 20.518 0.511 0.299 20.518 0.366 0.525 0.589 0.553 0.588 0.654 20.184 2 0.491 2 0.527 2 0.550 20.564 2 0.548 20.597 20.384 20.497 20.285 20.381 20.405
0.554 0.271 2 0.514 2 0.554 20.561 20.552 20.559 20.485 2 0.588 0.093 20.599 20.403 20.510 20.430 20.493 0.369 20.378 0.622 0.690* 0.673* 0.581 0.623 0.691* 0.055 2 0.494 2 0.535 2 0.551 20.557 2 0.554 20.589 20.319 20.475 20.195 20.341 20.376
0.461 0.126 20.442 20.463 20.475 20.474 20.476 20.447 20.499 20.007 20.512 20.346 20.442 20.391 20.456 0.326 20.417 0.507 0.598 0.608 0.562 0.572 0.602 20.077 20.440 20.459 20.471 20.475 20.484 20.518 20.297 20.413 20.204 20.303 20.325
20.631* 2 0.529 0.360 0.314 0.259 0.206 0.229 0.082 0.456 20.345 0.431 20.009 0.225 20.019 0.447 20.355 0.185 20.283 20.333 20.373 20.242 20.353 20.447 20.368 0.364 0.338 0.305 0.276 0.460 0.449 2 0.052 0.143 0.091 20.007 0.014
2 0.666* 20.524 0.466 0.398 0.326 0.261 0.283 0.094 0.528 2 0.408 0.503 0.032 0.284 0.019 0.453 2 0.428 0.117 2 0.564 2 0.536 2 0.498 20.270 2 0.469 2 0.601* 0.410 0.446 0.424 0.391 0.360 0.524 0.499 20.031 0.191 20.082 0.022 0.066
20.614* 20.524 0.386 0.350 0.298 0.250 0.273 0.135 0.491 20.336 0.467 0.034 0.264 0.027 0.455 20.302 0.203 20.392 20.396 20.393 2 0.275 20.415 20.492 20.353 0.390 0.371 0.341 0.315 0.490 0.479 20.015 0.177 20.053 0.030 0.051 (Continued)
Studies on fabric objective measurement 45
Table AV. The correlation coefficients between extracted features and the KES tensile parameters (106 cases of heavy fabrics)
IJCST 15,1
46
Table AV.
Par.
WT1
WT2
WT3
EMT1
EMT2
EMT3
SCP 0.056 0.154 0.079 2 0.054 0.025 2 0.025 ABC 0.196 0.140 0.100 0.176 0.127 0.097 AF 0.169 0.095 0.091 0.298 0.212 0.192 AH 0.196 0.099 0.066 0.339 0.239 0.190 AHF 0.236 0.125 0.072 0.392 0.281 0.212 ATH 0.280* 0.152 0.081 0.443* 0.317 0.230 AAC 0.268 0.144 0.075 0.431 0.309 0.224 ALF 0.356* 0.198 0.094 0.527* 0.372 0.257 LAAC 0.248 0.137 0.103 0.404 0.278 0.223 LABC 0.123 0.075 0.010 0.135 0.084 0.032 AUD 0.261 0.151 0.106 0.427 0.297 0.232 TGA 0.242 0.117 0.053 0.377 0.261 0.192 AGA 0.187 0.061 0.045 0.326 0.202 0.180 AGL 0.348* 0.170 0.094 0.506* 0.333 0.255 EMGL 0.137 2 0.022 0.050 0.262 0.088 0.149 LAMG 0.030 0.055 2 0.042 20.103 2 0.045 2 0.115 ET 0.142 0.020 2 0.007 0.268 0.123 0.106 SI 2 0.115 0.005 0.127 2 0.190 2 0.156 2 0.058 SUC 2 0.102 0.050 0.128 2 0.219 2 0.152 2 0.083 PBS 2 0.082 0.049 0.093 2 0.227 2 0.149 2 0.112 PSL 2 0.155 2 0.076 0.052 2 0.299 2 0.255 2 0.147 P 2 0.195 2 0.125 2 0.058 2 0.335 2 0.288 2 0.223 ES 2 0.009 0.192 0.178 2 0.160 2 0.013 2 0.018 ECP 0.280* 0.138 0.068 0.335 0.179 0.128 EF 0.176 0.104 0.097 0.303 0.215 0.196 EH 0.190 0.104 0.078 0.329 0.233 0.193 EHF 0.215 0.117 0.079 0.362 0.258 0.205 EML 0.241 0.131 0.082 0.395 0.280 0.217 ECD 0.200 0.111 0.094 0.340 0.238 0.201 EBC 0.208 0.105 0.083 0.359 0.244 0.202 CPD 0.206 0.100 0.039 0.323 0.226 0.161 EMGD 0.204 0.073 0.051 0.346 0.215 0.188 EE 0.188 0.106 0.038 0.282 0.191 0.135 EBCC 0.181 0.090 0.032 0.297 0.206 0.152 MG 0.192 0.092 0.043 0.313 0.217 0.168 Note: Bold values are used to show the highest correlation among the three sets of data and the sign * is used to show the three highest correlation values among all the features.
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
A study on the relaxation behavior of fabric’s crease recovery angle
Study on fabric’s crease recovery angle 47
Xia Dong Dong Hua University, Shanghai City, People’s Republic of China
Jianchun Zhang and Yan Zhang
Received July 2000 Revised July 2002 Accepted July 2002
The Quartermaster Research Institute of the General Logistics Department of the CPLA, Beijing, People’s Republic of China
Mu Yao Northwest Institute of Textile Science and Technology, Xi’an City, People’s Republic of China Keywords Textiles, Strain Abstract On the basis of Viscoelasticity Theory of textile materials, the Maxwell’s Model, Kalvin Paunting Model and Burger’s model were analyzed in the paper. The regression equation of the strain relaxation curve was established and the experiments verify that the crease recovery angle can be predicted by the regression equation of Burger’s model. The relaxation behavior of crease recovery angle varying with time was according to the creep-relaxation equation.
Introduction Materials of fabric are generally high-polymers. If a fabric was folding pressed or heated or effected by both of them, the fabric would be deformed. Its deformation would have three formats: fast-elastic deformation, delayed deformation, and permanent deformation. Fast-elastic deformation and delayed deformation are reversible, but permanent deformation is irreversible. In the paper, the reversible processes of fast-elastic deformation and delayed deformation were studied with experiment and theoretic analysis. Fabric samples and experiment Six samples were selected in the paper. Sample number 1 and 2 are wool serge, 3 and 4 are wool/polyester (45/55) blending serge, and 5 and 6 are wool-like polyester serge. These six samples were finished with the same pressure decatizing techniques. The fabric texture parameters of these six samples are listed in Table I. The sample is rectangle in shape and its size is 60 mm £ 50 mm: In the experiment, the sample was folded and pressed by 1,500 g wt, firstly heated at 1008C temperature for 3 min, then chilled at 08C for 3 min. And then maintain the sample at 208C, 65 percent RH environment and remove the weight.
International Journal of Clothing Science and Technology Vol. 15 No. 1, 2003 pp. 47-55 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310461150
IJCST 15,1
Subsequently test the crease recovery angle of the sample going with relaxation time. The schematic diagram of the experiment is shown in Figure 1. The experiment results are listed in Table II.
48
Theoretic analysis The fabric material is a macromolecule polymer. The deformation behavior is according to the viscoelasticity characteristics of high-polymer. Therefore, we
Sample code Weave
Table I. Fabric texture parameters of six samples
No. 1 No. 2 No. 3 No. 4 No. 5 No. 6
Serge Serge Serge Serge Serge Serge
Warp counts Weft picks (threads/10 cm) (threads/10 cm) Weight (g/m2)
Material Wool Wool 45/55 wool/polyester 45/55 wool/polyester Wool-like polyester Wool-like polyester
623.0 296.0 332.1 331.8 334.0 326.6
282.0 263.0 242.4 241.6 264.8 251.0
400.0 199.0 247.1 245.3 219.3 228.6
Figure 1. Schematic diagram of experiment
Relaxation time Sample No. 1 No. 2 No. 3 No. 4 Table II. Crease recovery angle and its corresponding time
No. 5 No. 6
Warp Weft Warp Weft Warp Weft Warp Weft Warp Weft Warp Weft
0s
5s
10 s
30 s
1 min
5 min
10 min
30 min
60 min
120 min
0 0 0 0 0 0 0 0 0 0 0 0
48.7 35.3 30.7 26.0 4.7 6.3 4.5 5.6 27.3 25.7 12.7 16.3
51.0 37.3 33.7 28.3 5.3 6.7 5.6 7.3 29.0 26.7 16.0 17.0
53.7 40.0 35.7 29.3 10.3 8.3 8.7 10.3 29.3 27.3 18.0 18.0
55.7 40.3 38.0 32.7 10.7 18.3 10.3 12.5 30.0 27.7 18.3 19.0
58.0 44.7 39.7 36.0 30.7 29.3 28.5 29.6 30.7 29.0 20.3 21.7
58.7 47.3 41.0 36.7 31.0 29.7 29.6 32.4 32.7 29.3 21.7 23.7
61.7 48.3 43.7 39.3 31.3 33.0 30.2 34.5 34.3 30.0 23.0 26.0
63.0 49.0 44.3 40.7 32.0 33.0 31.0 34.5 36.0 31.7 25.1 27.6
65.9 52.0 47.0 43.6 32.1 33.0 31.0 35.0 37.8 33.5 26.0 28.2
analyze the crease recovery angle relaxation behavior with the viscoelasticity Study on fabric’s theory. There are some common models for mechanical analysis of textile crease recovery materials. The simplest model is Maxwell’s model. It is constituted by a seriesangle wound about a Hookean elasticity spring and a Newtonian viscosity pot. More complicated models are Kalvin Paunting model and Burger’s model. The detailed deduction process of the Burger’s model is described as follows. 49 Assume that the thickness of the fabric is b, and the weight is P in the schematic diagram of Burger’s model in Table III. So the outer surface deformation is 11 þ 12 þ 13 , where 11, 12, and 13 are the corresponding strains. E1 and E2 denote Hookean elasticity of spring numbers 1 and 2. h 2 and h 3 denote Newtonian viscosity of pot numbers 2 and 3. 11 denotes fast-elastic deformation of extending outer-surface of the fabric caused by spring number 1. 12 denotes delayed deformation caused by spring number 2 and pot number 2. Maxwell’s model u ¼ A1 + B1t A1 No. 1 Warp Weft No. 2 Warp Weft No. 3 Warp Weft No. 4 Warp Weft No. 5 Warp Weft No. 6 Warp Weft
47.0645 35.4084 31.6907 27.3961 14.0308 14.9963 13.2488 14.9069 25.8566 23.8167 24.3841 16.9421
Correlative B1 (£102 3) coefficient r1 3.3631 2.9487 2.7117 2.8401 3.5128 3.5015 3.4481 3.8685 2.0973 1.6709 11.9925 2.0639
0.4190 0.4667 0.4767 0.5410 0.6077 0.6206 0.6217 0.6428 0.4654 0.4162 0.3807 0.5934
Kalvin Paunting model u ¼ A2 eB2X A2 54.4257 40.8408 36.5508 31.5092 12.3370 13.9892 11.7227 13.7864 29.9271 27.5880 17.7789 19.4266
Correlative B2 (£102 5) coefficient r2 3.2033 4.1011 4.2599 5.5289 18.6748 16.7549 18.9785 18.0711 3.8714 3.0650 6.7163 6.7433
0.7856 0.7459 0.7517 0.7609 0.5589 0.5614 0.5728 0.5895 0.8703 0.8884 0.7179 0.7777
Burger’s model u ¼ A3 eB3t + C
No. 1 No. 2 No. 3 No. 4 No. 5 No. 6
Warp Weft Warp Weft Warp Weft Warp Weft Warp Weft Warp Weft
A3
B3 (£102 2)
C
Correlative coefficient r3
2 58.8122 2 45.2274 2 40.7613 2 36.1841 2 29.7624 2 29.5505 2 28.4957 2 30.7947 2 32.6469 2 29.5127 2 21.4156 2 22.8326
2 30.694 2 24.901 2 23.354 2 19.557 2 7.9100 2 1.1128 2 0.7202 2 0.6451 2 32.508 2 37.478 2 14.335 2 17.499
59.0625 45.6225 41.1223 36.7615 31.8993 31.8945 30.7020 34.2558 32.7439 29.5640 21.8375 23.3940
0.95519 0.92436 0.93236 0.88417 0.98720 0.98011 0.99012 0.98789 0.93178 0.95733 0.88487 0.81851
Table III. Regressive coefficients (part 1)
IJCST 15,1
50
13 denotes the plasticity deformation caused by pot number 3. t denotes the relaxation time of the crease recovery angle. Presume that deformation of the sample outer-surface is extension deformation. Therefore, when weight was removed, the fast-elastic deformation of the spring number 1 must be returned and made the sample regaining an angle u1. Then the delayed deformation of the spring number 2 and the viscosity pot number 2 would return laxly and made the sample regaining an angle (u22 u1). At that time, the sample has a crease recovery angle u2. Because the pot number 3 causes plasticity deformation, the permanent plasticity deformation of the fabric can not return, the crease recovery angle would be a constant finally. So, for spring number 1, the following equation comes into existence.
s ¼ E 1 11
ð1Þ
For spring number 2 and the viscosity pot number 2, the following equation comes into existence.
s ¼ E 2 12 þ h 2
d12 dt
ð2Þ
For the viscosity pot number 3, there is an equation shown as follows.
s ¼ h3
d13 dt
ð3Þ
Hence, d1 1 ds s E 2 12 s ¼ þ 2 þ dt E 1 dt h2 h3 h2
ð4Þ
d2 1 1 d2 s 1 1 ds E 2 d12 ¼ þ þ 2 dt 2 E 1 dt 2 h 2 h 3 dt h 2 dt
ð5Þ
hence,
There is an equation where 1 ¼ 11 þ 12 þ 13 : d12 d1 d11 d13 d1 1 ds s 2 2 ¼ 2 ¼ 2 dt dt E 1 dt h 3 dt dt dt Solving equations (5) and (6), the following equation can be deduced. d2 1 E 2 d1 1 d2 s 1 1 E 2 ds E2 þ þ þ þ s ¼ þ 2 2 dt h 2 dt E 1 dt h 2 h 3 E 1 h 2 dt h 2 h 3
ð6Þ
ð7Þ
If the exertion s equals the constant P, the following equation would be Study on fabric’s obtained. crease recovery
angle
d2 s ds ¼ 0: ¼ dt 2 dt Therefore, the creep equation would be described as follows. E2 P P P 2h t 1¼ þ 12e 2 þ t E1 E2 h3 where P/E 1 – fast-elastic deformation; E2 P 2 t 1 2 e h2 – delayed deformation; E2 P t – plasticity deformation. h3 So we can get the creep-relaxation equation from equation (8). E2 1 ¼ 12 exp 2 ðt 2 t1 Þ þ 13 h2
51 ð8Þ
ð9Þ
where 12 ¼
E P 2 2t 1 2 e h2 1 ; E2
13 ¼
P t1 h3
The analysis results of these three models are shown in Figure 2. In the experiment, the crease recovery angle is tested after the weight is removed from the sample; therefore, the value of the angle is actually dependent on the strains of the fabric. When the weight was removed, the strain of the fabric was fluxing with creep-relaxation equation. So, the flux of angle u must be according to the creep-relaxation equations in Figure 2. Given that thickness of the fabric is b, extension deformation of outer-surface was 1. The relationship equation of the crease recovery angle u and the strain 1 would be that 1 ¼ pbu/1808. Therefore, the following equation would be obtained.
u ¼ A1 þ B1 t
ð10Þ
u ¼ A2 eB2 X
ð11Þ
u ¼ A3 eB3 t þ C
ð12Þ
where Ai, Bi, C are all constant coefficients, i equals 1 or 2.
IJCST 15,1
52
Figure 2. Models and its creep-relaxation equation
Results and analysis Regress the testing data in Table II with equations (10)-(12). The estimate and asymptotic standard error of regressive coefficients Ai, Bi, and C are listed in Table III, where r is the asymptotic in computing process. Higher the values of r, closer the correlation. The conclusion can be deduced from Table III that the Burger’s model is closer to the actual status of samples. The values of r that were regressed with equation u ¼ A3 e B3 t þ C are all larger than 80 percent, in general larger than 90 percent. Whichever warp or weft of every sample would have a curve equation just as equation (12) and the values of coefficients Ai, Bi, and C are different from each other. Based on the values of Ai, Bi, and C in Table III and equation (12), the predicted values of crease recovery angle of samples are calculated and listed in Table IV. Predicted values of crease recovery angle of samples in Table IV are closer to the testing results in Table II. Therefore, equation (12) is similar in its behavior to the crease recovery angle relaxing with the relaxation time.
Relaxation time Sample No. 1 No. 2 No. 3 No. 4 No. 5 No. 6
Warp Weft Warp Weft Warp Weft Warp Weft Warp Weft Warp Weft
0s
5s
10 s
30 s
1 min 5 min 10 min 30 min 60 min 120 min
0.25 0.40 0.36 0.58 2.14 2.34 2.21 3.46 0.10 0.05 0.42 0.56
46.39 32.60 28.44 23.15 3.29 3.94 3.21 4.44 26.32 25.03 11.38 13.88
56.33 41.87 37.18 31.64 4.40 5.46 4.19 5.38 31.48 28.87 16.73 19.43
59.06 45.60 41.09 36.66 8.42 10.73 7.74 8.88 32.74 29.56 21.55 23.27
59.06 45.62 41.12 36.76 13.38 16.74 12.20 13.34 32.74 29.56 21.83 23.39
59.06 45.62 41.12 36.76 29.13 30.85 27.42 29.81 32.74 29.56 21.84 23.39
59.06 45.62 41.12 36.76 31.64 31.86 30.32 33.61 32.74 29.56 21.84 23.39
59.06 45.62 41.12 36.76 31.90 31.89 30.70 34.26 32.74 29.56 21.84 23.39
59.06 45.62 41.12 36.76 31.90 31.89 30.70 34.26 32.74 29.56 21.84 23.39
59.06 45.62 41.12 36.76 31.90 31.89 30.70 34.26 32.74 29.56 21.84 23.39
Study on fabric’s crease recovery angle 53
Table IV. Predicted values of crease recovery angle of the samples
Testing results and predicted values of crease recovery angles fluxing with relaxation time are shown in Figures 3-5. Analysing Figures 3-5, it can be concluded that every tested curve is similar to its corresponding predicted curve. Therefore, a Burger’s model is in point for the textile material. And the relaxation behavior of crease recovery angle is according to the creep-relaxation equation. The pure wool serge samples have larger crease recovery angle, but the crease recovery angles of the wool/polyester blending serge samples and pure polyester serge samples are much smaller. Tested results and predicted results
Figure 3. Crease recovery angle of sample numbers 1 and 2
IJCST 15,1
54
Figure 4. Crease recovery angle of sample numbers 3 and 4
Figure 5. Crease recovery angle of sample numbers 5 and 6
are listed in Tables II and IV. The tested and predicted angles of every pure Study on fabric’s wool serge sample would terminate at about 508. Wool/polyester blending crease recovery serge samples and pure wool-like polyester serge samples have a terminate angle press angle of about 308. But the press angles of pure polyester serge samples would be much smaller in appearance. In the experiment the serge fabric is folded and pressed under 1,500 g wt at a temperature of 1008C for 3 min, 55 because polyester fabric finalize the design much easier than wool fabric. In practice, more wool the fabric had, bigger press angle the sample would have. Closer the characteristics of wool-like polyester are to wool, the greater the press angle of its fabric. From the former three figures, the difference of wool serge, wool/polyester blending serge and wool-like polyester serge would be decided. If a fabric is pressed and heated, the fabric made up of wool-like polyester material would have good rapid-elastic characteristic and bad slow-elastic characteristic, but its terminating press stability is very good. The fabric made up of pure wool would have bigger press angle and would not be stable under press and heat. Conclusion On the basis of viscoelasticity theory of textile materials, Maxwell’s model, Kalvin Paunting model and Burger’s model were analyzed. The relaxation behavior of crease recovery angle can be predicted by the creep-relaxation equation of Burger’s model. The regression equation of the crease recovery angle relaxation curve could be established on the basis of the Burger’s model. And the estimated values approximate the actual crease recovery angle tested in experiments. The use of Burger’s model was found to be successful for simulating the relaxation behavior of crease recovery angle. Further reading Gu, P. et al. (1999), Material Mechanics (in Chinese). Shen, G.L. (1996), Mechanics of Composite Materials (in Chinese). Yao, M. (1990), Textile Material (in Chinese). Yu, T.Q. and Wu, Y.S. (1997), Trans. Mechanics of Solid Materials (Chinese Translated Version).
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
IJCST 15,1
56
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Mechanical properties of fabric materials for draping simulation Z. Wu
Received June 2002 Accepted October 2002
Department of Mechanical Engineering, Hong Kong University of Science and Technology, Kowloon, Hong Kong, People’s Republic of China
C.K. Au School of Mechanical and Production Engineering, Nanyang Technological University, Singapore
Matthew Yuen Department of Mechanical Engineering, Hong Kong University of Science and Technology, Kowloon, Hong Kong, People’s Republic of China Keywords Fabrics, Drape, Mechanical properties Abstract Most of the cloth simulation and modelling techniques rely on the energy function of the system. The geometric deformation is related to the energy function by the fabric material characteristics, which are usually difficult to measure directly. This paper discusses how the fabric material properties are related to the measurable mechanical properties of the fabric such as tensile modulus, Poisson’s ratio etc. These properties are incorporated into a cloth simulator to produce draping results. The simulated image and real object are then compared to show the realism.
International Journal of Clothing Science and Technology Vol. 15 No. 1, 2003 pp. 56-68 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310461169
Introduction Cloth modelling has received considerable attention recently. The applications of this modelling technique are mainly for graphic and engineering purpose. Both visual realism and physical accuracy are of equal interest in these two fields. A lot of researches about cloth modelling have been conducted, which mainly focuses on developing physically based models for the cloth object. These models can be classified into two categories: the continuum model and discrete model. For continuum model, the fabric is considered as continuous object with mass and energies distributed throughout. The governing equations are derived from the variational principle. For the discrete model, the object is modelled as a collection of point masses with some relations among each other. A cloth model can be continuous or discrete but the computational methods are ultimately discrete (Gibson and Miritich, 1997). In addition, these models always possess certain physical quantities, particularly the deformation energy, which is defined to afford simulation algorithms.
Literature review for cloth modelling The deformable surface proposed by Teropoulos and Fleischer (1988) is a typical continuum model for the cloth. Furthermore, various approaches (Chen et al., 1999; Teropoulos and Fleischer, 1988; Volino et al., 1995) are discussed to define the potential energy of a deformable model. The final equilibrium state of the model is obtained by minimizing the potential energy with respect to the material displacements. Finite element method (Eischen and Clapp, 1996; Tan et al., 1999) is also used to find an approximation for a continuous function that satisfies the equilibrium expression. Particle system (Breen et al., 1992, 1994; Eberhardt et al., 1996) is a typical example of the discrete model for modelling of cloth. The cloth is represented by a set of particles. The energies of each particle are defined. The final equilibrium shape of the cloth occurs at the minimum energy of the whole particle system. Mass-spring system is another example of the discrete model for cloth object (Howelett, 1997; Provot, 1995). The cloth is modelled as a collection of point masses connected by springs in a lattice structure. This is a significant approximation of the true physics that occurs in the cloth. The current status in cloth modelling and animation research can be found in Volino and Magnenat-Thalmann (2000). Recently, cloth modelling tends to focus on two major areas. (1) Incorporating fabric property into the cloth model to give an accurate and realistic simulation (Breen et al., 1992, 1994; Eberhardt et al., 1996). (2) Improving the computational efficiency to give a fast simulation (Baraff and Witkin, 1998). Both are crucial in simulating the cloth since the demand of accuracy, realism and fast computation is increasing for engineering and graphic animation. Objective It can be seen that most of the cloth modelling techniques rely on the energy function of the system. The geometric deformation is related to the energy function by the fabric material characteristics, which are usually difficult to measure directly. Kawabata Evaluation system (Kawabata, 1975) is a common equipment to measure the mechanical properties of a fabric. Breen et al. (1994) proposed a method to derive the energy equations of the cloth model from fabric measurement data produced by the Kawabata Evaluation System. Firstly, a function is determined to approximate the Kawabata plots. Then, the approximate function is related to the energy function of the model. Lastly, the resulting equations are scaled to produce energy values in standard physical units. Eischen and Bigliani (2000) used a fifth order polynomial to approximate the Kawabata plot and fitted the approximation into the constitutive equations to obtain the elastic constants of the fabric.
Mechanical properties of fabric materials 57
IJCST 15,1
This paper discusses the characteristics of the fabric material in terms of a set of measurable mechanical properties. These properties are incorporated into a continuum model for draping simulation and the results are then compared with the actual situations.
58
Cloth draping simulation Figure 1 shows the pseudo-codes of a cloth draping simulator based on Teropoulos and Fleischer (1988) continuum model. The state of a cloth element is described by its position x and velocity x_ . Based on this state, the cloth deformation is defined. For instance, the metric and curvature tensors are the typical deformation measurements. The behaviours of the cloth are characterized by the strain energy due to the deformation. Furthermore, internal forces are induced because of this strain energy. Applying the fundamental laws of dynamics such as Newton’s second law of motion on these elements, the acceleration x€ is obtained, which is used to calculate the new state of the element for the next time step through numerical integration. Once the new state is computed, collision detection is performed. Basically, there are two types of collision detection called cloth self-collision and cloth object collision. If the collision is detected at the new state, the new state will be given up
Figure 1. The pseudo codes of a cloth draping simulator
and collision responses are imposed as constraints at that instant. Both collision detection and response are implemented as separate module in the simulator for easy software management. Mechanical properties of fabric material The mechanical properties of a fabric material are measured by the Kawabata Evaluation System under the assumption that fabric is anisotropic and orthotropic in warp and weft directions. Three basic tests are performed: tensile, shear and bending test. Figure 2 shows the plotting of the tensile and bending test of a fabric sample with 60 per cent wool and 40 per cent polyester. The mechanical behaviours of the cloth are governed by the strain energy accumulated due to deformation. Two main deformations occur in cloth draping: metric deformation due to in-plane forces and curvature deformation due to out-of-plane forces. The strain energy density U is written as X U ¼ 1=2 l ab C abdg ldg ð1Þ abgd where lab is the strain tensor due to deformation and C is the elastic modulus tensors; P a, b, g and d are indices to denote the directions of the principle axes; is the aggregation of the strain energies due to metric and curvature deformation. The elastic modulus tensors C abdg are symmetric tensors, hence C abdg ¼ C abgd ¼ C badg ¼ C dgab
ð2Þ
When the warp and weft directions coincide with the principle coordinate system of the fabric, C 1112 ¼ C 1211 ¼ C 1222 ¼ C 2212 ¼ 0 and only five components of the tensor C 1111, C 1122, C 1212, C 2211 and C 2222, are non-zero. From Hook’s law, 2 3 3 2 32 D1 y 2 D1 0 111 s11 1 6 s22 7 6 y 1 D2 76 122 7 D2 0 ð3Þ 4 5¼ 5 4 54 1 2 y 1y 2 0 0 Ds ð1 2 y 1 y 2 Þ s12 2112 2
t11
3
2
H1
6 7 6 1 6 t22 7 ¼ 6m H 4 5 1 2 m1 m2 4 1 2 0 t12
m2 H 1 H2 0
0
32
k11
3
7 76 76 k22 7 5 54 H s ð1 2 m1 m2 Þ 2k12 0
ð4Þ
where s and t are the stress due to metric and curvature deformation, respectively; 1 and k are the strain due to metric and curvature deformation, respectively; D1, D2 are the tensile modulus in the principal axes; Ds are the shear modulus; n1, n2 are Poisson’s ratio; H1, H2 are the bending modulus in
Mechanical properties of fabric materials 59
IJCST 15,1
60
Figure 2. Tensile and bending test result from the Kawabata Evaluation System
n1 D 2 ¼ n2 D 1
ð5Þ
Mechanical properties of fabric materials
m1 H 2 ¼ m2 H 1
ð6Þ
61
the principal axes; Hs are the twisting modulus; m1, m2 are physical quantities analogue to the Poisson’s ratios. and
Hence, the material modulus tensors can be expressed in terms of a set of measurable quantities: 9 D1 > ¼ C 1111 G 12y 1 y 2 > > > 1122 2211 y 2 D1 > > = C G ¼ C G ¼ 12 y 1y 2 ð7Þ D2 > C 2222 ¼ G 12y 1 y 2 > > > > > C 1212 ¼ D s ; G ¼ C 1111 B
9 > > > > > m2 H 1 > ¼ 12m1 m2 > = H1 12m1 m2
C 1122 ¼ C 2211 B B C 2222 ¼ B
H2 12m1 m2
C 1212 ¼ B
Hs
ð8Þ
> > > > > > > ;
where the CG and CB are the elastic modulus tensors due to metric and curvature deformation respectively. If the cloth is stretched in a direction making an angle u with one of the principal axis direction, the tensile strain e u is related to the tensile stress su as su Du ¼ ð9Þ 1u Similarly, the bending strain ku and the bending moment tu can be expressed as tu Hu ¼ ð10Þ ku According to the transformation laws of tensor, 1u ¼ ½cos2 u
sin2 u
2
111
3
6 7 1 7 sinu cosu6 4 22 5 2112
ð11Þ
2
IJCST 15,1
62
s 11
3
2
s u cos2 u
3
7 6 7 6 6 s 22 7 6 s u sin2 u 7 7¼6 7 6 5 4 5 4 2s u cosu sinu s 12
ð12Þ
and 2
k11
3
7 6 7 6 ku ¼ ½cos2 u sin2 u sinu cosu6 k22 7 5 4 2k12
2
t 11
3
2
t u cos2 u
ð13Þ
3
6 22 7 6 7 6 t 7 ¼ 6 t u sin2 u 7 4 5 4 5 u 12 2t cosu sinu t
ð14Þ
Derivation of equations (9)-(14) can be found in most common strength of materials literatures (Ryder, 1973). Substituting equations (11)-(14) into equations (9) and (10) and combining with equations (3) and (4) yields the tensile modulus and the bending modulus in this direction
1 1 1 n1 n2 1 4 cos2 u sin2 u þ sin4 u ¼ cos u þ 2 2 D u D1 D s D1 D 2 D2
ð15Þ
1 1 1 m1 m2 1 4 ¼ cos u þ 2 2 sin4 u cos2 u sin2 u þ Hu H1 Hs H1 H2 H2
ð16Þ
Since the “Poisson ratio” of bending m1 and m2 are usually small, they can be neglected. Taking the angle u ¼ 458, the tensile Poisson ratio and the twisting modulus are written as
1 1 1 1 4 ; n 1 ¼ D1 þ þ 2 2 D1 D2 Ds D458
n 2 ¼ D2
n1 D1
ð17Þ
21 ð18Þ
Mechanical properties of fabric materials
Hence the mechanical properties of modulus D1, D2, Ds, H1, H2 and Hs can be measured directly by the Kawabata Evaluation System. Other parameters such as Poisson’s ratio and twisting modulus are obtained from the tensile, shear and bending modulus by using equations (17) and (18).
63
Hs ¼
4 1 1 2 2 H 458 H 1 H 2
Simulation examples A set of examples is used to illustrate the effects on cloth draping simulation with the incorporation of the mechanical properties of the fabrics. Example 1. A piece of cloth falling on a sphere. The mechanical properties of the fabric are listed in Table I. The size of the cloth is 1 £ 1 m with meshing size of 50 £ 50 nodes and the radius of the sphere is 0.125 m. The animation process is shown in Figure 3. Example 2. Cloth draping over a sphere with various fabric materials. The bending modulus of the cloth is changed to show the draping effects. Three cases are performed: Case 1, the bending modulus is enlarged by 5 times. Case 2, the original bending modulus is used. Case 3, the bending modulus is reduced to 1/5 of the original values. The draping behaviour is shown in Figure 4. The more rigid fabric in case 1 gives a 4-wrinkle draping mode while the softer fabric generates an 8-wrinkle draping mode. Example 3. Cloth draping over a sphere with various fabric materials. Two pieces of cloth with different fabric materials draping over a sphere is shown in Figure 5. The results are compared with the real images. The size of the cloth is 0:6 £ 0:6 m and the radius of the sphere is 0.09 m. The properties are listed in Table II. Example 4. Table cloth draping with various fabric materials. The same simulation in example 3 is performed by using a cloth with varying fabric materials. The mechanical properties are listed in Table III. Each simulation is then compared with the appearance of real table cloth situations as shown in Figure 6. Example 5. Dresses with various fabric materials. Three typical fabric materials, cotton, polyester and silk, are listed in Table IV. The results are simulated with a mannequin as shown in Figure 7. r kg/m2 0.125
D1 N/m
D2 N/m
Ds N/m
H1 mNm
H2 mNm
Hs mNm
n1, n2
3,657
3,627
36.8
5.07
4.41
1.55
0.096, 0.086
Table I. Material properties of cotton and rayon
IJCST 15,1
64
Figure 3. The animation of cloth falling on a sphere
Figure 4. Draping behaviours with various fabric materials
Mechanical properties of fabric materials 65
Figure 5. Different types of cotton cloth draping over a sphere
Material sample r kg/m2 D1 N/m D2 N/m Ds N/m H1 mNm H2 mNm Hs mNm Cotton 1 Cotton 2
0.218 0.23
3,475 2,405
2,865 5,315
Material sample
r kg/m2
D1 N/m
D2 N/m
Cotton
0.208
3,475
2,865
Rayon
0.129
1,847
3,644
191 39.6
Ds N/m 191 15.2
87.7 11.6
61.4 10.7
11.42 3.1
n1, n2 0.215, 0.177 0.185, 0.165
H1 mNm
H2 mNm
Hs mNm
n1, n2
17.6
12.7
2.7
4.4
2.3
0.71
0.215, 0.177 0.185, 0.165
Table II. Mechanical properties of two different cotton samples
Table III. Material properties of cotton and rayon
IJCST 15,1
66
Figure 6. Table cloth simulation with different fabric materials
Material sample Table IV. Material properties of cotton, polyester and silk
Cotton Polyester Silk
r kg/m2 D1 N/m D2 N/m Ds N/m H1 mNm H2 mNm Hs mNm 0.208 0.212 0.076
3,475 3,071 1,388
2,865 1,823 827
191 46.7 12.4
17.6 10.9 0.81
12.7 9.0 0.76
2.7 2.31 0.089
n1, n2 0.215, 0.177 0.264, 0.169 0.285, 0.170
Discussion Fabric is made from threads in woven or knitted patterns with the structures resulting from different weaving or knitting techniques. Due to the complexity of the fabric microstructure, it is necessary and practical to treat the fabric as an engineering material in draping modelling. One of the feasible solutions is to
Mechanical properties of fabric materials 67
Figure 7. Dresses simulation with various fabric materials
assume the fabric to be continuous elastic in both the modelling and the experimental aspects. Hence, it is equivalent to modelling the draping of a fabric sheet. Fabric material exhibits the orthogonal anisotropy behaviours, which leads to different mechanical properties in the weft and warp direction. Figure 8 shows the tensile modulus of cloth. The mechanical properties of the fabric are considered in three different directions: the weft and warp directions; and an angle of 458 from one of the weft direction. Once these properties are measured and calculated, they can be inserted into the cloth model for draping simulation.
Figure 8. The tensile modulus of cloth
IJCST 15,1
68
References Baraff, D. and Witkin, D. (1998), “Large steps in cloth simulation”, SIGGRAPH’98, Computer Graphics Proceedings, Annual Conference Series, pp. 43-54. Breen, D.E., House, D.H. and Getto, P.H. (1992), “A physically-based particle model of woven cloth”, Visual Computer, Vol. 8 No. 5-6, pp. 264-77. Breen, D.E., House, D.H. and Wozny, M.J. (1994), “Predicting the drape of woven cloth using interacting particles”, Computer Graphics (SIGGRAPH’94 Proceedings), Addison-Wesley, Reading, MA, pp. 365-72. Chen, M.X., Wu, Z., Sun, Q.P. and Yuen, M.M.F. (2003), “A wrinkled membrane model for cloth draping with multigrid acceleration”, Journal of Manufacturing Science and Engineering (in press). Eberhardt, B., Weber, A. and Strasser, W. (1996), “A fast, flexible, particle-system model for cloth draping”, IEEE Computer Graphics and Applications, pp. 52-9. Eischen, J.W. and Bigliani, R. (2000), “Continuum versus particle representation”, in House, D.H. and Breen, D.E. (Eds), Cloth Modeling and Animation, A.K. Peters, pp. 79-122. Eischen, J.W. and Clapp, T.G. (1996), “Finite-element modelling and control of flexible fabric parts”, IEEE Computer Graphics and Applications, pp. 71-80. Gibson, S.F.F. and Miritich, B. (1997), “A survey of deformable modeling in computer graphics”, Technical report TR-97-19, Mitsubish Electric Research Laboratory. Howelett, P. (1997), “Cloth simulation using mass-spring networks”, MSc dissertation, Department of Computer Science, University of Manchester. Kawabata, S. (1975), “The standardization and analysis of hand evaluation”, Hand Evaluation and Standardization Committee of the Textile Machinery Society of Japan, Osaka. Provot, X. (1995), “Deformation constraints in a mass-spring model to describe rigid cloth behavior”, Proceedings of Graphics Interface, pp. 147-54. Ryder, G.H. (1973), Strength of Materials, ELBS and Macmillan, NY. Tan, S.T., Wong, T.N., Zhao, Y.F. and Chen, W.J. (1999), “A constrained finite element method for modelling cloth deformation”, Visual Computer, Vol. 15 No. 2, pp. 90-9. Teropoulos, D. and Fleischer, K. (1988), “Deformable models”, Visual Computer, Vol. 4 No. 6, pp. 306-31. Volino, P. and Magnenat-Thalmann, N. (2000), Virtual Clothing Theory and Practice, Springer, Berlin. Volino, P., Courchesne, M. and Thalmann N.M. (1995), “Versatile and efficient techniques for simulating cloth and other deformable objects”, SIGGRAPH’95, Computer Graphics Proceedings, Annual Conference Series, pp. 137-44.
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
COMMUNICATIONS
Designing surgical clothing and drapes according to the new technical standards
Designing surgical clothing and drapes 69
M.J. Abreu and M.E. Silva Departamento de Engenharia Teˆxtil, Universidade do Minho, Guimara˜es, Portugal
L. Schacher and D. Adolphe Ecole Nationale Supe´rieure des Industries Textiles de Mulhouse, Mulhouse, Cedex, France Keywords Protective clothing, Performance, Medical Abstract Hospitals will continue to be the largest consumers of disposables, because of the diverse range of procedures they provide. Favourable growth is forecast for non-wovens. Increasing concern over contamination and nosocomial infections will boost the demand for consumable and disposable surgical gowns and drapes. But, until now neither the manufacturers nor the end users of surgical gowns and drapes could agree on standards. So, a mandatory European standard is being developed to establish basic requirements and test methods for disposable and reusable materials used for surgical gowns and drapes. Once this standard has been adopted, the continued use of cotton textiles and conventional cotton-polyester mixed textiles will become questionable.
Introduction According to the Medical Device Directive 93/42/EEC, surgical clothing, drapes and air suits are considered to be medical devices, whether they are reusable or disposable gowns and drapes. Medical products are divided into four classes: I, IIa, IIb and III. The classification is in accordance with Annex IX of the Directive. All non-invasive medical products are class I. This way surgical gowns and drapes are medical devices of class I. Only the abdominal towels as invasive products belong in class IIa (Werner, 2000). This Directive on medical devices is the basis for evaluating surgical materials and has been transformed into national law since 14 June 1998 by appropriate regulations in all the European member countries. As part of these legal specifications, CEN (European Committee of Standardisation) has requested that a working group in 1996 (TC 205/WG 14) should develop a standard for these directives. The proposed mandatory European standard “surgical drapes, gowns and clean air suits, used as medical devices, for patients, clinical staff and equipment”, is being developed by the Technical Committee TC 205
International Journal of Clothing Science and Technology Vol. 15 No. 1, 2003 pp. 69-74 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310461178
IJCST 15,1
70
responsible of the Non-Active Medical Devices – workgroup WG 14 “surgical clothing and drapes used in healthcare facilities”. This standard specifies the basic performance requirements and test methods for disposable and reusable materials used to protect the patient, surgical staff and surgical facility. Through a special procedure, each member state of CEN has national mirror groups of the workgroup 14 and has thus, the right to agree or disagree with the proposed new standard. In case of adoption, the standard might also be accepted by the European Comission and be published in the official journal of the EU. The coordination of the national workgroups takes place by means of the respective national standard institute. All medical textiles regarded as medical devices now placed in the market have to comply with the European Medical Devices Directive and bear the CE mark demonstrating that they have been declared fit for their intended purpose and meet the essential requirements of the Directive. If they are supplied sterile then the manufacturer or supplier also needs to consult a Notified Body (Mounter, 1999). European standard The European standard prEN 13795 – “surgical drapes, gowns and clean air suits, used as medical devices, for patients, clinical staff and equipment” is expected to consist of five parts: . Part 1: general requirements for manufacturers, processors and products; . Part 2: test methods; . Part 3: test method for resistance to dry microbial penetration; . Part 4: test method for resistance to wet microbial penetration; . Part 5: performance requirements and performance levels. The first part of the European standard concerning general requirements for manufacturers, processors and products has been published in January 2000 as a prEN 13795-1 and was now submitted to the relevant CEN Technical Committee for final vote. This standard gives general guidance on the characteristics of single-use and reusable surgical gowns, surgical drapes and clean air suits used as medical devices for patients, clinical staff and equipment and intended to prevent the transmission of infective agents between patients and clinical staff during surgical and other invasive procedures. Workgroup 14 is currently working out the other parts – uniform and expressive test methods, as well as the performance requirements and levels. Exceptions Surgical masks, surgical gloves, packaging materials, foot and head wear and incision drapes are not covered by this standard.
Requirements for medical gloves are given in the EN 455 and packaging Designing materials are covered by the EN 868. surgical clothing Requirements for surgical masks and head coverings will be specified in and drapes future CEN/TC 205 standards (CEN, 2001). Surgical drape, gown and clean air suit materials Cotton textiles or conventional mixed cotton-polyester textiles, the traditional materials for surgical gowns and drapes, meet a number of requirements, such as comfort, drapeability, good tensile strength, steam permeability and steam sterilisability. However, they will not meet the requirements of the new standard which includes resistance to microbial penetration, resistance to liquid penetration and minimal release of particles. Cotton containing gowns and drapes must be replaced and the future looks bright for quality, properly processed reusable products or quality single-use disposable products (Patel et al., 1998). There is a wide variety of non-wovens of all types including hydroentangled, bonded, stitched, and laminated, of a range of quality depending on the intended use by the manufacturer. There are also combinations of various fabrics in one product and there are reusables which range from basic cotton and polyester/cotton weave fabrics, to micro-filament, micro-weave polyester (Mounter, 1999). Various single-use products and reusable materials have been proposed for surgical gowns and drapes with the objective of reducing microbial contamination of the incision and protecting the operating room staff from infection. However, this is not concerned about the user awareness. At present, producers and users are not getting enough information about the suitability of the materials used in surgery. In the interest of patient and personnel protection, identical requirements must be applied to single-use products and to reusable surgical materials. Surgical gowns These products are used in the operating theatre to prevent transfer of infective agents. Surgical drapes These products are used in the operating theatre to cover the patient and equipment to protect them from pollutant particles in the air, which carry infective agents. Clean air suits These products intended to minimise contamination of the operating wound by the wearer’s skin scales carrying infective agents via the operating room air thereby reducing the risk for wound infection.
71
IJCST 15,1
72
Surgical drape and gown requirements There are a number of factors that need to be considered to determine what type of textile to use in the operating theatre, which will be based on a number of various, perhaps competing priorities: Performance requirements Other general requirements for surgical drapes and gowns include aesthetics (including conformability, tactile softness, and comfort), fibre tie-down properties (linting and abrasion resistance), flame resistance, static safety and toxicity (Tables I and II). For surgical drapes, stiffness is very critical because the barrier performance could affect the conformability to patient or equipment. As to gowns, comfort and stiffness may affect perspiration and movement. Strength requirements include tensile, tear, burst and puncture resistance. Linting has to be avoided, because particles from gown or drape may complicate the wound healing process. Good abrasion resistance is a basic requirement for the safety of barrier materials. Flame and electrostatic resistance is needed especially for laser applications and oxygen administration, because of the danger of explosion (Adanur, 1995).
Table I. Requirements for surgical gowns
Resistance to microbial penetration – dry Resistance to microbial penetration – wet Cleanliness – microbial Cleanliness – particulate matter Linting Resistance to liquid penetration Bursting strength – dry Bursting strength – wet Tensile strength – dry Tensile strength – wet
Table II. Requirements for surgical drapes
Resistance to microbial penetration – dry Resistance to microbial penetration – wet Cleanliness – microbial Cleanliness – particulate matter Linting Resistance to liquid penetration Bursting strength – dry Bursting strength – wet Tensile strength – dry Tensile strength – wet Adhesion for fixation for the purpose of wound isolation
The resistance to microbial penetration is important for both products and in a Designing hospital you can find micro-organisms in the air and in liquids (for example in surgical clothing the blood of the patient). The sizes of the solid particle, bacteria and viruses are and drapes 10-40, 0.5-20 and 0.01-0.3 mm, respectively. Type of surgical procedure Some procedures are very invasive with large amounts of body fluids involved and some are regarded as highly risky, such as Orthopaedic and Neurosurgery; others, perhaps less invasive, are regarded as lower risk. It is the responsibility of the end users to determine what fabrics they wish to use and for which surgical procedures. The user endorses the responsibility not to use fabrics for which they were never intended (Mounter, 1999). In the actual operating area, a covering material must be used which presents an adequate barrier throughout the entire duration of the operation, even under mechanical pressure. The same requirement for the protection of the patient must be placed to the same extent on surgical gowns for the protection of the personnel. This concerns the areas, which are exposed to mechanical pressure or liquid as, for example, the front of the gown and the sleeves up to the elbow. Every operation is associated with certain risks of contamination. The medical success of a procedure can be especially affected by post-operative infections. A post-operative infection depends on numerous factors such as the duration of the operation, the operation technique, the mechanical pressure and the patient’s vulnerability to infection (Werner, 2000). Processing requirements The primary purpose of sterilising an item is to render it safe for use by destroying all living microscopic organisms. Because bacteria multiply very quickly, the sterilisation process must be absolute. Even a few organisms invading the patient’s body during a surgical procedure can reproduce rapidly and contribute to post-operative complications. Four common types of sterilisation are in use today: gas, irradiation, steam autoclave and dry heat. The first two types of sterilisation are also called low temperature sterilisation methods, applied to single-use products and the last two types, high temperature sterilisation methods, applied to reusable products (Arau´jo Marques, 1997). The traceability of decontamination, disinfection and sterilisation shall be maintained. The properties of the materials will have to be maintained using the agreed processing or the other procedures that can maintain the properties of the material. A processing specification will have to be designed and validated for the product, including visual and hygienic cleanliness, decontamination, disinfection and sterilisation (Werner, 2000).
73
IJCST 15,1
74
Costs In Europe, 30 million operations are performed every year. Assuming that the cost of one surgical infection in Europe amounts to 2,500 Euros, 1 per cent reduction in the surgical infection rate saves 748 million Euros each year. So we can safely assume that the reduction of post-operative infection would generate considerable savings. Conclusions Users of surgical gowns and drapes can continue to use whatever type of textile product they wish, although they do have a responsibility to take manufacturers claims for their products into account and not to use devices for purposes for which they were never intended. The Medical Device Regulations do not prevent the surgical staff from using whatever type of textile they wish. Only if there is good collaboration between users, producers and hygienists, the creation of standards will succeed. From the medical point of view, the protection of the patient and that of the personnel must be equally guaranteed whether single-use materials or reusable medical products are concerned. Neither reusable materials nor single-use products can generally be designated as suitable or unsuitable. We acknowledge that there is possible difference between degrees of exposure between very minor and exploratory surgery and very major and open surgery, but we do not consider “medium” risks. Directive 93/42/EEC relating to medical devices identified that, although the emphasis on standards varies from country to country, the essential requirement of any product is to provide a high level of protection. It is the responsibility of hospitals to ensure optimal protection of patients and users. In a lawsuit, this could lead to the burden of proof being shifted to the hospital, with the risk of liability. Because of this risk alone, the question of what surgical materials to use in the future is going to take up extreme significance. References Adanur, S. (1995), Wellington Sears Handbook, Technomic Publishing, USA. Arau´jo Marques, M.J. (1997), “Contribuic¸a˜o para a definic¸a˜o de propriedades de materiais teˆxteis hospitalares descarta´veis”, Masters thesis. CEN/TC 205/WG 14, (2001), “prEN 13795: surgical drapes, gowns and clean air suits, used as medical devices, for patients, clinical staff and equipment – Part 1”, General Requirements for Manufacturers, Processors and Products. Mounter, S. (1999), “Enforcing medical device regulations for medical textiles”, Index 99 Congress-Medical Session, 27-30 April, Geneve, Switzerland. Patel, S.R., Urech, D. and Werner, H.P. (1998), “The surgical gowns and drapes of tomorrow”, Medical Device Technology, September 1998. Werner, H.P. (2000), “The standardisation of hospital textiles and hygiene aspects of hospitals”, FiberMed 2000 Conference, 12-14 June, Tampere, Finland.
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
IJCST 15,2
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
3D collar design creation Jing-Jing Fang
88 Received October 2002 Accepted January 2003
Department of Mechanical Engineering, National Cheng Kung University, Tainan, Taiwan Keywords Clothing, Design, Object-oriented computing, Computer aided design Abstract This preliminary research revolute the conventional clothing design process by true designs from three-dimensional (3D) rather than two-dimensional. The aim of the research is to develop a handy 3D clothing design software tool for general garment designers. Work carried out in this paper is the preliminary result of the 3D software infrastructure. In addition, 3D collar design based on a mathematical formula is accomplished as a template for other garment portions. Object-oriented technology is invoked as a tool for software developing. The system is divided into two major modulus, the user interface and the kernel. The user interface is used to collect messages from the users, and then send it to the kernel for further computations. Moreover, it exhibits realtime pictures received from the kernel. The major work of the kernel is to handle the operations that are called by the user interface. In this paper two basal collars, convertible collar and shirt collar, are illustrated as diversified figurations.
1. Introduction Human life has become much more convenient and comfortable than ever in the 21st century. As human living conditions are well satisfied, exterior appearance becomes more and more important in this modern society. What a person wears represents his or her own temperament and style. Even the ready-made-garment is popular and economical; while its duplicate style would far from satisfy a person’s individual appetite. Traditional procedures for customer-made costumes are time consumable and laborsome; in addition, it is costly. The development of clothing design automation is to fulfill the target of customer-made requirements, uniqueness, and creating of individual style. The fundamental requirement of clothing design is framed on customer’s body geometry. Through self-developed body scanner, body features are automatically extracted by geometrical analysis. Cheng (2001) reconstructed human body surface originally from scanned human data points. According to anthropometry and body measurement for garment manufacturing, crucial body features are extracted and regenerated by finite structuralized feature points. Phenomenally, structuralized points below 5,000 are capable to duplicate body shape in detail. Based on these structuralized data, garment is directly generated by ways of points, girths, lengths, widths, heights, and even International Journal of Clothing Science and Technology Vol. 15 No. 2, 2003 pp. 88-106 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310470088
The authors thank Professor Wu, L.J., Dr Tsai, Doris and Ms Huang, S.T. of the Fashion Design Department in Tainan Girl Technical School for their contribution on clothing consultancy. Our most appreciation goes to Professor Tsai from Mechanical Engineering Department, who constructed and developed the body scanner and its automatic anthropometry software.
gradients. The developing software provides the tools of conception, 3D collar design modification in three dimensions, as well as patterns flatness derived from creation outcomes of the concept. Creativity is carried out from three-dimensional (3D) garment models, which are originally generated and built from individual body structures. Through the given interface of clothing design software, the customer and the designer can easily communicate through 3D navigation on a 89 computer screen, as well as conception details. Complex and costly design processes of customer-made garments become simple and reasonable. Compared to conventional two-dimensional (2D) sketch on papers, it provides easier expression between customer and designer. In addition, it allows the designer and the customer to refashion and refine the final outcomes in 3D before going into the stage of manufacturing. The aim of the research is to develop automatic garment design tools to induce traditional textile industry elevation. Moreover, it improves the quality of design procedure by way of real-time 3D garment shape modification. To achieve the goal of the above destination, design and refinement of two representative basal collars are chosen and accomplished as a preliminary and template work for further study. Based on computer graphics theory, the kernel of clothing system is built. It contains fundamental interface components such as, spatial exhibition, different viewpoint selections, alternative display modes, store and restore for special view, zoom in and out, etc. In order to provide collars shape creation and modification, we add the mathematical magic of parameterized formula with NURBS surfaces inside the kernel. 2. Investigations of relevant work In recent years, progress in computer industry has triggered high performance within common personal computers. Computer assisted drawing, design, engineering, manufacturing, or even surgery, adopting the technique of computer geometric modeling are in its florescence. With the ability of 3D computer animation in real-time, relevant studies and applications are widely expanded in apparel industry. We survey common commercial products and the relevant research, classify the different approaches among these tools, and then compare them with our work later. In general, most of the commercially available garment design software is based on 2D pattern maker such as the productions from PatternMaker Inc. ( http://www.patternmaker.com), Geoffrey E. Macpherson Ltd (1938) and DoCAD Ltd (2001) highly relies on the experience of the tailors. Automatic marking, grading to manufacturing in Gerber Technology Company (1969) successfully apply pattern to apparel sewing. However, they merely concentrated on pattern design in 2D that does not investigate the distinctions among individual human body. Instead of designing in three dimensions, the appearance of 3D garment in Pad system Inc. (1988), Computer Design Inc. (1992) and Terzopoulos et al. (1987) is simply for visualization,
IJCST 15,2
90
and not for refashionment. Most of the commercial software for garment design are based on 2D pattern maker, and thus, sewing up a 3D garment for visualization. In addition, it lacks fitting survey and its wearing effects. The 3D approached design software for sark, named 3D Design Concept by Computer Design Inc. (1992), allows users drawing brassiere profile on a sample mannequin to flatten into patterns. Their projective method of 2D drawing a ready-made-bra on a grading mannequin top, which is anticipated that the design would not adapt to women’s bra that usually be discovered incompletely symmetry. To deal with such imperfectly symmetric human body is one of our research goals in the next stage. In 1989, a research group founded at the University of Geneva worked on the research of virtual human simulation and virtual worlds. Their work (Hadap et al., 1999; Volino and Magnenat-Thalmann, 1997) focuses on the phenomena of wearing effect of a virtual garment on a virtual human. Based on Terzopoulos’s theory (1987), simulation of deformity and elasticity in wearing clothes is one of their research topics. By employing a self-developed body scanner and associated structuralized software (Cheng, 2001), crucial features of human body is obtained automatically. Based on body measurement by tailor, and anthropometry, the body features for garment manufacturing, such as bust width, shoulder width, neck girth, chest girth, waist girth, and hip girth, etc. are automatic acquisition without manual measurements. Garment is then created from the structuralized body.
3. Kernel environment In general, the development of a magnificient software needs to accumulate amounts of outcomes by teamwork for years. Most of the problems are exposed not during the beginning stage but during integration. Therefore, preplanning the communication and integration mechanism among modulus becomes a fateful issue of the software. Regarding the development of any applicable software, it contains three levels, the kernel, the support, and the platform, which usually operate synchronously. The kernel level handles the core calculations demanded from the user interface. Whereas, the support level provides the definitions of file format and library used in the kernel level. The platform level handles communication mechanism among users, software, and operating system. All these three levels consists an ideal software of high expansibility, portability, and general applicability. Based on the frame of Microsoft Foundation Classes (MFC) and 3D drawing functions provided by OpenGL, the system invokes object-oriented methodology as software developing infrastructure. Object-oriented design method is adopted for developing every component of garment.
3.1 System organization 3D collar design This preliminary system mainly consists of two levels, the kernel and the user creation interface, whereas, the interface simply contains the support and the platform levels. The interface comprises of four classes: Mouse, ViewPoint, CDesign3dView, and CMainFrame as shown in Figure 1. It allows the users to manage and navigate within space via simply a 2D mouse. The 91 CDesign3dView is tied in with the members in CMainFrame to deal with the process of message receiving from the users. It contains kernel module, Mouse and ViewPoint classes. Once CDesign3dView receives message from CMainFrame, the messages are transferred to kernel for advanced calculation, or to Mouse or ViewPoint for corresponding operations. Therefore, object navigation in 3D is reached by using a mouse. Figure 2 exhibits the communication mechanism between the kernel and the user interface. The kernel transmits messages through CDesign3dView. Once CDesign3dView receives the commands from the users, a dialog window is triggered to provide users the inputs of specific data. Geometric objects are constructed accordingly. Drawing modalities, such as point, line, lattice, and surface, one of them is selected and then transmitted to the kernel for calculation before canvas drawing. The kernel includes at least Mannequin, Top, Collar, Skirt, Sleeve, and Trouser classes. Figure 3 shows the primary kernel constructions. Currently, only Mannequin and Top with Collar has been built. Top correlates to the Mannequin as well as Collar correlates to Top. In other words, Top is not proportional offset generated from the Mannequin, whereas the collar base, by
Figure 1. User interface
Figure 2. System communication mechanism
IJCST 15,2
92
means of collar girth, connect both Top and Collar. For the purpose of figures refashionment, Collar class contains the B-Spline class to allow collar generating. A number of transformation Matrices stores all the modified operations in B-Spline. Due to the varieties of every garment components as well as distinct styles refashionment, we invoked object-oriented techniques as manageable mechanism among garment components. For example, as shown in Figure 4, the CConvertibleCollar and CShirtCollar are both derived from Collar, which retain the features of B-Spline and Matrix for style design and advanced modification. Yang et al. (1992) first applied the object-oriented mechanism to individual component structure for a garment system. They proposed and defined the hierarchical structure to illustrate the relationships for garment components and its templates. In our research, the base class Collar defines all the common features, virtual functions, and implementation functions in order to provide the reusability for derived classes. Based on Collar definitions, the CConvertibleCollar inherits these and develops its own specific features. Similar to CConveritbleCollar, the CShirtCollar represents various styles of shirt collar that would consequently bring into its patterns. 3.2 Results The user interface level provides the display modes such as structure point, line, lattice, and surface rendering to illustrate the outcomes in Figure 5. Currently, eight default views include top, bottom, front, rear, left, right, isometric, and recall stored view. It also allows the users the ability of navigation controlled by a mouse. Navigation to the specific view of interest is stored for intended recalling (Figure 6).
Figure 3. Kernel structure
Figure 4. Inherited relationships of collar
3D collar design creation
93
Figure 5. Display modes
Figure 6. Views
4. The kernel The kernel level in this system includes computation staffs of garment components described in Figure 3. In this section, we are going to describe how to generate both 3D convertible and shirt collars from a given neck girth. Collar is worn surrounding the neck of human body. Based on neck girth, the 3D collar feature and pattern are made possible. According to structuralized
IJCST 15,2
94
Figure 7. Collar design flowchart
Figure 8. Collar structure
neck girth from mannequin features (Cheng, 2001), the 3D curved girth is given as represented points which are not necessary on the same plane. Based on the structuralized points on neck girth, collar shape is generated by those geometrical parameters controlled by the users. Figure 7 displays the collar generating processes in the kernel. Starting from structured neck points, a B-Spline curve is generated by these structuralized neck points. The users can select either convertible or shirt collar thereby their correlative feature parameter is demanded. Collar surface consists of collar band and collar wing two parts. Figure 8 shows half of the collar frame, whereas, mirrors along sagittal plane would obtain the whole collar. The technicality for basal collar is defined as follows. (1) Collar wing – the domain contained in continuous lines of points A0 , B, F, H, D, E, A0 is half of the collar wing. (2) Collar band – the domain surrounded by connected lines of points A0 , B, F, G, C, A, A0 is half of the collar band. Collar band supports collar wing extension, which determine the height of a turnover collar.
(3) Collar girth – the base of collar band is named collar girth. Half of the 3D collar design curve begins from point A through C to G is half of the collar girth. creation It surrounds the base of human neck by means of neck girth. (4) Collar waistline – the line connects collar wing and collar band. Half of the collar waistline passes through points A0 , B, then stops at point F. (5) Collar height – collar height includes three types: rear height (FG), 95 side height (BC), and front height (AA0 ). (6) Collar width – a turnover collar width counted from waistline. It includes three types of rear (FH), side (BD), and front (A0 E). (7) Wing curve – curve passing through points E, D to H is half of the wing curve. 4.1 Collar girth generation Neck girth is one of the crucial feature girths on human body. According to the feature-based anthropometry, bottom neck girth is represented as structuralized neck points that are never coplanar. These structuralized neck points are automatically generated by image processing and curve fitting techniques (Cheng, 2001). To design a comfortable shirt or convertible collar wore on human, practical formula is given by experienced garment designers to fit the length of a collar girth. Here we would describe how to generate 3D collar girth from the structuralized neck points. In order to provide the functions of 3D collar shape refashionment, B-Spline curve and surface are invoked to increase the flexibility of tailor. B-Spline is used to fit the structuralized neck girth as shown in Figure 9. The dots reveal the control points of the B-Spline curve which smoothly constructs the bottom neck girth that connects the human body and neck. The collar girth is formed along the neck girth via these control points. Assume that B-Spline curves P(u) is represented in terms of their blending functions Ni,k(u) (Yang et al., 1992). PðuÞ ¼
X N i;k ðuÞV i
ð1Þ
i
Figure 9. Top view of a neck girth
IJCST 15,2
Vi represents the control points shown in dots in Figure 9. Suppose Qi represents the structuralized neck girth points, then Qj ¼ Pð uj Þ ¼
n X
N i;k ð ui ÞV i
ð2Þ
i¼0
96
Equation (2) is rewritten in matrix form, then control points Vi are gained. P ¼ NV ) V ¼ N 21 P
ð3Þ
Based on the neck girth fitted by the B-Spline curve, an adjustable collar girth is constructed by radiant to expand these control points. Currently, our research starts at basal collar styles. Collar girth and associated basal collars design are focused at those collar girths near neck girth, such as convertible and shirt collars (Figure 10). Beyond that, the collar girth that draped on the shoulder or chest are not considered in this preliminary work. Paranormal penetration between body and collar will be overcome without adapting consumable collision detections. The neck girth, by means of a B-spline curve, that fits the structuralized neck points are never coplanar. Figure 11 shows a top view diagram of a collar girth generated by a neck girth. Suppose Ai is one of the control points in neck girth, Pc is the centroid of the girth, and Bi is the control point in collar girth radial stretched from Ai. Pfn notates the front neck point of a human body. Hence, ui ; /Pfn Pc Ai : Define distance
!
dðui Þ ¼ jAi Bi j;
i ¼ 0; 1; 2; . . .; n
ð4Þ
For each ui, ui that depends on associated Ai obtained from equation (3) are not equivalent. The predefined extending ratios, in terms, every defined ratio is greater than 1, are setup to fit the nature collar girth. Both shirt collar girth and convertible collar girth are initially constructed as shown in Figure 12 for further modification. In addition, based on the parameterized collar girth platform,
Figure 10. Convertible collar (left) and shirt collar (right)
3D collar design creation
97
Figure 11. Collar girth generated from neck girth
we can generate associated collar band and/or its connected collar wing. In the next section, we are going to describe how to create general shirt and convertible collars from collar girth. 4.2 Collar segmentations Collar girth is the path that collar segmentation sweep along in order to generate a collar surface. However, the appeared segmentation (Figure 15(b)) are not always consistent. They vary depending on the orientation and the position beyond the collar girth. Shapes of general shirt collar and convertible collar are defined according to wearing figure in practical. Mathematical formula is given to build its dimensions and segmentations. In the research, parameterized geometrical dimensions and orientations determine the original collar figures. It comprises two parts of user-defined parameters and internal parameters in order to provide unique collar externality in three dimensions. The user-defined parameter allows the designer to instinctively introduce pivotal lengths based on their professional knowledge. On the other hand, the internal parameter maintains fundamental 3D figure of the collar. 4.2.1 Convertible collar. In general, variation of convertible collar frequently appears in garment design. Three parameters are given from the user designer as shown in Figure 13. Upper and lower limits of these three user-defined parameters are preset so that the users can easily uncover it through dragging dialog bars. l1 notates the rear collar height (FG in Figure 8), l2 represents rear collar width (the length of FG þ FH ), and variable angle w (Figure 14) is the wide-open angle from sigittal to curve A-A0 -E defined in Section 4. The internal presets are used to retain the profile of fundamental convertible collar. It includes collar waistline curve, L1(ui), collar width variation, L2(ui) (Figure 15(a)), and the various angles, a1 and a2, between collar band and collar face (Figure 15(b)).
IJCST 15,2
98
Figure 12. Top views of shirt and convertible collar girths
3D collar design creation
99 Figure 13. Convertible collar dialog
Figure 14. The wide-open angle w
It was initially defined that
ui 2 w p þ a1 ðu0 Þ; a1 ðui Þ ¼ ða1 ðun Þ 2 a1 ðu0 ÞÞ sin p2w 2
i ¼ 1; 2; . . .; n 2 1 ð5Þ
and
ui 2 w p a2 ðui Þ ¼ ða2 ðun Þ 2 a2 ðu0 ÞÞ sin þ a2 ðu0 Þ; p2w 2 L1 ðui Þ ¼
3 X
a j ui j ;
i ¼ 0; . . .; n
i ¼ 1; 2; . . .; n 2 1 ð6Þ ð7Þ
j¼0
Rewrite equation (7) in matrix form L1 ¼ QA ) A ¼ Q21 L1 and
ð8Þ
IJCST 15,2
100
Figure 15. Internal presets of half convertible collar
L2 ðui Þ ¼
3 X
b j ui j ;
i ¼ 0; . . .; n
ð9Þ
j¼0
To be a fundamental convertible collar, b0 ¼ constant; and b1 ¼ b2 ¼ b3 ¼ 0: Figure 16 shows the screen displays of the user-defined values as l1 ¼ 50 mm; l2 ¼ 100 mm; and w ¼ 98: 4.2.2 Shirt collar. The major difference between convertible collar and shirt collar is that the shirt collar consists of two patterns instead of one complete pattern as in convertible collar. The prolonged collar band length provides space for button and buttonhole as shown in Figure 17. Therefore, in 3D profile, the collar band of shirt collar is closer to human neck than convertible band for the sake of necktie applying. Figure 18 shows the notations of user-defined parameters, such as rear collar band height l1, rear collar wing width l2, front collar width l3, and wide-open angle w. The preset formulas are given collar band height L1(ui), and collar wing width L3 ðui Þ 2 L2 ðui Þ as shown in Figure 19(a). Define L1(ui) and L3(ui) as L1 ðui Þ ¼ l1 ; 3 P L1 ðui Þ ¼ a j ui j ; j¼0
p=3 # ui # p 2p=6 # ui # p=3
ð10Þ
3D collar design creation
101
Figure 16. Four views of the given convertible collar
Figure 17. Shirt collar
Figure 18. Shirt collar dialog box
IJCST 15,2
102
Figure 19. Half shirt collar parameters
ui 2 ul ðl3 2 ðl2 þ l4 ÞÞ þ l2 þ l4 ; L3 ðui Þ ¼ w 2 ul C 1 ðL3 ; ui Þ ¼ ðl2 þ l4 ; ul Þ;
i ¼ 0; . . .; ul
ð11Þ
C 2 ðL3 ; ui Þ ¼ ðl3 ; 0Þ
Curves C1 and C2 are defined by L3(ui) and ui, which determine the expansion of the collar wing area near the tip of the collar wing. ui 2 w p þ a1 ðu0 Þ; i ¼ 1; 2; . . .; n 2 1 a1 ðui Þ ¼ ða1 ðun Þ 2 a1 ðu0 ÞÞ sin p2w 2 ð12Þ ui 2 w p þ a2 ðu0 Þ; a2 ðui Þ ¼ ða2 ðun Þ 2 a2 ðu0 ÞÞ sin p2w 2
i ¼ 1; 2; . . .; n 2 1 ð13Þ
It is given that l1 ¼ 30; l3 ¼ 40; and l4 ¼ 60 mm; then the outcomes of the four-view are shown in Figure 20.
3D collar design creation
103
Figure 20. Shirt collar
4.3 Comparison For the purpose of examining the computer-designed 3D collars, in this section, we compare the area from 3D collar design and its collar pattern based on the given parameters described in the previous sections. The collar patterns are drawn on a quadrille paper by the given internal and external (user-defined) parameters accordingly. Table I shows the comparison results of the convertible and shirt collars. There are two possible reasons that cause the differences. (1) The wearable collars are practically generated from 2D pattern. However, a 3D B-Spline collar is inflatable due to tensible essence of cloth material. (2) The lines drawn on quadrille paper approximate pattern edges that contain incomplete squares which are not able to precise calculation. However, the minor differences can be ignored since it does not affect the shape of the 3D designs. To design the basal collars by merely setting parameters may not completely fit in the diversity of collar styles. Hereby, we provide refashionment tool in order to produce changeable collar styles described in the next section. 5. Refashion mechanism and results Based on B-Spline curves and surfaces techniques, 3D collar is generated and described in previous sections. Through the adaptable control points on B-Spline curve or surface, surface profile is multifarious. For the reason of
Collar type
Dimensions of 3D collar
Dimensions of pattern
Difference (percent)
Convertible Shirt
255 cm2 141 cm2
249 cm2 137 cm2
2.41 2.92
Table I. Comparisons of 3D collars and patterns
IJCST 15,2
104
reasonable refashionment, refashioned tool in the kernel allows the users the control point selections by either a single point or a series of sectional points. Meanwhile, the thereabout movement of these control points are separated into instinct two-way directions such as forward/backward, left/right traverse, upward/downward, and radial, and planar directions such as horizontal plane, transversal plane, and saggital plane of mannequin base accordingly. Concern about the collar variant, it is demanded to retain coherence of the collar girth connecting both collar and top, the modifiable components contain three parts, collar wing, collar band, and collar girth. The former two belong to B-Spline surface, whereas, the latter one is B-Spline curve. Refashionment on collar girth would affect the shape of collar band and top synchronously. There are several reasonable restrictions for shape refashionment. According to the infrastructure of collars, convertible collar and shirt collar are given in nearly similar constraints. Shirt collar constraints . Sectional curves of rear collar band and rear collar wing, notates FG and FH in Section 4, are not allowed to move left or right, as well as radiation. Since the collar is designed in symmetry in this stage, these constraints would avoid doubling between either side of half collar located in left- and right-hand-side. . In order to retain C0 continuity between collar band and collar wing, modification in collar waistline would renew both collar band and collar wing synchronously. Convertible collar constraints . Convertible collar constraints are same as shirt collar constraints, only two more constraints are added in. Collar wing is not allowed to modify except the edge along wing curve, in terms of curve E-D-H in Figure 8. Modification process is shown in Figure 21 that demonstrates the selections of single control point (a) and a series of sectional control points (b). Starting from the user-defined collar type and its relevant parameters; choose specific modifiable component and its movement modes. The component control points are displayed in dots for selection by ways of mouse select-drag-and-drop. After a sequence of modification processes, the refashioned convertible collar and shirt collar are shown in Figures 21 and 22, respectively.
6. Conclusions In this research, we employ the object-oriented methodology and geometric modeling theory to develop 3D clothing design software. The developed software using object-oriented design technique can be easily expanded and
3D collar design creation
105 Figure 21. Tailor process of collar wing curve
Figure 22. An alterable (a) convertible collar and (b) shirt collar
inherited by the succeeding researchers. Based on OOT, the interface and the kernel of software are developed. The user interface provides the communication mechanism of human-machine interface. In turn, it would send commands to the kernel for further calculation. In addition, the kernel addresses the computation mechanism that contains internal and external mathematical formulas in it. Most of the apparel design software available in the market are based on 2D pattern maker which are not substantially created in three dimensions. Based on anthropometric and tailor techniques, individual structuralized body (Cheng, 2001) is created automatically. Stand on these structuralized lattices; garment can be generated without penetrating human body. In this research, we demonstrate that parameterized models possibly carry out garment design in truly three dimensions. Achieving 3D design and navigation would assist the garment industry enter spatial revolution. Based on human neck girth, the preliminary work of this research exhibits two fundamental styles of designed collar. Design interface provides not only
IJCST 15,2
106
parameterized setting but also refashionment for collar features. Primary formative methodology of elemental collars is accomplished as a template for other garment components design in the next stage. The undergoing research is working on ultimate collars creation. Except the basic convertible and shirt collars, changing the values set in the given parameters could luxuriously develop diverse collars, such as Madarin collar, choker collar, Dutch collar, sailor collar, open collar, ring collar, etc. Meanwhile, based on the mathematical methodology of convertible and shirt collar creations, additional 3D garment components are built easily. They are toupee, sleeve, skirt, dress, trouser, pocket, etc. Objects will be built in the software according to the OO methodology. Flatten the created 3D garment component to produce patterns is possible, is one of the crucial issues of undergoing work. Meanwhile, texture mapping on 3D garment components without print cycle distortion is also under investigation. References Cheng, K.H. (2001), “Feature extraction from 3D human body’s data point and construction of computer mannequin”, Master thesis, National Cheng Kung University, Taiwan. Computer Design Inc. (1992), “3D design concept reference manual”, Vol. 1 and 2, 1992-1995. DoCAD Ltd (2001), “DoCAD drawing software reference manual”, Taiwan. Geoffrey E. Macpherson Ltd (1938), http://www.macphersons.co.uk Gerber Technology Company (1969), http://www.gerbertechnology.com/ Hadap, S., Bangarter, E., Volino, P. and Magnenat-Thalmann, N. (1999), “Animating wrinkles on clothes”, IEEE Visualization ’99, http://miralabwww.unige.ch/newMIRA/ IEEE Computer Society Press, San Francisco, USA, pp 175-82. Hong, S.S. (1997), “Shih-Shien Drapping”, Shih Chien University, Taiwan. Pad System Inc. (1988), http://www.padsystem.com Terzopoulos, D., Platt, J., Bar, A. and Fleischer, K. (1987), “Elastically deformable models”, Proceedings of SIGGRAPH on Computer Graphics, Vol. 21, pp. 205-14. Volino, P. and Magnenat-Thalmann, N. (1997), “Developing simulation techniques for an interactive clothing system”, Proc.VSMM’97, Geneva, Switzerland, pp.109-18, http:// miralabwww.unige.ch/newMIRA/ Yang, Y., Thalmann, N.M. and Thalmann, D. (1992), “3D garment design and animation – a new design tool for the garment industry”, Computers in Industry, North-Holland, Amsterdam, Vol. 19, pp. 185-91. Further reading DB ARTIST Inc., http://www.dbhk.com/main1.html Gomes, J., Darsa, L., Costa, B. and Velho, L. (1998), “A system’s architecture for warping and morphing of graphical objects”, Proceedings of SIGGRAPH International Symposium on Computer Graphics, Image Processing, and Vision, pp. 192-9. Lectra Ltd (1973), http://www.lectra.com/
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Forecasting women’s apparel sales using mathematical modeling
Forecasting women’s apparel sales 107
Celia Frank and Ashish Garg Philadelphia University, Philadelphia, PA, USA
Amar Raheja California State Polytechnic University, Pomona, CA, USA
Les Sztandera Philadelphia University, Philadelphia, PA, USA Keywords Apparel, Forecasting, Computing, Time series, Modelling Abstract Traditionally, statistical time series methods like moving average (MA), autoregression (AR), or combinations of them are used for forecasting sales. Since these models predict future sales only on the basis of previous sales, they fail in an environment where the sales are more influenced by exogenous variables such as size, price, color, climatic data, effect of media, price changes or campaigns. Although, a linear regression model can take these variables into account its approximation function is restricted to be linear. Soft computing methods such as fuzzy logic, artificial neural networks (ANNs), and genetic algorithms offer an alternative taking into account both endogenous and exogenous variables and allowing arbitrary non-linear approximation functions derived (learned) directly from the data. In this paper, two approaches have been investigated for forecasting women’s apparel sales, statistical time series modeling, and modeling using ANNs. Four years’ sales data (1997-2000) were used as backcast data in the model and a forecast was made for 2 months of the year 2000. The performance of the models was tested by comparing one of the goodness-of-fit statistics, R 2, and also by comparing actual sales with the forecasted sales of different types of garments. On an average, an R 2 of 0.75 and 0.90 was found for single seasonal exponential smoothing and Winters’ three parameter model, respectively. The model based on ANN gave a higher R 2 averaging 0.92. Although, R 2 for ANN model was higher than that of statistical models, correlations between actual and forecasted were lower than those found with Winters’ three parameter model.
Introduction What is forecasting? Forecasting is ubiquitous; nearly everyone, in almost every walk of life, forecasts to some extent. A forecast is a probabilistic estimate of a future value. The underlying assumption in most forecasting methods is that the past patterns or behavior will continue in the future. It is commonly said that a good forecast requires a good “backcast”; patterns of the past are modeled and those patterns are projected into the future. This research has been supported by the United States Department of Commerce/National Textile Center Grant IP0-P10. Additional research support has been provided by Mothers Work, Inc.
International Journal of Clothing Science and Technology Vol. 15 No. 2, 2003 pp. 107-125 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310470097
IJCST 15,2
108
Figure 1. Forecasting procedure
For many organizations, their investment in forecasting has an immediate, as well as, long-term impact on profitability, customer service, productivity, etc. A good forecasting system is essential in avoiding problems such as inventory shortages and excesses, missed due dates, plant shutdowns, lost sales, lost customers, expensive expediting, and missed strategic opportunities. Forecasting process The process of forecasting can be relatively simple or complex depending on the situation. A typical forecasting procedure is shown in Figure 1 and involves the following. . Identify the problem. There is a need to establish clearly the targeted variable that is to be forecasted, e.g. the demand of a particular product as a function of time. . Assemble the data. Past data must be carefully collected to find any patterns (as well as randomness) associated with the targeted variable. . Formulation of the model. Based on an analysis of the data, a hypothetical model is formulated that includes those factors that influence the targeted
.
.
.
variable. For example, the demand for a particular garment can be Forecasting seasonal and trending, a function of selling price, influenced by women’s apparel advertising, etc. sales Execution of the model. Based on the hypothesis, one or more models are then applied to the real data. The data set is generally divided into two subsets: one subset is used to formulate the model and the 109 other subset is used to test how well the model performs in predicting unseen data. Analysis of the result. By performing statistical significance tests, and appropriate error measurements, the model may be accepted, modified or rejected. Ongoing improvements. A model must be constantly monitored for its performance and improved whenever unacceptable deviations emerge.
Classification and overview of forecasting methods In general, forecasting methods are divided into three categories: (1) univariate; (2) multivariate; and (3) qualitative. Univariate methods ( DeLurigo, 1998). Univariate modeling techniques, generally use time as an input variable with no other outside explanatory variables; this forecasting method is often called time series modeling. For example, a simple seasonal model might be, Y t ¼ Y t212 þ ðY t2l 2 Y t213 Þ; where Y t ; Y t2l ; Y t212 ; Y t213 are sales in weeks t, t21, t 2 12; and t 2 13; respectively. A few commonly employed methods in time series models are as follows. . Moving averages. Time series are smoothed using moving averages that reduce the period-to-period variation; local movements above and below a long-run mean are tracked. . Exponential smoothing (EXPOS)/Holts-Winters. Time series are smoothed in such a way that the most recent observations receive greater weight. Advanced methods incorporate decomposition to explain trend and seasonality. . Fourier series. This method models trend, seasonality, and cyclical movements using trigonometric sine and cosine functions. This method is used in automated forecasting systems, however, it is not without its detractors. . ARIMA (Box-Jenkins). This method models a series using trend, seasonal, and smoothing coefficients that are based on moving averages, autoregression, and difference equations. In this approach, a user is not
IJCST 15,2
110
constrained to a single model and, hence, to a single specific functional form, but can select from a wide range of models (Newbold, 1983). Models using time series methods can help managers make prudent business decisions, but they exclude causal relationships, and they should not be followed automatically without consideration of other business factors (Wheelwrigth and Makridakis, 1997). Causal/Multivariate methods. Explanatory models are used to establish “cause and effect” relationships in a system, for example, sales as a function of price, advertising, competition, and so forth. Although, multivariate methods are often used to model “cause and effect”, many multivariate models are focused primarily on the accuracy of the forecast. A few commonly used causal models are listed below. . Multiple regression approach. This approach uses the method of least squares, to model a relationship between one dependent and many independent variables. From a causal standpoint, multiple regression models may not be as valid as those of econometric. Nonetheless they may forecast as accurately. . Econometric method. In this method, using generalized least squares techniques, relationships between one or more endogenous and exogenous variables are estimated. Small-scale, simple models are multiple regression models; however, the theoretical foundation of econometric models is more rigorous. Mutual causality using several simultaneous equations can be modeled with econometric methods. . Multivariate ARIMA (Box-Jenkins-MARIMA) method. This method combines the strengths of the econometric and ARIMA time series methods. It is quite effective in applications when the effects of the independent variables lead one or more dependent variables. Qualitative methods. Qualitative methods of forecasting include Delphi, market research, panel consensus, historical analogy, etc. These suggestive qualitative methods are most frequently used to make long-run predictions when there is little objective data concerning relevant past patterns or relationships. Qualitative methods are useful when there is minimal data to support quantitative methods. In business, they are used to predict the demand for new products, new technologies, new market shares, the cost or development time for new products or technologies, or the best competitive strategy. A few commonly used qualitative models are given below. . Delphi. It is an iterative process in which experts respond to questionnaires, and the results are subsequently tabulated and modified to reach conclusions.
.
.
.
Panel consensus. This method is based on the assumption that the Forecasting consensus of several experts will yield a better forecast than that from one women’s apparel expert. The opinions of complementary experts yield improved sales predictions. Historical analogy. This method models a time series using a similar event from the past and is useful with new products and emerging technologies 111 where there is no past data. Soft computing. Methods based on soft computing mimic some of the parallel processing capabilities of the human brain to model both simple and complex situations. These models can identify non-linear and interactive relationships, which were anticipated by the analyst. Common methods include artificial neural networks (ANNs), fuzzy logic, and genetic algorithms.
Apparel sales forecasting In the present world, all industries need to be adaptable to a changing business environment in the context of a competitive global market. To comply with higher versatility and disposability of products for consumers, firms have adopted new forms of production behavior with names such as “just-in-time” and “quick response” (Vorman et al., 1998). To react to worldwide competition, managers are often required to make wise decisions rapidly. In fact, they must often anticipate events that may affect their industry. More than 80 percent of the US textile and apparel businesses have indicated an interest in time-based forecasting systems and have incorporated one or more of these technologies into their operations (Kincade et al., 1993). To make decisions related to the conception and the driving of any logistic structures, industrial managers must rely on efficient and accurate forecasting systems. Better forecasting of production, predicting in due time a sufficient quantity to produce, is one of the most important factors for the success of a lean production. Present research Forecasting garment sales is a challenging task because many endogenous as well as exogenous variables, e.g. size, price, color, climatic data, effect of media, etc. are involved. Our approach is to forecast apparel sales in the absence of most of the above-mentioned factors and then use principles of fuzzy logic to incorporate the various parameters that affect sales. The forecasting modeling investigates the use of two statistical time series models, seasonal single exponential smoothing (SSES) and Winters’ three-parameter model. It then investigates soft computing models using ANNs. A foundation has been made for multivariate fuzzy logic based model by building an expandable database and a rule base. After a substantial amount of data is collected, this model can be used to make predictions for sales specific to a store, color, or size of garment.
IJCST 15,2
112
Database A US-based apparel company has been providing sales data for various types, or classes, of apparel for the current and previous years. Data format From January 1997 until February 2001, sales data were collected for each day for every class for every store. From March 2001 onwards, more detailed information about each garment is available including its size and color. The data for the previous years (1997-2000) has much fewer independent variables as compared to the data for the current year, 2001. This is evident from Tables I and II. Row 1 of Table I is read as: 1 unit of Class X sold for $19.90 in Store_1 on 28 June 1997. Row 2 of Table I is read as: 1 unit of Class Y returned in Store_2 on 28 September 1998. Similarly, Row 1 of Table II is read as: 1 unit of Class X of size M, color 1 sold for $19.90 in Store_1 on 28 July 2001. Building of database Raw data available in the formats shown in Tables I and II was processed to form a database, which was subsequently used as input for time series analysis and a fuzzy logic based multivariate forecast model. Using the data from January 1997 till March 2000, time series analysis was performed. A data file for each class that contains information about its total sales (in dollars) each day was prepared. For example, input file for class X looks as shown in Table III. Total sales in terms of number of units would have been a better option but it was not possible since the number of units sold was missing from many rows in the raw data. From March 2001 onwards, aggregation was performed in several steps. Initially, from daily sales files, a database containing information about total
Table I. Sales data format for 1/1997-2/2001
Units
Price
Class
Store
Date
1 21
19.9 19.9
X Y
Store_1 Store_2
28/6/1997 28/9/1998
Base
Table II. Sales data format for 3/2001 onwards
Base_1 Base_2
Color
Size
Units
Price
Class
Store
Date
Label
QOH
1 49
M XL
1 1
19.9 19.9
X Y
Store_1 Store_2
28/7/2001 8/01/2001
Label_1 Label_2
0 0
Notes: Base: garment description; label: manufacturing division identity; QOH: quantity on hold.
sales was formed for a particular garment-class for each color, each size, and Forecasting each store. For example, input file for class X is shown in Table IV. women’s apparel The total number of rows in this format was too large to be used as an input sales file for the multivariate model. In order to reduce the dimensionality of the database, two compressions were performed. First, daily sales data were converted into monthly sales data. Secondly, color information was compressed 113 by aggregating sales of similar colors. For example, color code 23 represents light green, 25 dark green, 26 medium green. Instead of having different rows for different tones of green, we compressed this information into a single row by assigning color code 2 for all greens. Finally, input data file for multivariate fuzzy logic based model looks as shown in Table V. Pre-analysis Data were analyzed for trends and seasonality. This analysis helps in choosing an appropriate statistical model, although this kind of preparation is not necessary for soft computing based models. For pre-analysis, three classes were chosen, one each from the Spring, Fall and Non-seasonal garment categories. Analysis was spread among all the categories to remove any bias due to the type of class.
Date
Total sales ($)
28/6/1997 05/9/1998 14/3/1999 08/1/2000
Table III. Format of input data file for class X for time series analysis
2,378 1,405 3,546 5,983
Date
Size
Color
Store
Units
28/7/2001 28/7/2001 28/7/2001
M M XL
23 26 45
Store_1 Store_1 Store_2
6 2 7
Date
Size
Color
Store
Units
7/2001 7/2001
M XL
2 4
Store_1 Store_1
12 19
Table IV. Initial input data file for class X for multivariate model
Table V. Final input data file for Class X for multivariate model
IJCST 15,2
114
Weekly trend Sales data often showed a weekly trend with sales volume increasing during weekends as compared to weekdays. This trend was evident from the daily sales graphs for all the classes. Figure 2(a)-(c) shows this trend graphically for three classes, one each from Spring, Fall and Non-seasonal category. In all the figures, it can be seen that sales volume peaks on 4th, 11th, 18th, and 25th of January, which are all Saturdays, and generally second highest sales are on Sundays. This observation was further supported by qualitative means by calculating auto correlations functions (ACFs). ACF is an important tool for discerning time series patterns. ACF were calculated for 350 observations for all the three classes. ACF for a given lag k is given by equation (1): n X ðY t 2 Y ÞðY t2k 2 Y Þ ACFðkÞ ¼
t¼1þk n X t¼1
Figure 2. (a) Daily sales class AS for January 1997; (b) daily sales class AF for January 1997; (c) daily sales class CN for January 1997
ð1Þ ðY t 2 Y Þ2
It can be seen from Table VI that in every case there is a very high value of Forecasting ACF at lag 7. This suggests that in daily sales data, sales pattern repeats after women’s apparel every 7 days. sales Upon further analysis of the data, it was observed that fraction contributions towards total sales in any week of the year and for any class remain significantly constant. Hence, information about fraction contribution can be 115 used to forecast daily sales after forecasting weekly sales. Table VII and Figure 3 shows average fraction contribution towards total sales of a week. Annual trend Garment sales are generally seasonal with demand increasing for a particular type in one season and for a different type in another season. To investigate seasonality, the same methodology was used as was used to establish weekly trend. Both graphically as well as using ACF, it was shown that sales of all three classes under consideration show strong and distinct seasonal trend. Interestingly, class CN has been categorized as non-seasonal; still it showed increase in seasonality although not as distinctively as shown in other two classes. While investigating the annual seasonal trend, data were aggregated into weekly increments. Figure 4(a)-(c) graphically shows the seasonality of the three classes. Figure 5(a)-(c) graphically shows ACFs. All the graphs of Figure 5 are sinusoidal in shape. This reflects the relationships in the low order (i.e. 1 to 12 lags) and high order (i.e. 48 to 52 lags). The sinusoidal pattern in ACFs of Figure 5 is typical of many seasonal time series.
Lag ACF
1 0.653
2 0.502
ACFs for class AS 3 4 0.444 0.433
5 0.501
6 0.637
7 0.874
Lag ACF
1 0.767
2 0.676
ACFs for Class AF 3 4 0.685 0.659
5 0.650
6 0.757
7 0.87
Lag ACF
1 0.670
2 0.514
ACFs for class CN 3 4 0.486 0.470
5 0.490
6 0.655
7 0.871
Day Fraction (percent)
Sunday Monday Tuesday Wednesday Thursday Friday Saturday 13
10
11
11
13
18
24
Table VI.
Table VII. Fractions of weekly sales distributed among 7 days
IJCST 15,2
116
Figure 3. Fractions of weekly sales distributed among 7 days
Figure 4. (a) Weekly sales data for class AS; (b) Weekly sales data for class AF; (c) Weekly sales data for class CN
Methodology and results Evident from the format of the database for years 1997-2001 only sales information with respect to time for various garments is available. Hence, only univariate time series and soft computing models were investigated using this data.
Forecasting women’s apparel sales 117
Figure 5. (a) ACF (k) for class AS; (b) ACF (k) for class AF; (c) ACF (k) for class CN
From March 2001 onwards, much more elaborate sales data were available. Using this data set, a multivariate forecasting model was implemented which could prove useful for inventory maintenance. Six classes, two each from the Spring (AS and BS), the Fall (AF and BF), and the Non-seasonal (CN and DN) categories, were chosen for each model. They were built using 4 years sales data, and the next 2 months data were forecasted. The forecasted data were then compared with actual sales to estimate the forecasting quality of the model.
IJCST 15,2
118
Univariate time series model SSES. EXPOS is one of the most widely used forecasting methods. As discussed previously regarding seasonality in sales data, a single exponential smoothing (SES) model with a factor of seasonality was investigated. A SSES model requires four pieces of data: the most recent forecast, the most recent actual data, a smoothing constant, and length of the seasonal cycle. The smoothing constant (a) determines the weight given to the most recent past observations and therefore controls the rate of smoothing or averaging. The value of a is commonly constrained to be in the range of zero to one. The equation for SSES is: F t ¼ a At2s þ ð1 2 aÞF t2s
ð2Þ
where Ft is the exponentially smoothed forecast for period t, s the length of the seasonal cycle, At2 s the actual in the period t 2 s, Ft2 s the exponentially smoothed forecast of the period t 2 s, and a (alpha) the smoothing constant. Weekly sales data were used for the forecast model given by the above equation. Hence, s was chosen to be 52 (number of weeks in a year). There are many ways of determining alpha. Method chosen in the present work was based on minimum squared error (MSE). Different alpha values were tried for modeling sales of each class and the alpha that achieved the lowest SE was chosen. After choosing the best alpha value, a forecast model was built for each class using 4 years weekly data. Using the model, a weekly sales forecast was conducted for January and February of 2001. In order to forecast daily sales, the fraction contribution of each day (given in Figure 3) was multiplied by the total forecasted sales of each week. Figure 6(a) shows the actual versus fitted values of 3 years sales data and Figure 6(b) shows the actual versus forecasted daily sales for class AS. The data for the remaining five classes are available on the project Web site. Table VIII gives the alpha value, R 2 of the model, and correlation coefficient between the actual and forecasted daily sales from 3 January 2001 to 27 February 2001, for all classes. It can been seen that even with single parameter SSES, R 2 is on an average more than 0.75 implying that the model is able to explain 75 percent of the variation in the data. Correlation coefficients between the actual and forecasted sales from 3 January 2001 to 27 February 2001 are also quite high except for class DN. Winters’ three parameter EXPOS. Winters’ powerful method models trend, seasonality, and randomness using an efficient EXPOS process. The underlying structure of additive trend and multiplicative seasonality of Winters’ model assumes that: Y tþm ¼ ðS t þ bt ÞI t2Lþm
ð3Þ
Forecasting women’s apparel sales 119
Figure 6. (a) Actual vs fitted sales value for class AS using SSES model; (b) Actual vs forecasted sales value for class AS using SSES model
Class
AS
Alpha 1.4 R2 0.738 Corr. 0.906 Note: Correlation coefficient 27 February 2001.
BS
AF
BF
CN
DN
1.4 0.9 1.3 1.3 1.0 0.832 0.766 0.872 0.762 0.831 0.893 0.862 0.910 0.892 0.722 between the actual and forecasted sales from 3 January 2001 to
where St is the smoothed non-seasonal level of the series at the end of t, bt the smoothed trend for the period t, m the horizon length of the forecasts of Y tþm ; and I t2Lþm the smoothed seasonal index for the period t þ m: That is, Y tþm the actual value of a series, equals a smoothed level value St plus an estimate of trend bt times a seasonal index I t2Lþm : These three components of demand are each exponentially smoothed values available at the end of period t. The equations used to estimate these smoothed values are:
Table VIII. Values of alpha, R 2, and correlation coefficients for SSES model
IJCST 15,2
120
S t ¼ aðY t =I t2L Þ þ ð1 2 aÞðS t21 þ bt21 Þ
ð4Þ
bt ¼ bðS t 2 S t21 Þ þ ð1 2 bÞbt21
ð5Þ
I t ¼ g ðY t =S t Þ þ ð1 2 gÞI t2Lþm
ð6Þ
Y tþm ¼ ðS t þ bt mÞI t2lþm
ð7Þ
where Yt is the value of actual demand at the end of period t, a the smoothing constant used for St, St the smoothed value at the end of t after adjusting for seasonality, b the smoothing constant used to calculate the trend (bt), bt the smoothed value of trend for the period t, It2 L the smoothed seasonal index L periods ago, L the length of the seasonal cycle (e.g. 12 months or 52 weeks), g (gamma) the smoothing constant, for calculating the seasonal index in period t, It the smoothed seasonal index at end of period t, and m the horizon length of the forecasts of Y tþm : Equation (4) calculates the overall level of the series. St in equation (5) is the trend-adjusted, deseasonalized level at the end of period t. St is used in equation (7) to generate forecasts, Y tþm : Equation (5) estimates the trend by smoothing the difference between the smoothed values St and S t21 : This estimates the period-to-period change (trend) in the level of Yt. Equation (6) illustrates the calculation of the smoothed seasonal index, It. This seasonal factor is calculated for the next cycle of forecasting and used to forecast values for one or more seasonal cycles ahead. For choosing a (alpha), b (beta), and g (gamma) MSE was used as a criterion. Different combinations of alpha, beta, and gamma were tried for modeling sales of each class and the combination that achieved the lowest RSE was chosen. After choosing the best alpha, beta, and gamma values, the forecast model was built for each class using 4 years of weekly sales data. Using the model, a weekly sales forecast was done for January and February of 2001. In order to forecast daily sales afterwards, the fractional contribution of each day (given in Figure 3) was multiplied by the total forecasted sales of each week. Figure 7(a) shows the actual versus fitted values of 3 year sales data and Figure 7(b) shows the actual versus forecasted daily sales for class BS. This data for the five remaining classes are available on the project Web site. Table IX gives the alpha, beta, gamma, R 2 of the model, and the correlation coefficient between the actual and forecasted daily sales from 3 January 2001 to 27 February 2001. R 2 values for all the classes except BF are much higher than those obtained from SSES. Higher R 2 values and the ability of Winters’ model to better define this is due to the additional parameter beta utilized for trend smoothing.
Forecasting women’s apparel sales 121
Figure 7. (a) Actual vs fitted sales value for class BS using Winters’ model; (b) actual vs forecasted sales value for class BS using Winters’ model
Class
AS
BS
AF
BF
CN
DN
Alpha 0.60 0.50 0.50 0.80 0.60 0.50 Beta 0.01 0.01 0.01 0.01 0.01 0.01 Gamma 1.00 0.47 0.91 0.72 1.00 0.82 R2 0.923 0.969 0.951 0.685 0.941 0.933 Corr. 0.903 0.920 0.869 0.667 0.927 0.777 Note: Correlation coefficient between the actual and forecasted sales from 3 January 2001 to 27 February 2001.
Although curve fitting is very good, correlation coefficients between the actual and forecasted sales are not as high. On observing the graphs of the actual versus forecasted values for all the classes, it can be observed that trend (growth or decay) has always been over estimated and, hence, forecasted values
Table IX. Values for alpha, beta, gamma, R 2, and correlation coefficients for Winters’ model
IJCST 15,2
122
are too high or too low. As in any multiplicative model, the division by very small numbers or multiplication by extremely large values is a problem with equation (3) which could have resulted in overestimation of the trend. Soft computing methods Soft computing methods are a rapidly growing area of computer science. These methods are being used to solve many problems such as optimization, pattern classification, forecasting, learning, etc. ANNs, genetic algorithms, fuzzy logic based reasoning, etc. are some of the popular soft computing methods being used to solve many real world problems. In our current research, we have used ANN to learn the patterns of sales of garments in the past to forecast sales in the future. ANN model Neural networks mimic some of the parallel processing capabilities of the human brain as models of simple and complex forecasting applications. These models are capable of identifying non-linear and interactive relationships and hence can provide good forecasts. In our research, one of the most versatile ANNs, the feed forward, back propagation architecture was implemented. The architecture of the feed forward neural network is shown in Figure 8. The hidden layers are the regions in which several input combinations from the input layer are fed and the resulting output is finally fed to the output layer. {x1 ; . . .; xM } is the training vector and {z1 ; . . .; zM } is the output vector. The error E of the network is computed as the difference between the actual and the desired output of the training vectors and is given in equation (8): E¼
NL X P 1 X ðtð pÞ 2 sðnpÞ Þ2 PN L n¼l p¼1 n
ð8Þ
where tðnpÞ is the desired output for the training data vector p and sðpÞ n is the calculated output for the same vector. The updated equation for the weights of individual nodes in different layers is defined using the first derivative of the error E as given in equation (9):
Figure 8. Multilayer perceptron model used for forecasting sales
ðnÞ ðnÞ wðnÞ i ðlÞ ¼ wi ðlÞ þ Dwi ðlÞ;
where
DwðnÞ i ðlÞ ¼ 2h
›E ›wðnÞ i ðlÞ
ð9Þ
ANN consisted of three layers: input layer, hidden layer and output layer with 10, 30 and 1 neuron, respectively. Two hundred and seventeen weeks sales data were divided into three parts. The first part consisted of 198 weeks, which was used to train the network. The second part with 10 weeks data were used to test the network for its performance. The third part with 9 weeks data were used to compare forecasting ability of the network by comparing the forecasted data with the actual sales data. In order to forecast daily sales afterwards, the fraction contribution of each day was multiplied by the total forecasted sales of each week. Figure 9(a) shows the actual versus fitted values of 3 years sales data and Figure 9(b) shows the actual versus
Forecasting women’s apparel sales 123
Figure 9. (a) Actual vs fitted sales value for class CN using ANN model; (b) actual vs forecasted sales value for class CN using ANN model
IJCST 15,2
124
forecasted daily sales for class CN. This data for the five remaining classes are available on the project Web site. Table X gives the R 2 of the model, and correlation coefficient between the actual and forecasted daily sales from 3 January 2001 to 27 February 2001. R 2 values for all the classes are much higher than those obtained from SSES, and Winters’ model. High R 2 values and the strength of the ANN model are due to the ability of ANNs to learn non-linear patterns. Although curve fitting is very good, correlation coefficients between the actual and forecasted sales are not that good. This might be due to over learning of the network. A potential problem when working with noisy data is the so-called over-fitting. Since ANN models can approximate essentially any function, they can also overfit all kinds of noise perfectly. Typically, sales data have a high noise level. The problem is intensified by a number of outliers (exceptionally high or low values). Unfortunately, all three conditions that increase the risk of over-fitting are fulfilled in our domain and have impacted correlations. Conclusion Time series analysis seemed to be quite effective in forecasting sales. In all the three models, R 2 and the correlation coefficients were significantly high. The three parameter Winters’ model outperformed SSES in both explaining variance in the sales data (in terms of R 2 ) and forecasting sales (in terms of correlation coefficient). ANN model performed best in terms of R 2 among three models. But correlations between the actual and forecasted sales were not satisfactory. A potential problem when working with noisy data, a large number of inputs, and small training sets is the so-called over-fitting. Since big ANN models can approximate essentially any function, they can also over fit all kinds of noise perfectly. Unfortunately, all three conditions that increase the risk of overfitting are fulfilled in our domain. Typically, sales data have a high noise level. The problem is intensified by a number of outliers (exceptionally high or low values). A multivariate fuzzy logic based model could model the sales very well, as it would take into account many more influence factors in addition to time. This
Table X. Values of R 2, and correlation coefficients for ANN model
Class
AS
BS
AF
BF
CN
DN
R2 Corr.
0.963 0.878
0.941 0.906
0.953 0.704
0.906 0.793
0.953 0.914
0.916 0.845
Note: Correlation coefficient between the actual and forecasted sales from 3 January 2001 to 27 February 2001.
naturally leads to the first extension of this work. Extensions of the concept of Forecasting discovery learning are of current interest and are being investigated. women’s apparel References DeLurigo, S.A. (1998), Forecasting Principles and Applications, 1st ed., McGraw Hill, NY, USA. Kincade, D.H., Cassill, N. and Williamson, N. (1993), J. Text. Inst., Vol. 84 No. 2, p. 2. Newbold, P. (1983), Journal of Forecasting, Vol. 2, p. 28. Vorman, P., Happiette, M. and Rabenasolo, B. (1998), J. Text. Inst., Vol. 1 No. 2, p. 78. Wheelwrigth, S.C. and Makridakis, S. (1997), Forecasting Methods for Management, 2nd ed., John Wiley, New York.
sales 125
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
IJCST 15,2
126 Received June 2002 Accepted December 2002
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Factors during tumble drying that influence dimensional stability and distortion of cotton knitted fabrics L. Higgins, S.C. Anand, M.E. Hall and D.A. Holmes Faculty of Technology (Textiles), Bolton Institute, Bolton, UK Keywords Woven fabrics, Shrinkage, Physical properties, Cotton yarns Abstract The length and width shrinkages, skewness, spirality and moisture content of three weft knitted cotton structures, plain single jersey, interlock and lacoste, were determined at regular intervals during tumble drying. Significant length and width shrinkages occurred in all three structures with the amount of shrinkage increasing rapidly in plain single jersey and lacoste as their moisture contents fell below 30 per cent. Distortion was less affected by tumble drying. An attempt was made to isolate the effects of heat and agitation during tumble drying. It has been demonstrated that similar patterns of shrinkage and distortion occur whether heat is applied during tumble drying or not. The tumbling action in a tumble drier has the greatest influence on the dimensional stability and distortion of weft knitted cotton fabrics.
International Journal of Clothing Science and Technology Vol. 15 No. 2, 2003 pp. 126-139 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310470132
Introduction Since 1970, there has been growing consumer demand for cotton knitted fabrics in an increasingly diverse range of end uses (Burkett, 1986). Combined with the increasing and widespread use of domestic tumble dryers, this has led to recent consumer awareness of negative properties, such as dimensional changes, distortion etc. associated with cotton knits (Thomas, 1994). This investigation is concerned with understanding the dimensional instability and distortion in weft knitted cotton fabrics as they occur within a domestic tumble drying environment. It is an inherent consequence of the manufacturing processes that knitted fabrics and garments are left in a highly distorted state (Hepworth, 1989). As these fabrics relax, they tend towards a state of minimum total energy (Hepworth and Leaf, 1976; Munden, 1959). In order to achieve this state, the knitted loops within the fabric structure must change their shape to that of total minimum bending thus causing dimensional changes and distortion within the fabric or garment (Lo, 1989; Postle and Munden, 1967). Different states of relaxation can occur. Munden (1959) suggested that cotton knitted fabrics require wet relaxation to acquire their minimum energy state. Other authors have shown that reproducible and complete relaxation will only occur after cotton knitted fabrics have been washed and tumble dried five times (Burkett, 1986; Heap, 1984; Lo, 1989). Previous work has confirmed the need to use
tumble drying as opposed to line or flat drying to facilitate total relaxation and hence maximum shrinkage (Brown, 2000; Greenwood, 1986; MacKay, 1992). When wet cotton fibres swell, they do so mainly laterally causing changes in the loop shape, which alter the dimensional and shape retention properties of the knitted fabrics (Suh, 1967). As the fabric is dried below approximately 30 per cent moisture content, the cotton fibres deswell causing voids to appear between the fibres. The application of mechanical energy at this stage causes the voids to collapse and hence further shrinkage can occur (Leah, 1986). Experimental data, which indicate that the level of shrinkage significantly increases during tumble drying as the moisture content of cotton knitted fabrics falls below 30 per cent, have been reported. This occurs irrespective of whether heat is applied during tumbling (Leah, 1986) or not (Gordon et al., 1984). In this investigation, an attempt has been made to isolate and study the individual influences of moisture content, mechanical energy and heat on dimensional stability and distortion during tumble drying for a limited range of commercially available weft knitted cotton structures.
Factors during tumble drying
127
Experimental method The aim of the experimental design was to isolate the contributions of moisture content, agitation and heat to fabric shrinkage and distortion during tumble drying. To that aim, the wash cycle was kept constant throughout the experiment and three drying regimes were studied (as detailed in Table I). Moisture content, length shrinkage, width shrinkage, skewness and spirality were determined immediately after washing, at regular intervals during tumble drying and after 48 h in a standard atmosphere of 20 ^ 28C and 65 ^ 2 per cent relative humidity. Samples of three commercially finished, weft knitted, cotton fabrics (plain single jersey, interlock and lacoste, Table II) were obtained from an outside source. After conditioning for 48 h at 20 ^ 28C and 65 ^ 2 per cent relative humidity, specimens were prepared of an open pillowcase construction approximately 50 £ 50 cm2 : They were marked according to AATCC test method 150-1995 “Dimensional changes in automatic home laundering”. The same specimens were marked according to AATCC test method 170-1996 “Skewness change in fabric and garment twist resulting from automatic home Tumble dried at 65-758C
Tumble dried at 228C
Flat dried at 65-758C
Interrupted every 10 min Loads P(1), I(1), L(1) Loads P(2), I(2), L(2) Loads P(3), I(3), L(3) Interrupted every 5 min Loads P(4), I(4), L(4) – – Notes: P – Plain single jersey, I – Interlock, L – Lacoste. Numbers in brackets denote load number.
Table I. Summary of drying regimes
IJCST 15,2
128
laundering”. The open pillowcase construction of the specimens allowed for the determination of the angle of spirality. At present, there is no AATCC test method for spirality; the method used during this investigation is detailed in the Appendix. After preparation, specimens were left for further 48 h to condition before testing. In order to determine the moisture content of the specimens during testing, the moisture contents of the three fabrics were measured before testing. Small specimens of each fabric were conditioned for 48 h at 20 ^ 28C and 65 ^ 2 per cent relative humidity before being dried to a constant mass in an oven at 1058C. The specimens were washed in an AATCC approved Kenmore super capacity automatic washing machine (top loading). The specimens were washed on a normal fabric setting without detergent. The wash water temperature was approximately 428C and the rinse water temperature was approximately 228C. Both wash and rinse waters were made up to a water hardness level of 155 ppm by using a stock water hardness solution. Each wash load consisted of three specimens of any one type of fabric made up to a mass of 3.6 kg with ballast squares of similar construction. The specimens were dried in an AATCC approved Kenmore super capacity plus clothes dryer using three different regimes: (1) the full 3.6 kg load was dried on a normal cotton tumble drying cycle, temperature 65-758C, until constant mass was achieved – loads P(1), I(1), L(1) and loads P(4), I(4), L(4); (2) the full 3.6 kg load was tumble dried on an air dry setting, temperature 228C, until the specimens reached approximately normal regain (drying beyond normal regain was not possible under these conditions) – loads P(2), I(2), L(2); (3) the specimens were dried individually by placing flat on a stationary rack fitted within the tumble dryer and using a normal cotton setting, temperature 65-758C, until constant mass was achieved – loads P(3), I(3), L(3). The drying cycles were interrupted every 5 or 10 min, as detailed in Table I, for the measurement of length shrinkage, width shrinkage, skewness, spirality and moisture content. After drying, the specimens were placed in a standard Property
Table II. Fabric specifications
Machine gauge Stitch length Yarn linear density Tightness factor Fabric area density
Plain single jersey
Interlock
Lacoste
E28 2.60 mm 17.5 tex 1.61 126.7 g/m2
E28 2.66 mm 9.8 tex 1.18 153.3 g/m2
E20 3.26 mm (average) 31.7 tex 1.73 219.1 g/m2
atmosphere of 20 ^ 28C and 65 ^ 2 per cent relative humidity for 48 h before final sets of measurements were taken. Results and discussion Early analysis of the data obtained from loads P(1), I(1) and L(1) suggested a need to further clarify the effect of moisture content on dimensional stability and distortion in the knitted fabrics. Hence, further specimens of plain single jersey, interlock and lacoste were washed and tumble dried at 65-758C with interruptions for testing every 5 min (loads P(4), I(4), L(4)). The results were found to be similar to those obtained when the drying regime was interrupted every 10 min. However, when the drying cycle was interrupted and the specimens tested, more often, the collected data became more scattered. This was, particularly, apparent when examining the spirality results where the amounts concerned were relatively small.
Factors during tumble drying
129
Effect of moisture content on dimensional stability during tumble drying For plain single jersey, interlock and lacoste, the relationship between moisture content and length and width shrinkages during the three different drying regimes when determined at 10 min intervals (loads 1, 2 and 3) is shown in Figures 1-6. The behaviour of the different fabric structures during normal tumble drying conditions, that is at 65-758C, can be seen in the load (1) plots within these figures. As demonstrated in the previous work, all three fabrics exhibited length and width shrinkages after the wash cycle due to wet relaxation. The level of length
Figure 1. Length shrinkage vs moisture content during drying (fabric: plain single jersey)
IJCST 15,2
130
Figure 2. Width shrinkage vs moisture content during drying (fabric: plain single jersey)
Figure 3. Length shrinkage vs moisture content during drying (fabric: interlock)
Factors during tumble drying
131
Figure 4. Width shrinkage vs moisture content during drying (fabric: interlock)
Figure 5. Length shrinkage vs moisture content during drying (fabric: lacoste)
IJCST 15,2
132
Figure 6. Width shrinkage vs moisture content during drying (fabric: lacoste)
shrinkage seen in the interlock specimens was 8.2 per cent, well above the commercially acceptable level of 5 per cent, suggesting that the interlock sample used had been poorly finished. Length and width shrinkages increased during tumble drying, with plain single jersey and lacoste showing a significant increase in the rate of shrinkage as the moisture content decreased, particularly below 30 per cent. That the interlock fabric showed a more linear rate of increase of length and width shrinkage values was perhaps a consequence of its poor finishing. For all the three fabrics, length and width shrinkage continued when the fabrics were over-dried below the level of moisture content determined for the conditioned fabrics. When the specimens were left for 48 h in a standard atmosphere at 20 ^ 28C and 65 ^ 2 per cent relative humidity, there was a limited recovery from this excessive shrinkage, but never more than 1.7 per cent. For the three weft knitted cotton fabrics tested, the level of length and width shrinkage values found after tumble drying on a normal cotton cycle was above the commercially acceptable standard of 5 per cent. There was evidence that the different knitted structures had an influence on the levels of shrinkage observed. Plain single jersey and interlock suffered similar levels of shrinkage in the length and width direction. However, the level of width shrinkage in the lacoste fabric was lower (5.9 per cent) than the width shrinkages exhibited by the plain single jersey (8.7 per cent) and interlock fabrics (11.3 per cent) and it was approximately half the level of length
shrinkage found in lacoste (10.3 per cent). Previous work by Anand and Yanmaz (2000) has demonstrated that this is due to the presence of tuck stitches within the structure of lacoste which act to push the wales apart restricting the relaxation process in the width direction.
Factors during tumble drying
Effect of moisture content on distortion during tumble drying The relationship between moisture content and skewness and spirality in plain single jersey, interlock and lacoste during the three drying regimes when determined at 10 min intervals (loads 1, 2 and 3) is shown in Figures 7-12. All three fabrics exhibited some degree of skewness and spirality after the wash cycle due to wet relaxation. The influence of the knitted structure on distortion is evident immediately after washing. Plain single jersey exhibited commercially unacceptable levels of skewness (7.6 per cent) and higher degrees of spirality than interlock and lacoste after the wash cycle due to its unbalanced structure. The degree of skewness and spirality decreased slightly for all fabrics during tumble drying at 65-758C with no significant changes occurring as the fabrics were over-dried.
133
Isolating the effect of agitation and heat The results from the two drying regimes which isolated the components of agitation and heat in the tumble dryer are plotted in Figures 1-12, on the same axes with the data obtained from the normal tumble drying cycle. If, as suggested (Gordon et al., 1984; Leah, 1986), it is the input of mechanical energy
Figure 7. Skewness vs moisture content during drying (fabric: plain single jersey)
IJCST 15,2
134
Figure 8. Spirality vs moisture content during drying (fabric: plain single jersey)
Figure 9. Skewness vs moisture content during drying (fabric: interlock)
Factors during tumble drying
135
Figure 10. Spirality vs moisture content during drying (fabric: interlock)
or agitation that facilitates shrinkage and distortion during drying we would expect to see two things. Firstly, we would see the same trends occurring when the fabric specimens were tumbled at 228C and at 65-758C. Secondly, the levels of shrinkage and distortion imparted after the wash cycle should remain constant during flat drying at 65-758C.
Figure 11. Shewness vs moisture content during drying (fabric: lacoste)
IJCST 15,2
136
Figure 12. Spirality vs moisture content during drying (fabric: lacoste)
Whilst comparing the data it should be observed that, for the tumble drying at 228C, drying time was considerably extended from under 60 min for loads P(1), I(1), L(1) to approximately 240 min for loads P(2), I(2), L(2). The increased number of interruptions for testing caused data scattering. This is particularly evident in Figures 6-12 which deal with distortion. It was not possible to dry the specimens below approximately 10 per cent moisture content when using the tumble dryer at 228C and consequently, no data was obtained for the effect of over-drying. It should be noted that when flat dried at 65-758C (loads P(3), I(3), L(3)), the size of the load being dried was reduced from a 3.6 kg load containing three specimens and ballast to a single specimen. Hence, the vapour pressure within the drum was reduced causing more rapid drying times. The distribution of thermal energy would also change. Dimensional stability When plain single jersey fabric was tumble dried at 228C, the patterns of length and width shrinkages were similar to those observed when tumble dried at 65-758C (Figures 1 and 2) with a very significant increase in the rate of shrinkage as the moisture content fell below 20 per cent. Flat drying plain single jersey specimens at 65-758C gave a much flatter trend line with lower overall levels of shrinkage. Similarly, the data obtained from lacoste shows good correlation between tumble drying at 228C and 65-758C (Figures 5 and 6). A flatter trend line is seen for length shrinkage when lacoste was flat dried at 65-758C with the overall level of shrinkage being lower than when the specimens were tumbled during drying (Figure 5). Width shrinkage also
increased during flat drying of lacoste specimens at 65-758C but the rate of increase was lower than that achieved when specimens were tumbled during drying. The data on length and width shrinkage values for interlock (Figures 3 and 4) shows fewer differences between the three drying regimes, although the overall level of shrinkage resulting from flat drying at 65-758C was lower than those resulting from the regimes that involved tumbling. Distortion The scattering of data when testing during tumble drying at 228C is immediately evident when examining Figures 7-12. Tumble drying at 228C produced similar patterns of skewness and spirality in plain single jersey and interlock (Figures 7-10), with similar overall levels of distortion at the end of the drying cycle. For lacoste, skewness and spirality increased when the specimens were tumbled at 228C whereas they decreased when tumbled at 65-758C, however, the overall changes were small. In all cases, except for skewness in the lacoste fabric, a much flatter trend line was obtained when the fabrics were flat dried at 65-758C compared to either of the tumble drying regimes. In the case of lacoste, the trend for skewness (Figure 11) was similar to that obtained in tumble drying; however, it should be noted again that the changes involved were small. Conclusions It has been demonstrated that tumble drying plain single jersey, interlock and lacoste cotton weft knitted structures under a normal cotton cycle causes significant length and width shrinkages with the amount of shrinkage increasing sharply in plain single jersey and lacoste as their moisture contents fall below 30 per cent. This confirms the work reported by Leah (1986). Distortion was less affected by tumble drying, the levels of skewness and spirality decreased slightly in all fabrics when tumbled under normal conditions. The knitted structure has been shown to have an effect on the dimensional stability and distortion during laundering. The balanced structure of interlock and the single jersey lacoste structure both exhibited lower levels of distortion than plain single jersey structure. The presence of tuck stitches in lacoste reduces the level of width shrinkage compared to the other two structures, thus confirming the work carried out by Anand and Yanmaz (2000). An attempt has been made to isolate the effects of agitation and heat during tumble drying. Tumble drying without heat, that is at 228C, caused similar overall levels of shrinkage and distortion to that determined when tumble drying at 65-758C. Crucially, the level of length and width shrinkage in plain single jersey and lacoste increased rapidly as their moisture contents fell below 20 per cent, whether heat was applied or not. The over-riding feature of the plots obtained for flat drying at 65-758C is their flatness when compared with those obtained for either of the regimes that
Factors during tumble drying
137
IJCST 15,2
involved tumbling. The fact that shrinkage and distortion did not substantially increase when there was no tumbling action present reinforces the conclusion that it is the tumbling action in a tumble dryer that has the greatest influence on the dimensional stability and distortion of weft knitted cotton fabrics.
138
References Anand, S.C. and Yanmaz, Y. (2000), “Some properties of single jersey weft knitted structures”, Melliand Textilberichte, pp. 181-4. Brown, K.S.M. (2000), “The effect of laundering on the dimensional stability and distortion of knitted cotton fabrics”, PhD thesis, Bolton Institute. Burkett, F.H. (1986), “The starfish project: an integrated approach to shrinkage control in cotton knits. Part 1 – introduction and definition of the reference state”, Knitting International, pp. 40-2. Gordon, B.W., Bailey, D.L., Jones, B.W., Stone, R.L. and Noell, R.D. (1984), “Shrinkage control of cotton knits by mechanical techniques”, Textile Chemist and Colourist, Vol. 35, pp. 25-7. Greenwood, P.F. (1986), “Influence of the laundering treatment”, Knitting International, Vol. 93, pp. 58-60. Heap, S.A. (1984), “Computer predictions for cotton shrinkage”, Knitting International, pp. 23-5. Hepworth, B. (1989), “Investigating the dimensional properties of 1£1 rib fabrics”, Knitting International, Vol. 96, pp. 48-51. Hepworth, B. and Leaf, G.A.V. (1976), “The mechanics of an idealised weft-knitted structure”, Journal of the Textile Institute, Vol. 67, pp. 241-8. Lau, Y., Tao, X. and Dhingra, R.C. (1995), “Spirality in single-jersey fabrics”, Textile Asia, pp. 95-102. Leah, R.D. (1986), “The starfish project: an integrated approach to shrinkage control in cotton knits. Part 5 – finishing to achieve specification”, Knitting International, pp. 96-9. Lo, T.Y. (1989), “The starfish project: cotton knitgoods shrinkage”, Textile Asia, pp. 134-8. MacKay, C. (1992), “Effects of laundering on the sensory and mechanical properties of 1 £ 1 Rib knitwear fabrics”, MPhil thesis, Bolton Institute. Munden, D.L. (1959), “The geometry and dimensional properties of plain knit fabrics”, Journal of the Textile Institute, Vol. 50, pp. 448-71. Postle, R. and Munden, D.L. (1967), “Analysis of the dry-relaxed knitted-loop configuration. Part 1 – two-dimensional analysis”, Journal of the Textile Institute, Vol. 58, pp. 329-51. Suh, M.W. (1967), “A study of the shrinkage of plain knitted cotton fabric, based on the structural changes of the loop geometry due to yarn swelling and deswelling”, Textile Research Journal, Vol. 37, pp. 417-31. Thomas, K. (1994), “From liable to stable: dimensionally stable knitted fabrics today”, Knitting Technique, Vol. 16, pp. 360-1. Appendix Spirality is a distortion of circular knitted fabrics, such as those used in this investigation. Residual torque is present in twisted yarns used in these knitted structures. As this torque is released, for instance whilst laundering, the yarns tend to untwist causing one side of the knitted loop to lift out of the plane of the fabric. The consequent distortion of the fabric is most commonly seen as seam displacement or twist (Lau et al., 1995) (Figure A1).
Factors during tumble drying
139
Figure A1. Measurement of spirality When preparing specimens, an open pillowcase construction, with two side seams, was produced by overlocking three of the outside edges of two fabric squares. After laundering, the seams twist as seen by the displacement of the seams at the bottom edge of the specimen. S1 and S2 are measured from the original seam to the seam created after laundering. L is the length of the specimen after laundering. Spirality Angle ðDegreesÞ ¼ tan21 u where tan u ¼ ðS 1 þ S 2 Þ=2L:
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
IJCST 15,2
140 Received February 2002 Revised September 2002 Accepted September 2002
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Modelling the changing apparel supply chain R.P. Rollins University of Huddersfield, Huddersfield, UK
K. Porter and D. Little Liverpool John Moores University, Liverpool, UK Keywords Apparel, Case studies, Modelling, Production planning and control, Supply chain Abstract This paper describes a research project funded by the Engineering and Physical Sciences Research Council. Twenty case-study companies operating across a range of industrial sectors participated in the project. Sectors chosen for the development of these architectures were those where the use of the traditional manufacturing resource planning (MRPII) model is not the optimum operating solution. In particular, the paper describes the process mapping and analysis approach applied to the study of a sector-based group of apparel manufacturing companies who collaborated in the research. The planning issues that confront the companies, the control solutions they employ in response to their present commercial environment as they seek to address the changing demands being made of the industry are outlined. A generic planning and control reference architecture developed from the study for the apparel sector is presented.
International Journal of Clothing Science and Technology Vol. 15 No. 2, 2003 pp. 140-156 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310470141
Introduction In recent years the key focus for many European apparel manufacturing companies has been to optimise the cost structure of their supply chains to enable the cost-effective delivery of goods to the point of sale. Driven by competitive pressures to reduce unit item price, the initiative has been to search out sources of low cost direct labour and then to subsequently downsize in-house manufacturing activity. The ensuing re-engineering programmes have been very successful. By the late 1990s most of the larger retailers and many leading manufacturers are purchasing high volumes of completed garments from low cost centres of production. The result of this large-scale move to out-sourcing has been competitive levelling. Price, once under control like quality before it, is no longer the single critical competitive issue for many producers. Changing business factors, which include the new demands being placed upon the industry by changing consumer demographics and a realisation by many western countries of the need to revitalise their manufacturing industries is exerting new pressures on the industry. For apparel businesses to operate competitively these pressures necessitate a modification to the existing cost optimising supply model to one which is more responsive to the customer service. Recent research has shown that one key factor for effective manufacturing responsiveness is the performance of the manufacturing control function.
Whilst there is a wide range of generic proprietary software available that Modelling the meets a company’s planning and scheduling requirements with varying changing apparel degrees of success, few of these products are designed for the apparel industry. supply chain The difficulty experienced by many firms lies in understanding the match between their business needs and the capabilities of the software. A poor match typically limits the ability of a firm to respond to market changes and is likely 141 to inhibit company growth. This is a particular problem for the small and medium enterprise (SME) which are common in this industry. The research project covered by this paper is aimed at developing novel planning and control reference models for industrial sectors where the traditional manufacturing resource planning (MRPII) paradigm is not a good fit. The paper gives an overview of the research method developed, explains the data capture method used for the case study companies and the use of extended event process chains and planning and control process models which underpin the development of the supply chain orientated reference model for the sector. Current status of the apparel manufacturing industry Despite a formidable challenge from the newly industrialised nations, which include low cost producers like China and India, garment manufacture remains a significant part of the trading base of the United Kingdom and the European Union as a whole. European statistics for the garment industry (EU Document, 1995) describe the sector as comprising of 80,500 companies employing 1.2 million people or 3.4 per cent of the total EU workforce. UK is the fourth largest garment producer within the EU, contributing 10 per cent of total product. However, even with a recorded 11 per cent increase in garment consumption across all EU states over the period 1983-94, making them the world’s largest clothing consumer (op. cit.) there has been a decline in the numbers employed within the European garment manufacturing industry. This is due in part to industry having relatively low barriers to entry. Even in those sectors that require higher technology inputs, the costs of business start-up are lower than in many other industries. This, along with a strong and still growing domestic demand has made garment manufacturing very attractive to developing countries wishing to expand their manufacturing base into export generating areas. The result of this has been a steady increase in the pressures upon high volume-lower cost EU producers leading to a trend, especially by the small to medium sized garment manufacturing companies, to leave the high volume market sector. UK manufacturers, along with many others within the EU, have turned instead to concentrate upon the manufacture of lower volume and generally higher priced garments for niche markets. This market shift has been a direct result of the home industry’s inability to compete with the lower wage production centres outside the community where wages are typically £4 per day or less compared to £4 per hour within UK.
IJCST 15,2
142
In most cases, success and market position are maintained in these niche areas by use of innovative design and ensuring a high product quality whilst being responsive to varying demand and fashion changes. Penetration of these markets by the low cost manufacturers, which are mainly based in the far East, has until recent times been limited by logistical complexity, their limited exposure to the relevant market forces and the cost of meeting the above criteria for success. In recent years, even these specialised low volume and high profit areas have come under threat. Many of the logistical obstacles and prohibitive costs have been reduced as the newly industrialised nations become more sophisticated and with the improved reliability and cost of global communications. Added to this threat are the new low wage and relatively local producers of Eastern Europe. A recent report (EU Document, 1994) shows that in the last 10 years, 50,000 jobs have been lost in UK and 280,000 in total within the EU. The social and economic effect of this decline is compounded by the regional concentration of the industry and the subsequent effect on the local supply chains. Manufacturing systems Historically, for each stage of industrial and technological development there has been a corresponding evolution in manufacturing information systems. Each phase of development has been followed by a period of industry stability and systems consolidation. Examination shows that the time for such consolidation between successive periods of change has been continually reducing. Today, manufacturing companies are being faced with a continual need to change in the market driven quest for responsiveness and flexibility. Regulating drivers from the market place are constantly modifying the operational parameters within companies. The result is that the time periods and horizons for established planning and operational control mechanisms within manufacturing systems continue to be driven down. These market pressures for change have led to a requirement for more reactive manufacturing control systems to accommodate the falling volumes and increased variety experienced by surviving garment manufacturers. Experience and research show that established approaches to planning and scheduling (Little and Yusuf, 1997; Little et al., 1995) are not suitable for application in all industrial sectors. The broad popularity of current systems has often been a product of their successful development or application within a specific industry. This success is not necessarily transferable to another sector. Higher expectations of performance or new drivers from the market place have initiated change in what had previously been seen to be successful and optimum performing systems (Katayama and Bennett, 1996). Although often giving an improvement over existing methods, migration of systems to other
industries or sectors has not always been successful in optimising their Modelling the performance and this has particularly applied to the garment manufacturing changing apparel industry. If manufacturing industry in general and garment production in supply chain particular are to remain major contributors to economic and social well being, it is critical that the best suited (i.e. most market responsive and cost-effective) form of production control system is adopted. 143 When operating in a rapidly changing business environment, those companies that quickly adapt to change, gain competitive advantage. Rembold et al. (1993) suggested that: for companies to be successful the manufacturing systems of the future would need to be flexible and programmable.
In many ways, the converse will be true if present trends continue. At the operational level, manufacturing control systems are becoming increasingly integrated into total business and enterprise wide systems. The product of this increase in integration and potential functionality is that there is a corresponding increase in the number, complexity and inter-dependence of the parameters that control and measure the systems. Consequently, systems change is becoming increasingly more complex and time consuming. Kochhar (1992) states that: the process of selecting and implementing computer based manufacturing control systems in different manufacturing environments is very complex, time consuming and costly.
The move to give central management more information and hence control of the extended enterprise could therefore be said to be in direct conflict with the prevailing paradigm for business success at the operational level. For many businesses this paradigm is structured around simplifying the supply process. Pull production systems providing mechanisms that synchronise manufacture with customer demand have been established. Current practice strives to increase flexibility and shorten response times by devolving decision making and responsibility to the lowest appropriate level within the business. These structures and practices are intended to remove not only the unnecessary costs of excess inventory and its management but also of excess information and its management. Awareness of businesses requirement for responsive adaptability in their systems has focused academic and vendor efforts in working towards developing speedier and simpler procedures for systems installation and subsequent systems re-configuration as evidenced by the introduction of accelerated SAP by SAP. This has led to the initial development of some general, industry specific standard architectures or reference models. Companies are encouraged, under the pressure of extending what are already costly implementation phases to match themselves, as far as is possible, to what the proprietary system offers. This results in their new system being configured to the nearest general architecture available
IJCST 15,2
144
(sometimes referred to as the “vanilla flavour”). Research has shown that this is not necessarily the optimum solution for an effective company that has successfully developed its structure and operational methods in response to its environment over many years (Little et al., 1999 Case-study no. 4). Although systems providers and some academics would state that the new, standard systems software contains the optimum “best of breed” solutions to the broad spectrum of business processes (Wortmann, 1998), this is arguably not the case in all sectors, particularly among the smaller manufacturers. Most enterprise resource planning (ERP) solutions still have MRPII modules at their core. Research has shown that although very successful in many applications, MRPII on its own is not always an optimum planning and scheduling solution for all sectors (Little et al., 1995). To further refine the process of matching manufacturing systems control architectures to a greater range of individual company requirements it has been identified that a selection of more specific reference architectures is required (Little et al., 1996). These will assist in the specification of the order fulfillment process for a particular sector. Within the context of the above, a reference architecture or reference model can be described as a general model for a sector or group of similar companies that is used as a foundation for the design or configuration of business systems. A reference model contains those elements that are common to the business systems operating within the sector. The degree of model genericity is critical when using a reference model to optimise a manufacturing control system. To be worthwhile a reference model must not be too generic, in that it should describe well, those system components or configuration structures that are distinctive to meeting the desired objectives of the system. In the case of the apparel industry, the simplification of in-house systems and the increasing move towards outsourcing must be accommodated. Enterprise modelling A model is a representation of reality. The aim of enterprise modelling is to acquire a realistic representation of systems in relation to the environment in which they operate. Contained within the representation is a description of the processes and functions of the system, their interactions and the information that supports them. There are a number of well-proven techniques available to model manufacturing systems and relevant approaches will be discussed later in this paper. It is difficult to identify one single technique that will satisfy the need to model all relevant issues in all cases. Pandya (1995) concludes that: all of the tools and techniques that he reviewed in his comprehensive study could be used to model a business in general. However some tools were better suited to modelling particular aspects of a business in detail.
Yet the modelling needs to be sufficiently comprehensive to be useful. Doumeingts et al. (1995) states that:
Modelling the changing apparel Within formal systems there are diverse conceptual issues that are difficult supply chain it is necessary to use a modelling system which takes into account decisions, functions, information and resources as well as external factors.
to capture and record using any one technique. This issue is further complicated by the need to capture the detail of the informal, people-driven systems that always evolve to support those formal systems and relieve their inadequacies. Another important question to consider when deciding upon a modelling method is that of the ease and effectiveness of model validation. Doumeinghts states (op. cit.) that: to successfully model, it is necessary to involve and mobilise all people concerned with the system being modelled.
It is a concern therefore that the model must be presented in a form that can be fully understood by all those concerned in its conception. Consequently a technique must be developed that will capture a realistic and useful description of a system, then portray this description in an universally discernible form. Research frame of reference The work detailed in this paper forms part of a broader study of manufacturing control systems funded by the UK Engineering and Physical Sciences Research Council. Whilst investigating manufacturing control systems within case study companies and specifically systems concerning planning and scheduling, the aim of the research was to detail and gain an understanding of the processes they contained and how these related to different industrial sectors. The hypothesis is that by coupling a process modelling approach with an understanding of higher level business drivers, it is possible to determine architectures with characteristics that are likely to promote business success within the specific operating environments. Once identified, these architectures can then be used to facilitate self-evaluation and encourage positive change within a particular manufacturing firm. To this end, a prototype self-audit tool has been developed from the research findings. Once the operating environment of a candidate company has been identified, the tool provides the company with an appropriate planning and control architecture drawn from the reference model. This paper specifically describes the use of the above approach in developing a generic reference architecture for the apparel industry. Apparel manufacturers are being forced by competition and changing markets to make radical changes to the structure of their industry. The competition critical issues of quality and price are now augmented and complicated by those requirements of speed to market, flexibility and responsiveness. In recognition of this, the proposed architecture incorporates those mechanisms identified by the research, which enable the new operational requirement.
145
IJCST 15,2
146
Figure 1. Schematic of first phase data capture and modelling
Research methodology Case company selection Initially, case study companies are identified by their responses to a survey that establishes high-level business context and operating variables. Using a question set developed by the research team the complexity and uncertainty of planning and control operations are then examined. The uncertainty and complexity grid developed for, and reported in the DTI publication “Manufacturing into the late 1990s” is then used to identify similar companies. This technique has the effect of grouping together companies into sectors where like environmental characteristics and responding business processes from the companies can be anticipated. Once these processes are identified and understood by an appropriate generic planning and control reference architecture for the group can then be determined. The research methodology adopted by the research team has a four-phase structure: Phase 1: Data capture. To facilitate study within companies a novel structured method of data capture and validation has been developed. Four techniques, refined after testing within a trial group of companies, were used to construct sets of first phase company models. Selection of the data capture and first phase modelling techniques was based upon the criteria of suitability for gathering the required information and for ease of validation by case company personnel. The method specifically provides models of the companies order fulfillment process (OFP) and associated planning and control activities. Figure 1 shows the bi-directional approach of data capture. The techniques are briefly described below.
(1) Top down Modelling the . Function Relationship Diagrams describe the high level sequencing changing apparel and organisational relationships of the sub-process activities that supply chain comprise the OFP within the company. . Grai Grids capture the decision making and information structures 147 that support the OFP within a context of time. (2) Bottom up . IDEF0 Diagrams are used to map the activities and detail data flows within the OFP, recording initiators, consumed inputs, facilitating resources and planned outputs. . Time Capture Grids (TCGs) were developed by the group in recognition of the difficulty of associating the dynamics of time with IDEF0. TCGs record individual activity times and in-process elapsed time. Top down, interviews, with senior company managers provide an overview of the OFP and the decision and information structures that support it. Bottom up interviews at the operational level, provide detailed process activity information. Each activity is investigated sequentially through the process by questioning the people responsible for its execution. The interviews are structured to elicit the detail of both the official documented procedures and the informal ones that have been developed over time to lubricate the process. The recording of informal procedures is key to recording the actual operation of most SMEs. Phase 2: Detailed modelling. Once the initial data capture and validation phase has been completed the data is input into the ARIS tool-set software to produce an extended event process chain (eEPC) for order fulfillment within the company. In turn, representatives of all functions within the case study company validate this eEPC model. Phase 3: Model analysis. From analysis of the eEPC model for each company a specific planning and control architecture using industry standard descriptors for process activities is constructed. Again this model is validated by representatives of the case study company. Phase 4: Sector reference architecture. Examination and comparison of a set of eEPCs, planning and control architectures and supporting business environmental and operating data (critical factors) enables the production of a sector specific AS-IS reference model. When input from specialists concerning industry trends and requirements is included these models can be further developed to provide sector specific planning and control reference models. The process of deriving the sector architecture: shown in Figure 2 below, is completed by a series of expert validations carried out by knowledgeable industrial practitioners and experienced academics.
IJCST 15,2
148
Figure 2. Data capture and analysis overview
Key apparel industry features Manufacture Although historically large volumes of goods have been produced, garment manufacture is not well suited to most mass production techniques. Short product life-cycles dictate regular changes to item configurations and material properties vary from season to season. Standard, high volume production equipment is therefore difficult and costly to design and its layout often uneconomical to implement. Driven by price competition, early initiatives to reduce labour content and cost have led to some procedures that transfer manipulative skills from workers to machines. Machines for in-process work transfer is now common place and recent advances in the use of computers have been applied to guide cutting machines and to automate pattern making
and grading. Even so production capability is still highly dependent on the Modelling the experience and manual dexterity of its skilled work force and the flexibility of changing apparel simple plant. supply chain Production planning and control Production planning and control within the industry has traditionally followed the general manufacturing paradigm in use at any one time. Much of the manufacturing control software currently used by the industry is based upon the logic underpinning MRPII. Within this there is an apparel specific feature required of MRP and bills of materials. There is a need to define each finished product in terms of its colour, length and size. Along with MRP, piecework payment and work-study are key features of most operations within apparel manufacturing companies. In support of this, most management systems in the industry have at their core a work recording procedure. This uses some form of shop-floor data collection system. These are often bar-code labeling systems with the potential of operating in near real-time. Work is tracked through the OFP by monitoring at specific order points. As contract orders often include a critical path reporting agreement, it is common for these order points to span the whole supply chain. The trend to increased outsourcing is placing different demands upon the planning and control functions. There is also a continuing change from cost critical to time critical manufacturing. The key competitive differentiator is now customer service, which requires logistics sophistication and efficient use of integrated information technology. This change forces a changing emphasis away from internal planning and monitoring to one of supply chain coordination and efficient inter-company interfaces as more and more production moves off site and more products are procured from external suppliers. Research carried out within firms in the UK apparel manufacturing sector (Little et al., 1996) highlighted major changes required in their OFPs. Along with other apparel producers in Europe and those of most developed industrial nations, the companies studied face one major competitive disadvantage, that of relative high labour costs. Wage costs in Europe are between 10 and 50 times greater than those of some emerging industrial nations especially of India, Pakistan and China. When compared to the developed countries of Eastern Europe, South East Asia, Japan and USA, labour costs in Europe can be relatively high (Cahill and Ducatel, 1997a). Outsourcing to these low cost centres of production is now an everyday strategy for most European companies operating in the high to medium volume part of the industry and this has to be recognised by any manufacturing control system. Initially problems with ensuring good quality production and efficient communications over the extended supply chain have been overcome. However, new difficulties are now appearing. Costs in the new partner
149
IJCST 15,2
150
countries are beginning to rise to uncompetitive levels. In China, for example, labour costs have risen by over 300 per cent in the last 5 years (Ip, 1998). New low cost producers around the world are continually entering the industry competing for work. Optimising competitiveness by partnering with lowest cost suppliers and producers requires that supply chains be configured almost on an order by order basis. Software with the capability to efficiently support the management of this type of responsive business practice was not available within any of the case study companies. Apparel planning and control reference model development Case study company overview The apparel companies studied in the course of the research ranged in size of turnover from over £60 million down to £1 million per year. Although serving different markets, they have all developed and implemented a combination of similar strategies. . They have each extended their involvement along the supply chain. Integrating their manufacturing capability with either end product retailing/service provision and/or raw materials supply. . They have or are moving towards some form of product or niche market specialisation where although price is still important, it is non-price factors that are critical to competitiveness and profitability. . They provide a product range and manufacturing mix that comprises both out-sourced and in-house produced items. . They are developing closer and mutually beneficial relationships with a number of reliable suppliers for critical items. . They are starting to configure their supply chains on an order by order basis. Case study company A In the late 1980s this company operated as a supplier to high street retailers and as a jobbing sub-contractor to the larger branded clothing manufacturers. Trading conditions at that time dictated that the company had very little influence over its own destiny. Competition was cut-throat and customers were becoming more demanding in terms of price, quality and delivery. The company was becoming an intermediate warehouse for its main customers, only being paid for goods on delivery. To ease the company’s dependency on the large retailers, who had driven the firm to the verge of failure, a supply chain integration strategy was embarked upon. At a time when retail parks were becoming popular across the country the company opened outlets on several of the most popular developments. These first outlets supplied a range of specialist, high value, and high profit products and were very successful. A decision was made to quickly increase the number
of stores across the country and to gradually change the manufacturing Modelling the business completely over to their supply. The company now runs a nation wide changing apparel retailing chain of 40 stores. It continues to extend the range of products it offers supply chain by applying its now well-known brand name to items produced by other manufacturers. To maximise profitability and meet demand for retail stocks a new supply 151 strategy has been developed. Bulk orders for high volume items are placed on Far Eastern factories based upon six monthly sales forecasts. Shipped by sea, batches of high quality products are supplied store-ready and are warehoused in UK. Lower volume items are produced in the home factory, as are any of the higher volume items that are out of stock and in demand. Relationships have been built with a number of quality approved sub-contracting manufacturers within UK. If demand cannot be met from off-shore sources or from the home factory and the sub-contractors have capacity available, orders are placed with them. Branded items produced by other manufacturers are ordered to contract and called off to meet demand. Currently over 200 companies regularly supply raw materials or finished items. Increased demand and a more certain market for its goods has enabled the company to ensure its supplies by taking over and managing three businesses that had previously operated as its suppliers. To minimise stocks held at any one store, daily point-of-sales data is electronically transmitted to a central warehouse. The company uses a mix of its own and contracted transport to replenish each store from stock on a daily basis. Case study company B Part of a large international group, this company operates from two manufacturing sites within UK. It has for many years produced a range of high quality garments for famous high street retailers and its own stores. Although the company has retailed under its own name for many years it is highly dependent upon the large volume contracts from high street retailers. In support of the increasing demands for flexibility and responsiveness these retailers have placed upon the company and to enhance profitability, there has been heavy investment in recent years to improve production equipment, information systems, warehousing and logistics technology. The raw supplier base has been rationalised and closer links formed with supply chain partners from cloth mill to high street store. Over 1.5 million garments are produced from the UK sites each year and exports are made to 45 countries. The company also has brand name licensee agreements with companies operating in India and the Far East. To help shift some of the dependency from the big name retailers in UK, the company is expanding its own international retailing operation. Using supply chain and production techniques that were developed to support the retail contact trade, the company can speedily and economically produce very high
IJCST 15,2
152
quality products in small batches or even on an almost bespoke basis to its own stores. By extending its brand name to products produced by other manufacturers a complimentary range of high quality, high value goods and service can be provided in all stores. The company’s strategy is based upon associating its already internationally known brand name with a fashionable range of complimentary premier quality products. Coupling this with systems that enable the provision of a second-to-non service around the world is seen to be critical for market success and profit maximisation.
Case study company C For many years this company operated from a number of sites in UK, providing products and service to manufacturing industry. Decline in heavy manufacturing and low cost competition from abroad has radically reduced demand for this established trade. As a result the company has been forced to consolidate its manufacturing operations onto one site and diversifying the product and service range it offers. This has required a corresponding change in its methods of working. Whilst still remaining a major force in its traditional markets the company has used its design and manufacturing expertise to develop specialist apparel collections for use across a range of industrial and service applications. This strategy has been extended by incorporating products produced by other manufactures to provide off the shelf complete kits of apparel and equipment to meet specific requirements. Orders are usually on a contract basis with the company bidding to provide products and service for a specific period. Business is most profitable when existing contacts are renewed. This is highly dependent upon the quality of service offered throughout the existing agreement and having the ability to remain price competitive. Traditionally the company operated as a large batch, in-house manufacturer, relying upon extensive stock holdings of quality assured products to compete in terms of market-critical and delivery lead-time. Now, operating with a much reduced manufacturing and warehousing capacity and producing a wider range of products has necessitated the introduction to new methods of supply and support systems. Although still, supplying mostly from stock the company has moved to a system of small batch manufacture and quick response replenishment. Whilst maintaining stocks at 25 per cent of their old levels the company still usually meets or beats the lead-times demanded by the market on most products. The majority of raw materials are supplied on a consignment basis. Some problems do occur when flame or chemical resistant fabrics are required from specialist suppliers. These expensive materials, usually ordered in small to medium quantities, traditionally have had long purchase lead-times. The criticality of these materials to some major contracts has forced
the company to improved forecasting of usage to allow it to forewarn its Modelling the suppliers of expected future requirements. changing apparel Case study company D This is a single owner company with a highly skilled workforce that produces a range of high quality bespoke garments in any size on a make-to-order basis. From a catalogue of suggested designs, products are supplied worldwide, usually via mail order and often to the British ex-patriot community. In recent years, reduced demand for the product from these traditional customers and the availability of quality goods from competing suppliers around the world has led to a fall in sales. Under new ownership, the company is introducing strategies which are opening up new markets. Recognising the need to reduce total order lead-time to compete with local suppliers the company has started to trade electronically. The company has started to trade over the internet, providing an order-form for customers to register their style selections and for new customers to enter their own measurements. Customers are encouraged to pay before orders are put into production. The company is also building relationships with other complimentary bespoke tailoring businesses. Either floor space is being taken in the partner establishments or the partner takes orders as part of their own business and passes the work on. The ability to demand premium payments for this made to measure service permits the company to use relatively expensive courier services to speedily transport garments from their factory to the partner’s premises for customer fittings or final delivery. Apparel reference model The changes and problem countering business strategies described above have placed many new demands upon the planning and control systems of the companies studied. The research has shown that the existing procedures and production control software used by the case companies does not fully meet the new demands that are being placed upon them. In all of the companies studied a high level of human input was required to maintain product supply. There has been a shift in requirement from systems to solely control internal processes, material availability and movement. The need for monitoring and control of the OFP has moved beyond the boundaries of the company to the whole supply chain. New systems are now required that have the capability to aid the planning and control of an entire product range from a multi-supplier base. All of the case study companies are using or have in the recent past used some form of materials requirement planning. From the evidence of the case studies, it is questionable if this tool has ever provided an optimising solution to the planning and control of apparel manufacture. In common with many other soft goods the bill of materials for articles of apparel are shallow with few
supply chain 153
IJCST 15,2
154
common items. That is finished products are made up from raw materials with few intermediate, sub-assembly processes and with few common components. This feature is at odds with the basic logic of MRP that is best suited to the medium term planning of products with complex structures and common items. Although the logic of MRP is capable of managing products through extended apparel supply chain processes, its automation into usable software systems degrades its capability as a medium to short term planning tool in this application. The presence of uncertainty is known to greatly complicate the operation of MRP systems (Brennan and Gupta, 1993). In an environment where responsiveness and flexibility have become critical, uncertainty is the prevailing operational feature of business and must be well managed. To do so effectively the planning and control system must be able to adapt promptly to change. This requires integrated information systems that enable and support the optimisation of design, organisation and control of the whole supply network. Interfaces must be in place to ensure timely transfer of appropriate information between individual supply elements. Once relevant information is available to decision-makers they must then have the facility to quickly simulate and discern the subsequent effect of any planning adjustment they make. Thus making visible the effect of operational change on the performance of each supply chain element across the whole product range. In attempting to address the issues outlined above, a generic planning and control reference architecture is proposed. The detail of the case study work describes the AS-IS operating conditions, business drivers and key factors for success for a group of companies operating in the apparel sector. The proposed To-BE architecture is the product of insight gained by the research team, discussion with the companies who participated in the project also from input given by a validating group of experienced industrial practitioners. This has been developed from analysis of the empirical research data and with input from case study companies and other industrial practitioners. A high level schematic of this reference model is shown in Figure 3. The detail of the model outlined above describes the architecture for reactive apparel supply. Operating within parameters set by high-level business planning and the externally influenced demand generation process the architecture enables synchronisation of in-house manufacture and/or external source capability. Deviations from the supply plan are monitored and reported in near real time to facilitate re-evaluation of schedules and possible reallocation of resources. Decision-making is aided by the what-if capability of the finite capacity scheduling element. Conclusions There have been several beneficial outputs from the research to supplement the proposed sector reference architectures now being validated by industry.
Modelling the changing apparel supply chain 155
Figure 3. Overview of proposed apparel supply planning and control architecture
Other researchers are using the data capture and modelling method developed by the research team. It is also being evaluated for use by industrial practitioners in the requirements definition stage of systems implementations. The comprehensive review of OFPs has benefited individual case study companies by providing the following. . An independent audit of a company’s OFP has been undertaken. . A comprehensive process map of information flow and workflow has been produced for each company. . A review of planning and control processes within the context of the businesses operating environment has been explored. . Opportunities for improvement have been identified. The research has identified key areas where each company’s existing manufacturing control systems are seen to be inadequate and have indicated, through the reference model, how this can be addressed. Using audit software now in development, companies will be able to examine and compare their current OFPs and operating requirements against the sector reference model to identify those key elements in planning and control that differentiate it. This can lead to the identification of a new architecture that more closely aligns the business to its current needs.
IJCST 15,2
156
References Brennan, L. and Gupta, S.M. (1993), “A structured analysis of materials requirement planning systems under combined demand and supply uncertainty”, International Journal of Production Research, Vol. 31 No. 7, pp. 1689-707. Cahill and Ducatel (1997a), Advanced Technology and the Competitiveness of European Industry, EUR 17732 EN, European Commission Joint Research Centre Document, Institute for Prospective Technological Studies, Seville, p. 4. Doumeingts, G., Vallespir, B. and Marcotte, F. (1995), “A proposal for an integrated model of a manufacturing system”, Control Engineering Practice, Vol. 3. EU. Document (1994), Eurostats OETH DEBA, E.U. Commission Eurostats E.U Comitextil: textilwirtchaft: Dable. EU. Document (1995), Eurostats Nace 453 Clothing, pp. 14-7. Ip, W.H. (1998), “Manufacturing integration strategy using MRP II and RTMs: a case study in South China”, Integrated Manufacturing Systems, Vol. 9 No. 1, ISSN 0957-6061. Katayama, H. and Bennett, D. (1996), “Lean production in a competitive world (Lean Production and Work Organisation)”, International Journal of Operations and Production Management, Vol. 6 No. 2, pp. 8-16. Kochhar, A.K. (1992), “A structured methodology for the effective implementation of manufacturing control systems”, EPSRC ACME Research Conference, Brunel University, September 1992. Little, D. and Yusuf, Y.Y. (1997), Manufacturing Control Systems – Moving Towards the Enterprise Model Factory 2000. Little, D., Kenworthy, J., Porter, K. and Jarvis, P. (1995), Investigation of Best Practice in Short Term Scheduling EPSRC (CDP), Final Report Gr/H/20473, University of Liverpool. Little, D., Porter, K., Rollins, R. and Peck, M. (1996), Planning and Scheduling Reference Models for Different Industrial Sectors, EPSRC Project GR/L2203 Final Report. Little, D., Porter, K., Rollins, R. and Peck, M. (1999), Planning and Scheduling Reference Models for Different Industrial Sectors, EPSRC Project GR/L2203 Final Report. Pandya, K.V. (1995), Review of Modelling Techniques and Tool for Decision Making in Manufacturing Management, I.E.E Process Science Measurement Technology, September 1995, Vol. 142 No. 5. Rembold, U., Naji, B. and Storr, A. (1993), Computer Integrated Manufacturing and Engineering, Addison-Wesley, Reading, MA. Wortmann, J.C. (1998), “Evaluation of ERP systems”, in Umit and Allan Carie (Eds), Proceedings of the International Conference of the Manufacturing Value-chain, 11-22 August 1998, Troon, Scotland, Kluwer, Dordecht. Further reading Cahill and Ducatel (1997b), Advanced Technology and the Competitiveness of European Industry, EUR 17732 EN, European Commission Joint Research Centre Document, Institute for Prospective Technological Studies, Seville, p. 39. PA Consulting Group/DTI (1989), “Manufacturing into the late 1990’s”, A Report by PA Consulting Group, DTI, 1989, ISBN 0 11 515206 7, pp. 54-8. Rollins et al. (1999), Case Study Reports and Interviews. Planning and Scheduling Reference Models for Different Industrial Sectors, EPSRC Project No. GR/l2203.
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Determining reaction abilities of sewing machine operators in joining curved seams Snjezˇana Firsˇt Rogale, Zvonko Dragcˇevic´ and Dubravko Rogale
Determining reaction abilities
179
Department of Clothing Technology, Faculty of Textile Technology, University of Zagreb, Zagreb, Croatia Keywords Garments, Sewing Abstract In accordance with his/her psycho-physical abilities and degree of skill, regulating system in the operator’s brain will match the speed of sewing operations, e.g. actual machine stitching speed, using a simple movement of the foot, to adapt it to his/her sensory and motoric reaction abilities. Workpiece in sewing is fed employing three degrees of freedom movement, i.e. using three independent regulating circuits: position of the seam, edges and workpiece layer length. Cybernetic bases of the regulating system man-machine-workpiece are presented, as well as the development of the necessary and maximum possible sewing machine operator’s reactions in joining curved seams. The procedure is explained by constructing a mathematical model and obtaining the equations used in calculating necessary and maximum operator’s reaction abilities. Measurements necessary were performed in the Department of Clothing Technology, Faculty of Textile Technology, University of Zagreb, using modern measuring equipment while garmentsewing processes were recorded under actual in-plant conditions. The results obtained were computer-processed and a 3D graphic representation was obtained together with a mathematical model appropriate to calculate interdependence of the mentioned parameters. Appropriate software package was developed, which allows a quick, accurate and continuous determination of normal machine-hand times, as dependent on all the influential parameters.
1. Introduction Mo¨ller (1990) was the first to investigate the possibilities of defining mathematical expressions to determine normal times for machine-hand sewing sub-operations, based on the correction of sewing machine stitching speed. He noticed that curved seams were sewn at a stitching speed considerably lower than the machine’s nominal stitching speed, meaning the technological abilities of the machine were not utilised to their full extent. The same author established the ratio of so-called necessary number of reactions and maximally possible number of reactions on the part of machine operator, which is used to rectify the value of the machine-hand time calculated, according to the well-known equation for machine work. He also introduced into his calculations the time necessary to activate pedal regulator in the starting and stopping phases of the driving electric motor work. W. Mo¨ller proposed the following equation for the purpose of calculating normal times of machine-hand sewing sub-operations:
International Journal of Clothing Science and Technology Vol. 15 No. 3/4, 2003 pp. 179-188 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310478260
IJCST 15,3/4
180
tar ¼
l s S d 1667 Rp þD vn Rmax
ð1Þ
where ls is the seam length; Sd is the specific stitch density; vn is the nominal sewing machine stitching speed; Rp are reaction abilities necessary; Rmax are maximum reaction abilities; D is the value added for activating pedal regulator and electric motor acceleration and deceleration. He also proposed the equation for the so-called effective stitching speed of sewing (equation (2)), where nominal sewing machine stitching speed is rectified by the coefficient of the ratio between maximum and necessary number of reactions: vef ¼
vn Rmax Rp
ð2Þ
where vef is the effective stitching speed in rpm. W. Mo¨ller also presented a detail of joining a curved seam, marking the segments to be joined, as well as the deviations that should be rectified in the course of seam joining (Figure 1). He investigated three groups of seam curvatures, in order to determine the necessary number of the operator’s reaction abilities. Slightly curved seams, with curvature radius of more than 60 cm, seams of medium curvature, with curvature radii between 30 and 60 cm,
Figure 1. The system man/machine as a regulation circuit
and finally strongly curved seams with curvature radius less than 30 cm were Determining investigated (Hopf, 1978). He also defined necessary reaction abilities as reaction abilities dependent upon the seam curvature radius, joining accuracy and the method of guiding the seam (Table I). The knowledge of the operator’s reaction abilities is necessary in order to guide seam trajectory, which deviates from the direction of sewing planned, so 181 that the seam is positioned as planned and with a pre-determined accuracy (Figure 1). Reaction abilities necessary are determined on the basis of the level of seam trajectory deviation (in mm) from the planned sewing direction, on the seam segment of 5 cm, presuming adequate joining accuracy, according to the following equation: Rp ¼
deviation joining accuracy
ð3Þ
The table depicting the maximum number of the operator’s reactions, as dependent upon the nominal stitching speed and seam density, is obtained in a similar way. W. Mo¨ller described the concepts of necessary and maximal
Method of feeding workpiece Single fabric layer for sewing or common feeding the fabrics to be sewn
Separate feeding of the fabrics to be sewn
Seam description
rc (cm)
x (mm)
Rp
Straight and slightly curved
.60
Medium curved
.30-60
Strongly curved
#30
#^ 3 #^ 2 #^ 1 #^ 3 #^ 2 #^ 1 #^ 3 #^ 2 #^ 1
0.3 0.5 1.0 0.7 1.0 2.0 1.0 1.5 3.0
Both straight or slightly curved
.60 and . 60
#^ 3 #^ 2 #^ 1 #^ 3 #^ 2 #^ 1 #^ 3 #^ 2 #^ 1 #^ 3 #^ 2 #^ 1 #^ 3 #^ 2 #^ 1
0.7 1.0 2.0 1.3 2.0 4.0 1.3 2.0 4.0 1.7 2.5 5.0 2.0 3.0 6.0
Straight or slightly curved/medium curved
, 60 and .30-60
Medium curved/medium curved to the other side
. 30-60 and . 30-60
Medium curved/strongly curved
. 30-60 and # 30
Strongly curved/strongly curved to the other side
#30 and # 30
Table I. Necessary operator’s reaction abilities (Rp) as dependent upon joining accuracy (x in mm), curvature radius (rc in mm) and the method of feeding the workpiece
IJCST 15,3/4
182
operator’s reaction abilities in an incomplete manner and was vague in his papers published; thus it can be assumed that he determined these values through experiments, using empirical methods. The authors have used the results obtained by W. Mo¨ller in order to clarify some of his assumptions and establish new ones that could be used as a basis for developing new methods of determining the time for machine-hand curved seam sewing sub-operations.
2. Measuring system and equipment Curved seam sewing operations were recorded under laboratory and in-plant conditions. Measurements were performed in the Department of Clothing Technology Laboratory, Faculty of Textile Technology, University of Zagreb. Measuring equipment and the method (MMPP) for computer-aided determination of basic processing parameters for sewing operations have been developed at the Department of Clothing Technology, Faculty of Textile Technology, University of Zagreb (Rogale and Dragcˇevic´, 1998). The equipment consists of a computerised system for measuring and storing processing parameters and a bi-plane system of video recording, intended to analyse the work at workplaces and to interpret properly the results obtained in measuring processing parameters of garment sewing operations. The computerised equipment can determine real and average sewing machine stitching speed, maximum stitching speed achieved, acceleration in the course of sewing, specific stitch density in a seam, seam length, specific time necessary to sew 1 m of a seam, as well as some other processing parameters. The second part of the system consists of a bi-plane camcorder system, which simultaneously records movements and other occurrences in the horizontal plane, offering continuous and comparable video recordings of the operator’s limb positions, body positions and movements in performing the operation of garment sewing, presented as ground plan and side view presentations.
3. The concept of a method based on cybernetic model of reaction abilities The concept of a method based on cybernetic model of reaction abilities deals with defining a mathematical expression and functional dependencies of the number of reactions necessary (Rp) and the maximum number of the operator’s reactions (Rmax ) in the course of joining curved seams. Defining a mathematical expression of the functional dependence of the number of the operator’s necessary reactions is based on a geometrical representation of the dependence of the factors involved in joining a curved seam (Figure 2).
The presentation of geometrical dependence of the factors involved in Determining joining a curved seam (Figure 2) is a development of the idea presented by reaction abilities W. Mo¨ller. The basic idea is used to develop and establish both groups of methods. Geometric relationship, as shown in Figure 2, offers a basis for a mathematical expression to link the distance (wx), after which seam trajectory should be rectified, seam contour curvature radius (rc ) and deviation (x) of the 183 seam trajectory should be planned and realised: qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi wx ¼ r 2c 2 ðrc 2 xÞ2 ð4Þ where rc is the seam curvature diameter in mm; x is the deviation of the real from planned seam trajectory in mm; wx is the distance after which a need arises to react, with a known rc allowed deviation x in mm A referral length of 50 mm is used in establishing the model, with the aim of determining the necessary number of reactions (Rp) due to which the operator should stop sewing and align the seam to rectify the deviation of the seam trajectory realised and the planned one. Taking this into account, necessary number of the operator’s reactions can be determined as a ratio of the referral length (50 mm) and the distance (wx) after which the need arises to react in order to rectify the deviations (x): Rp ¼
50 wx
ð5Þ
Substituting equation (4) into (5), the final mathematical form to calculate the necessary number of operator’s reactions for the referral length of 50 mm, with a known curvature radius and deviations in trajectory, i.e. joining accuracy is obtained. 50 ð6Þ Rp ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi r2c 2 ðrc 2 xÞ2 A mathematical model is constructed in this way and the functional dependence is established for the necessary number of reactions, seam
Figure 2. Geometrical dependence of the factors concerning joining a curved seam
IJCST 15,3/4
184
curvature radius and joining accuracy. It is evident that the necessary number of the operator’s reactions is directly related to the geometrical factors involved in joining curved seams. On the basis of the equation for functional dependence Rp ¼ fðrc ; xÞ; the distances (wx) are determined after which the necessary number of operator’s reactions occur (Rp) in joining curved seams with curvature radii (rc in mm) and for joining accuracy (x in mm) of 1, 2 and 3 mm. Their functional dependence can be seen in Figure 3. In the following stage of the investigation and construction of the mathematical model, it is necessary to establish the relationship of maximum possible number of operator’s reactions (Rmax ) and other influential factors acting in joining curved seams. It can be seen that Rmax is the ratio of the time necessary to sew a seam of the referral length of 50 mm (tSR50) at given parameters: nominal sewing machine stitching speed (vs in rpm), specific stitch density (Sd in cm2 1) as well as the operator’s reaction time (tr in seconds). Reaction time of 0.2 s is recognised in garment technology (Grandjean, 1991). If the above is true, the value of Rmax is the ratio of the time: Rmax ¼
300S d vn
tr
¼ 1500
Sd vn
ð7Þ
Obviously, maximum possible number of the operator’s reactions (Rmax) depends on the specific stitch density in the seam and sewing machine nominal stitching speed. It is possible to calculate, on the basis of equation (7), the dependence of the maximum number of the operator’s reactions, machine nominal stitching speed and specific stitch density. Figure 4 shows a 3D presentation of the curve depicting the relationship. According to the
Figure 3. Functional dependence of seam curvature radius (rc in mm) and joining accuracy (x in mm)
mathematical model established, and based on equation (7), maximum number Determining of the operator’s reactions (Rmax ) is compared with the tables proposed by reaction abilities W. Mo¨ller. Figure 5(a) shows an example of the nominal stitching speed of 2,500 rpm, for which the results recorded by W. Mo¨ller and the results obtained using the mathematical model match, as well as an example for the nominal stitching speed of 6,000 rpm (Figure 5(b)), where the results do not match 185 completely. Results obtained through the mathematical model are linear ones, which was expected, due to the nature of the assumptions, mathematical expressions and the model as a whole. Dotted line is used to present the results obtained by W. Mo¨ller through his measurements. The verification of the mathematical model described, as well as the comparison with the results obtained by W. Mo¨ller, concerning maximum possible number of operator’s reactions, show the model is adequate for the purpose. Results obtained by its application are more precise, and such accurate and correct mathematical expressions will be used further in computer-aided calculations of the time necessary to perform machine-hand sewing operations. 4. Method of determining normal time for machine-hand sub-operations of sewing seams with a single curvature radius (MONOR1 method) Results obtained are used to establish a comprehensive mathematical model called MONOR1. The model consists of equation (1) used to calculate the time of machine-hand curved seam sewing operation and equation (2) used to calculate the effective machine stitching speed, as published by W. Mo¨ller in 1990, as well as equation (6) used to calculate necessary (Rp) and equation (7), maximum possible number of operator’s reactions (Rmax). The model is
Figure 4. Functional dependence of the maximum number of operator’s reactions (Rmax), specific stitch density (Sd in cm2 1) and nominal sewing machine stitching speed (vn in rpm)
IJCST 15,3/4
186
Figure 5. Maximum possible number of operator’s reactions (Rmax), as dependent upon stitch density (Sd in cm2 1) for nominal sewing machine stitching speeds (vn in rpm). (a) 2,500 and (b) 6,000
developed for the purpose of the investigations described. All in all, the MONOR1 model consists of two equations that have been known already and two completely new ones. The new equations are used instead of the table values (Rp, Rmax) determined by Mo¨ller (1990), and offer only rough values. Consequently, computer software can be developed and constructed following the MONOR1 model, to be used in calculating normal times for machine-hand curved seam sewing suboperations. Figure 6 shows a diagram comparing normal times for machinehand curved seam sewing sub-operations according to W. Mo¨ller’s methods and MONOR1 method (Grandjean, 1991). Figure 6 shows that, using the MONOR1 method, a continuous calculation of normal times for machine-hand curved seam sewing sub-operations can be obtained for any curvature radius, while this is not possible by
Determining reaction abilities
187 Figure 6. Diagram comparing normal times for machine-hand curved seam sewing suboperations according to the W. Mo¨ller’s methods and MONOR1 method
implementing the method by W. Mo¨ller, as curvature radii of his division are in three areas. 4. Conclusion The investigations described resulted in establishing a method based on a cybernetic model of sewing machine operator’s reaction abilities. Taking W. Mo¨ller’s theoretical assumptions as a starting point, a new method was established using his initial expressions for calculating the time of machine-hand curved seam sewing sub-operations and effective sewing speeds; new mathematical model and new expression were founded to determine the necessary reactions and maximum possible number of the operator’s reactions, successfully substituting table values offered by W. Mo¨ller. The experiments performed, results obtained, as well as the method established and verified for determining normal times of machine-hand curved seam sewing sub-operations, indicate that the method proposed can be applied for each and every specific technological condition of sewing curved seams. It can also be concluded that the method proposed can be successfully used with the methods of pre-determined times (especially the MTM method), thus eliminating actual disadvantages of the otherwise acceptable methods for determining machine-hand time of sewing sub-operations. References Grandjean, E. (1991), Physiologische Arbeitsaltung, 4. Auflage, Ott Verlag Thun, Ecomed Landsberg, ISBN 3-609-64460-5, Deutschland. Hopf, H. (1978), “Beno¨tigen wir ein neues Unterweisungsprogram fu¨r Na¨herinnen?,” DOB + haka praxis, Vol. 13 No. 7, pp. 456-61.
IJCST 15,3/4
Mo¨ller, W. (1990), “Realistische Prozesszeit fu¨r das Maschinenna¨hen”, Bekleidungstechnik, Vol. 36 No. 17, pp. 30-2. Rogale, D. and Dragcˇevic´, Z. (1998), “Portable computer measuring systems for automatic process parameter acquisition in garment sewing processes”, International Journal of Clothing Science and Technology, Vol. 10 No. 3/4, pp. 283-92.
188
Further reading Firsˇt Rogale, S. (2002), “Metode odredivanja strojno-rucˇnih vremena tehnolosˇkih zahvata sˇivanja”, Mag. thesis, Faculty of Textile Technology, University of Zagreb, Croatia.
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Load-adapted 3D-reinforcement by means of function-adjusted ready-making process
Load-adapted 3D-reinforcement
189
Claudia Herzberg, Sybille Krzywinski and Hartmut Ro¨del Institute of Textile and Clothing Technology, Technische Universita¨t Dresden, Dresden, Germany Keywords 3D, Textiles, Clothing Abstract Complex material requirements for high-technology applications increasingly demand the use of hybrid material structures with properties tailored to the lines of loading. Textilereinforced multilayer composite structures are particularly suitable for the production of component structures in an optimised lightweight construction. In the loading case, however, delaminating phenomena occur between the individual layers due to the low interlaminar shear strength. The appropriate techniques and machines of the ready-made-clothing technology allow the specific sewing-up of the semifinished textile products into a three-dimensionally reinforced multilayer composite structure; the setting of a load-adapted and failure-tolerant characteristic of properties being possible in the z-direction through a versatile variation of sewing parameters. Moreover, the sewing technology makes possible a ready-made-clothing-technological preassembly of components of semi-finished products, and thus can perform position-fixing functions in the consolidation of the composites. The ready-made-clothing process is divided into sub-processes like product development, preparation of cutting, cutting, connecting and forming as well as packaging and shipping. The technical procedures and machines applied are chosen from economic aspects. Besides the large number of pieces, extreme thickness of the textile products of up to 20 mm and the required sewing precision demand precise and reproducible manufacturing processes.
1. Introduction The empirical examinations together with the experiments carried out on the textile and plastic composites, which are joined by sewing, impressively showed that the sewing technology is an excellent means to noticeably increase the interlaminary strength for the application of load-adapted z-reinforcement arrangement. The textile assembly of complex textile preforms and their further plastic processing is possible. The research work will be carried on, which will open new markets for new products to the textile and clothing industry. These textile-reinforced synthetic materials will make vehicles and machines lighter, more resistant, safer and, as a result, more economic in the future. Within the framework of a research group – “Textile reinforcement for highperformance rotors in complex applications” promoted by the German research society (DFG) – rotationally symmetric reinforcement structures in the shape of rotors were assembled technologically by means of optimal sewing process
International Journal of Clothing Science and Technology Vol. 15 No. 3/4, 2003 pp. 189-197 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310478279
IJCST 15,3/4
190
parameters and further processed plastics. Eight subprojects focus on the following issues: . carbon and thermoplastic hybrid yarn; . spinning of very thin PEEK filament yarn; . textile fabrics with variable axes for rotational loads; . application of procedures and machines used in the textile industry to manufacture and assemble textile preforms; . computations on the basis of material parameters; . experimental testing of the prototypes on the rotor test stand; . integration of sensor components in the rotors; . application of kinematic molding tools. It is interesting to note that a thermoplastic matrix is applied, which is introduced into the composite via the textile form by consistently using hybrid yarns. The plastic process is realized by applying the autoclaving technique for the manufacture of complex components. Autoclave technique, i.e. curing the laminate in the vacuum under pressure and heat, still requires a certain degree of manual work. Form tools adapted to the component are necessary to stabilize the made-up textile preform. In the autoclave, the component geometry should be subjected to a homogeneous pressure. Therefore, the form tools should ideally be tensile inside the component. 2. Production of preforms within the ready-made process The tests are carried out on high-performance rotors with a diameter of 500 mm containing blades that are arranged between two disks. Potential applications of these rotors may be in centrifugal substance separation process, chemical and laboratory technologies and also in flywheels running at extreme speeds. The disks of the model rotor 1 are still flat, while one of the disks of model rotor 2 has already been hyperbolically shaped (Plate 1). The following steps are necessary to make a textile preform with the component design being very complex: . pattern design in accordance with material behavior; . cutting;
Plate 1. Model rotors 1 and 2
. . .
stacking; prefabrication and placing of the z-reinforcement; assembly of the 3D preform.
The technical procedures and machines applied are chosen according to the economic aspects. Besides the large number of pieces, extreme thickness of the textile products of up to 20 mm and the required sewing precision demand precise and reproducible manufacturing processes. 3. Pattern construction with due regard to material behavior If curved element contours of lightweight textile structures are covered with an undefined shape of the reinforcing textile, the mechanical component properties may deteriorate. The patterns should be developed directly on the object to apply the reinforcing structures to the desired 3D shape according to the required load and thus rework is avoided. Three-dimensional CAD programs are mainly applied to design complex components (AUTOCAD 2000, Pro Engineer, Thinkdesign 4.0, CATIA. . .). The data obtained by the above programs may be transferred to the simulation programs via suitable interfaces (IGES – initial graphics exchange specification and VDAFS – interface suited for the exchange of free forms and curves). The component outline and/or the form tools, respectively may quite well be generated in the simulation programs 3D concept directly, which is however, not very comfortable in terms of handling. The textile and bendable (with low bending stiffness) preform should be in most exact accordance with the component geometry desired in the end. In particular, for the realization of free-form surfaces, it is necessary to cut the fabric or non-stitched fabric so that it may be shaped later without irregular folds. In the garment industry this is called “working in the close-to-body range close-fitting garments”, where the essential factors that influence the shaping behavior are the tension and stretching behaviors together with the shearing behavior of the material used. After the patterns have been developed with due regard to functional requirements using a 3D model, surface generation and the development of the 2D patterns in the 2D level are made feasible by an efficient software tool (3D concept by CDI, recently belonging to Lectra systems, France). The software 3D concept is based on the polygone computation of non-uniform-rational-Bsplines (NURBS). It is considered the state-of-the-art computation method to design complex polygon surfaces. Shearing that occurs in the pattern as well as material tension stresses and stretching may be analyzed to provide the designer with information that would enable him to produce suitable patterns of the reinforcing textile material. For this purpose, detailed knowledge about the mechanical properties of the reinforcement structures that are to be processed is
Load-adapted 3D-reinforcement
191
IJCST 15,3/4
192
inevitable. The instrument system developed by KAWABATA may be used to investigate the mechanical parameters of textile fabrics, such as measurements of compression, surfaces, bending, shearing and strain/tension. In some cases, the common standardized test procedures may also be applied. The thermoplastic matrix materials used are polypropylene (PP), polyamides (PA), polyethylene terephthalate (PET) and even highperformance plastics such as polyetheretherketone (PEEK). Thanks to the use of commingling yarns in the form of thermoplastic hybrid yarns for the production of the intermediate textile stuff in the form of fabrics and/or noncrimp fabrics, an additional matrix component that changes the shaping behavior of the textile surface is not necessary. The material data obtained for shearing, material tension stress and also the stretching behavior may be implemented by scanning the measurement curves and subsequent scaling or by loading a file in the ASCII format. This investigation starts from an orthotropic structure for the majority of fabrics tested so far. When high modulus carbon yarns are processed (E modulus .650.00 N/mm2), we may start from the fact that the potential deformation between the 2D cutting and the multiplied curved component surface results from the shearing deformation. Problems characterized by large deformations may be described by incremental formulations to determine the state of deformation and tension stress. For this purpose, a network is generated on the component surface to be shaped. The network may be generated automatically or interactively. The accuracy of computation depends on the triangle size. Material behavior is attributed to the network to simulate the development in the 2D level depending on the material type. After the computation has been completed, the shearing that occurs in the shaped patterns may be read. A comparison with the critical shearing angle, which indicates how far the share of threads can be twisted/compressed without crease formation (folds), helps the designer to decide whether the cut-up pattern is suitable for the component surface (Plate 2). 4. Cutting The pattern design is followed by the cutting process. Very often multiple-layer cutting of geometrically identical pattern is not feasible due to the load-specific arrangement of the threads in the fabric. In addition to the conventional cutting techniques (knife cutter - drawing knife, oscillating knife), unconventional cutting media (hydrocutter, laser cutter) are particularly suitable. If the fabric to be cut is made of 100 percent glass fiber or carbon fiber, the obtainable quality of the cutting edges will not be satisfactory. There will be slippage of the edge fibers or falling out of the fabric during successive stacking and ready-made processes. When thermoplastic hybrid yarns are used by cutting with a laser cutter (1,000 W) a consolidated cut edge will be obtained, which positively influences the exact further ready-made processing on the one
Load-adapted 3D-reinforcement
193
Plate 2. Sample rotor with conical shell, wrap flat simulation 3D to 2D
hand, but does not influence the shaping behavior of the fabric. The thermoplastic fixation is effective from 1 to 2 mm into the fabric. 5. Investigation of sewing processes Shaping and joining requires the use of sewing-machine technology. The extremely high number of layers, the geometry of the blank parts and preforms, along with the rows of stitching required for reinforcement, all currently point to a thrust to further develop the sewing technology involved. CNC system control of the stitching processes is a prerequisite for the reproducible fabrication precision required for mechanical engineering and automotive sector applications. There are additional processing problems in this area in the form of the normally very delicate fibrous materials used as sewing thread. Multiple parameter variations with respect to the type of stitch, sewing needle geometry, type of sewing thread, sewing thread precision, stitch length, seam clearance and direction can be used to establish the load-bearing characteristics of textile seams. 5.1 Establishing needle penetration force The needle penetration force is one of the factors selected, with respect to research, for sewing techniques in the application of multiple-layer, textilereinforced semi-finished components with a total thickness of up to 20 mm. It is assumed that dimensions should be easy to handle with measuring instruments, thus permitting evaluation of the efficiency of the sewing needle precision, the design of the needle point, and the turning speed of the main shaft required for the needle to penetrate the material being sewn.
IJCST 15,3/4
194
The dynamic process is carried out with reference to the angle of rotation of the main shaft of the sewing machine. 5.2 Perforation effects The penetration of the needle into the laminated, reinforced textile material is inevitably associated with a certain perforation effect. An important influencing factor, apart from the geometry of the sewing needle, is the length and density of the stitching. The degree of weakening attributable to the perforation effect is determined by carrying out tensile strength testing on the reinforcement textile layers that are partially perforated by the stitching action and which lie crossways to the direction of strain. 5.3 Reduction in seam strength The stitching process in particular is related to paired elements regarding the effects on sewing threads of larger correlated dynamic loads and the results of rubbing and friction. Stronger (and thus thicker) sewing threads made of the same fiber material can in fact increase seam strength and the reliability of the stitching process, but they require the use of larger-diameter sewing needles and greater-sized perforations in the surface of the textile material. Thus, evaluation of the loss of strength caused is extremely difficult to carry out, although a comparison of the maximum tensile strengths of stitched and non-stitched threads, using appropriate testing apparatus, can produce results of the desired accuracy if a sufficient number of tests are carried out. 5.4 Selection of stitch type Double lock stitching is particularly suitable for sewing textile-reinforced semifinished items. However, the principle behind this technique produces stitches where the loop in the thread normally lies in the center of the article being sewn, which has a significant weakening effect on delicate sewing threads made of materials such as glass or carbon fiber. In the case of bend-stressed textileplastic composites, the maximum interlaminate shear stresses likewise arise in the center of the composite, while in the case of compound textile-reinforced structural items such as overlapping joins, it is the maximum z-standard tensions that occur at this point. By varying the tension in the upper and lower threads, it is possible to displace the kink in the stitch towards the outer or inner surface of the material being sewn. This stretches the thread in the breakdown-critical area, thus allowing this weak point to be avoided. 5.5 Thread tension Thread tension also influences the alignment of the seam in a consolidated textile-plastic composite. The multiple-layer, stitched textile-reinforcement material is compressed, during the composite material consolidation process, to
a fraction of its original thickness, thus converting a loose set of textile layers Load-adapted into a compact composite structure. An increase in the tension of the upper and 3D-reinforcement lower thread, and the effect of high sewing foot pressure, can lead to a tendency of the thickness of the multiple-layer reinforcement textile material to be compressed down to the thickness produced by consolidation. Using the available technological knowledge regarding possible processing parameters, 195 and textile-fabrication techniques and machines, it has now been possible to manufacture complex textile preforms for high-performance rotors. 6. Pre-fabrication of the rotor disk The disk consists of several reinforcement textile cuttings, which can yet be supplemented by specifically shaped parts. The sewing unit must guarantee the defined multi-layering of the reinforcement textile cuttings with positioning seams. Taking the concrete loads situation under consideration, defined reinforcement seams are to be applied. The problems of multi-directional sewing known from the x – y-cross-table are to be avoided for the most part, as the carbon multi-filaments, or carbon/thermoplastic multi-filaments and/or stable fiber yarn that might be used as sewing thread – all are extremely difficult to work with on account of the high brittleness of carbon fibers. The sewing unit (Plate 3) allows the most part, a normal drawing-off of the thread from the sewing needle to the seam in the material. As the basic seam forming concentric circles and spirals were chosen which, by a variation in diameter, the number of rotations, the stitch length, as well as the beginning angle in a polar coordinate system, they are changeable and make possible the arrangement of the positioning and the reinforcement of seam far-reachingly in favor of the composite properties.
Plate 3. Sewing unit for prefabrication of the rotor disc
IJCST 15,3/4
196
Plate 4. Stages of the rotor production
The sewing unit uses a commercial sewing head. The material-leading equipment is a specially constructed one whereby steering and driving are moved by means of the CNC-technology of the SIEMNENS AG. The construction of the material-loading equipment was established by a contract with the CETEX Chemnitzer Textilmaschinenentwicklung gGmbH Chemnitz. 7. Assembly of the 3D preforms Complex component geometries often cannot be made in a single step. Only by following the assembly of individual textile components in two or more substeps, the complex preform is created. The stitching used here is known as assembly stitching and it fulfils its function primarily in the textile preform. Depending on component geometry, it may also fulfil a reinforcement function in the consolidated component. The use of assembly stitching is conceivable not only to the normal surface, using special sewing technology, but it may also make good sense to employ a variable angle to the surface in order to lend shape or effect reinforcement. The component geometry determines the shape and dimensions of the textile cut parts, whose contours are developed using industrially applied CAD systems. The cut parts are arranged on the flat structures in a way that is tailored to the yarn lines, and then cut using the state-of-the-art CNC automatic cutting systems designed for this purpose. The sewing technology is specifically suitable for shaping and connecting. The extreme number of layers, the geometry of the cut parts and all the preforms as well as the arrangement of the sewing yarns in the component due to reinforcement requirements give a strong impetus to the sewing technology (Plate 4). The CNC control of the sewing process is a precondition for
the reproducible production precision. Another aspect is the problematic Load-adapted processing of these normally very brittle fibers when they are used as sewing 3D-reinforcement yarns. Packaging and shipping of the textile preforms from the garment industry to the plastics enterprise should be handled in a careful and clean manner. Component-integrated sensors are necessary for measuring the extension 197 field that occurs as a result of operational and extreme extraordinary loads. Ready-made clothing technologies are used to integrate these sensors into the textile preform. A vacuum high-speed rotor test stand verifies the data. Further reading Herzberg, C., Krzywinski, S. and Ro¨del, H. (2001), “Load-adapted 3D-reinforcement through function-adjusted stitching technique”, Vortragsband 13th International Conference on Composite Materials, 25-29 Juni 2001, Peking, China. Offermann, P., Ro¨del, H., Leopold, T., Choi, B.-D. and Herzberg, C. (2000), “Fertigung textiler Versta¨rkungshalbzeuge fu¨r CF-PEEK-Hochleistungsrotoren – vom Garn bis zur Faserpreform”, Vortrag Deutscher Luft-und Raumfahrtkongress, 18-21 September 2000, Leipzig. Ro¨del, H., Herzberg, C. and Krzywinski, S. (2001), “Einsatz konfektionstechnischer Verfahren fu¨r die Herstellung von komplexen textilen preforms”, Tagungsband. 8. Chemnitzer Textilmaschinentagung, 24-25 October 2001, Chemnitz. Ro¨del, H., Schenk, A., Herzberg, C. and Krzywinski, S. (2000), “Links between design, pattern development and fabric behaviours for cloths and technical textiles”, Tagungsband 3rd International Conference IMCEP, 11-13 October 2000, Maribor, Slowenien.
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
IJCST 15,3/4
198
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Mechanical properties of sewing stitch performed in frozen state Gojko Nikolic´, Zˇeljko Sˇomodi and Diana Franulic´ Sˇaric´ Department of Clothing Technology, Faculty of Textile Technology, University of Zagreb, Zagreb, Croatia Keywords Sewing, Mechanical properties, Textiles Abstract The proposed solution for increased automation in clothing production by freezing the textile involves sewing operations in frozen state. The realistic applicability of such technology could be limited, amongst other issues, by the possible loss of mechanical properties of the sewing stitch performed in that way. This work represents an experimental investigation for the evaluation of the influence of textile freezing to possible changes of mechanical properties with respect to the normally performed sewing stitch. A series of standard extension tests are done on specimens of sewing stitches performed both in frozen state and in usual way for comparison. The differences in the results are analysed statistically for a representative textile material.
International Journal of Clothing Science and Technology Vol. 15 No. 3/4, 2003 pp. 198-203 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310478288
1. Introduction The introduction of high automation into the clothing production process has been seriously limited by the mechanical behaviour of textile. The lack of stiffnes is the main reason why textiles are highly demanding for the automated manipulation in the process of assembly and sewing (Rogale and Knez, 1988). Nikolic´ and Sˇomodi (2000, 2001) tried to overcome these difficulties by freezing the textile workpieces – bringing them in the stiff state suitable for automated (robotic) manipulation. It seems reasonable to expect that this would result in a considerable reduction of labour built in the final product. The elaboration of the mentioned idea has been done in terms of numerical limit analysis of frozen textile modelled as rigid-plastic thin plate (Sˇomodi and Nikolic´, 2000), followed by an experimental evaluation of mechanical properties of frozen fabric in relation to freezing parameters (Nikolic´ et al., 2001). Our next concern is whether or not the quality of the sewing stitch shall be influenced by performing it in wet or frozen state, as opposed to normal sewing in dry conditions. This paper represents an experimental study meant to help clearing up possible questions on this matter. The authors are grateful to Anica Hursa, MSc, Alena Mudrovcˇic´, Ante Agic´, PhD and Tomislav Rolich, MSc for their kind assistance.
2. Experiments For the time being, the purpose of these experiments is not to investigate the behaviour of wide range of fabrics, but to establish and check out the methodology. Therefore, we choose to investigate the sewing stitches performed on a single material considered to be representative for possible sewing operations to be automated. The data for the chosen fabric are as follows: material 100 per cent cotton, thickness 0.77 mm (evaluated using Croatian norm HRN F.S2.021), mass per unit area 117 g/m2, both warp and weft densities 27 yarns/cm, cloth weave: 1/1. Standard testing of the quality of sewing stitch, as described in Cˇunko (1995), consists of the following: two rectangular pieces of fabric of dimensions 20 £ 30 cm are sewn together in the position face to face by a straight stitch 1 cm away from the longer edge. The specimens are cut out in the form of 5 cm wide strips, with the width in the region of the sewing stitch increased to 10 cm. Standard prescribes three specimens for each direction (warp and weft). After certain preloading, the specimen is fixed in the initial 200 mm distant clamps of the testing machine and loaded, with elongation rate of 100 mm/min, to a certain load corresponding to the given mass per unit area of the fabric (80 N in our case). At this load, the specimen is then visually inspected and the extent of shear deformation in the region of the sewing stitch is reported, and measured if necessary. In the preparation of specimens, the freezing temperature was 2208C and the sewing was at the room temperature with as short time as possible (10 s) from the freezer to the start of sewing – enough to well preserve the rigidity obtained by freezing. The fabrics sewn in wet and frozen states are first dried out and then cut into the specimens and tested. In our experiments, we have slightly modified the shape of the specimens by cutting the curved transition to the wider part in order to avoid the stress concentration (see photographs in Plate 1). Although this is not prescribed by the standard, after the visual inspection (photographing) we have continued the loading until the rupture when the ultimate tensile force, elongation and the position of the crack are recorded. These tests are carried out at the laboratory of the Department of Mechanical Technology, Faculty of Textile Technology, University of Zagreb. In addition to these tests, the load-elongation diagrams are also recorded, in order to obtain a more detailed insight into the possible differences in mechanical behaviour of the specimens sewn in dry, wet and frozen state. 3. Results and discussion The appearance of the specimens under the standard tensile load is shown in Plate 1. The marks on the specimens correspond to the sewing conditions (first letter S for dry, M for wet and Z for frozen) and the direction of the fabric (second letter O for warp and P for weft). The marks were not identically placed
Mechanical properties of sewing stitch 199
IJCST 15,3/4
200
Plate 1. Snapshots of the specimens in tensile loading. (a) Sewn dry, warp direction, (b) sewn dry, weft direction, (c) sewn wet, warp direction, (d) sewn wet, weft direction, (e) sewn frozen, warp direction, and (f) sewn frozen, weft direction
in all the specimens, so after cutting the pictures, some specimens appear unmarked. Dry sewn specimens are in the top row of Plate 1 (left warp, right weft direction), wet sewn specimens in the middle row and specimens sewn in the frozen state are in the bottom row. In Figure 1, the recorded force-elongation diagrams are given in the same formation as the snapshots in Plate 1. Experiments are performed on the machine Zwick 1445 of the Laboratory for Mechanical and Physical Testing of Materials, Faculty of Chemical Engineering and Technology, University of Zagreb.
Mechanical properties of sewing stitch 201
Figure 1. Force-elongation diagrams. (a) Sewn dry, warp direction, (b) sewn dry, weft direction, (c) sewn wet, warp direction, (d) sewn wet, weft direction, (e) sewn frozen, warp direction, and (f) sewn frozen, weft direction
IJCST 15,3/4
202
The results of the measurements are used to compute the mean value and standard deviation of both the ultimate tensile force and elongation for each group of three specimens. The computed data are given in Tables I and II. The results indicate no significant differences in tensile strength. Let us also mention that the final rupture was never in the position of the sewing stitch – the fabric broke more or less close to one of the clamps. In our opinion, the clearly increased extensibility, especially in warp direction, of the specimens which have been moisturised and dried, is to be attributed to the lack of the previous treatment of the fabric, and not to the fact that the sewing itself had been done in wet or frozen state. 4. Concluding remarks The main conclusion is that our expectations have been confirmed – there is no significant loss of strength caused by sewing in the frozen state. Of course, this holds for our selected material, and other fabrics, especially those with considerably different characteristics, which should be treated with care and examined separately. However, since the fabric employed in our tests can be considered as a good representative of fabrics used in the production of wide variety of clothes, the applicability of freezing textile for automated handling is thus confirmed, at least from the standpoint of mechanical properties of the sewing stitch. The lack of stabilising treatment prior to sewing can affect some mechanical properties of fabrics, probably the shrinking as well, but the strength of the sewing stitch itself remains basically unaffected. The next step towards the highly automated clothing production based on textile freezing should be, as we see it, the technological elaboration of the selected production process, considering the steps related to moisturising, freezing, and automated (robotic) manipulation of textile.
Table I. Ultimate tensile force – mean value and standard deviation
Ultimate tensile force (daN) Warp direction Weft direction Sewn dry Sewn wet Sewn frozen
22.8 19.2 24.8
1.36 3.30 2.68
21.3 19.5 19.8
2.45 1.47 0.62
Extensibility ( per cent) Warp direction Table II. Extensibility – mean value and standard deviation
Sewn dry Sewn wet Sewn frozen
17.0 22.2 22.0
Weft direction 0.41 0.63 0.71
16.2 18.5 18.8
0.94 0.41 0.85
References Cˇunko, R. (1995), Ispitivanje tekstila, Tekstilno tehnolosˇki fakultet Sveucˇilisˇta u Zagrebu, ISBN 86-329-0180-X, Zagreb. Nikolic´, G. and Sˇomodi, Zˇ. (2000), “Investigation of grip parameters for automated handling using frozen textile in clothing production”, in Katalinic, B. (Ed.), Proceedings of 11th International DAAAM Symposium, October 2000, Opatija, DAAM International, Vienna, ISBN 3-901509-13-5, pp. 333-4. Nikolic´, G. and Sˇomodi, Zˇ. (2001), “Frozen textile workpieces – a step towards robotisation of clothing production”, OIAZ – Oesterreichische Ingenieur- und Architekten-Zeitschrift, Jahrgang Heft, Vol. 147 No 3, pp. 92-5. Nikolic´, G., Sˇomodi, Zˇ. and Agic´, A., (2001), “Loading capacity and bending characteristics of textile frozen for automated manipulation”, Proceedings of the 12th International DAAAM Symposium on Intelligent Manufacturing and Automation: Focus on Precision Engineering, Jena. Rogale, D. and Knez, B. (1988), “Robotisation of sewing process in clothing industry”, Tekstil, Vol. 37 No 10, ISSN 0492-5882, pp. 587-93 (in Croatian). Sˇomodi, Zˇ. and Nikolic´, G. (2000), “Numerical modelling of mechanical grip on thin rigid-plastic plates in automated handling”, in Marovic´, P. (Ed.), Proceedings of 3rd International Congress of Croatian Society of Mechanics, September 2000, Cavtat, Croatian Society of Mechanics, Zagreb, ISBN 953-96243-3-9, pp. 169-75.
Mechanical properties of sewing stitch 203
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
IJCST 15,3/4
204
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
The evaluation of fused knitted systems stability Eugenija Strazdiene and Matas Gutauskas Faculty of Design and Technologies, Kaunas University of Technology, Lithuania Keywords Textiles, Deformation Abstract The aim of the presented research is to apply the method of punch deformation for the simulation of textile systems behaviour in serve conditions and on the basis of it create original method and find new criteria for shape stability evaluation. The research was done with the help of three devices of punch loading originally created at Kaunas University of Technology (Lithuania), which were attached to the standard tensile testing machine. Creep tests were performed by a special device, clamping radius of which was R ¼ 56.5 mm. Creep process was controlled up to the stabilisation of shells height, i.e. after t ¼ 48 h. Tests were carried out in wet and dry states of the specimen. Two different types of textile systems (I+K and K+I) composed of two layers were investigated (where K – outer material; I – interlining).
International Journal of Clothing Science and Technology Vol. 15 No. 3/4, 2003 pp. 204-210 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310478297
1. Introduction In practice of apparel manufacture, systems composed of two textile layers having different properties and bonded together by adhesive joints are wide spread. Such systems (assemblies) improve garment appearance and performance stability. However, from the material science point of view of, there are still many unsolved problems concerning the formation of such systems and the evaluation of their mechanical stability. It is purposive to evaluate the stability of apparel garments or their elements taking into account the effects of all possible external loads (Gutauskas, 1978; Kobliakov, 1973) like the type of loading and ambient conditions. During their performance almost all zones of garments experience the state of biaxial loading and repeated effect of moisture, which essentially change the initial shape of clothing especially in zones such as, elbows or knees. There are a certain number of publications, which propose to simulate external biaxial loading conditions by punch deformation and on the basis of it predict the behaviour of textile materials or their systems (Gutauskas and Masteikaite, 1997; Strazdiene and Gutauskas, 1999). The experience of clothes wearing out in extreme conditions (e.g. soldier clothing which loses its initial shape in rainy seasons), has shown that the degree of loss depends on the mechanical properties of the material (Strazdiene and Gutauskas, 2000). However, more evident changes are observed in clothes made of thicker materials than in those made of thinner ones. It is explained by the ability of textile shell, e.g. at the zones of elbows or knees, to maintain the spatial shape due to different fabric flexibility, which is directly related to its thickness. In wet medium, both types of materials can form shells of similar
size (height), but thicker one visually will be more vivid because it will sustain Fused knitted spatial shape, while the thinner one will deflate and visually will look smaller. systems stability The aim of this research is to adapt the principle of punch deformation for the simulation of biaxial loading conditions of textile systems. Further more, on the basis of it set the testing methodics and find the criteria of textile systems stability evaluation.
205
2. Methodics Three devices of original construction were used for the research (Figure 1). With the help of these devices the strength parameters of fabrics and their systems in punch loading were determined. Later, on the basis of these parameters the values for loading simulation, regimes of long-term creep processes and conditions of spatial shell measurement were selected. Bagging tests were performed by a special device (Strazdiene and Gutauskas, 2001; Strazdiene et al., 1997), which operates together with the tensile testing machine. The ratio between the punch r and specimens R radii was r=R ¼ 0:51 ðR ¼ 28:2 mmÞ: Creep tests were performed with the device (Strazdiene and Gutauskas, 1999, 2001) where R ¼ 56:5 mm and r=R ¼ 0:53 or 0.85. The process was controlled till the moment when the height H of the shell was stabilized, i.e. after t ¼ 48 h and it comprised about 12-17.5 per cent of punch loading falling on the weaker system of the component. Tests were performed in dry and wet conditions. The size of the shell was evaluated on the basis of its height, which was measured by a special device. It allowed to define the value of the shell’s initial deflection H0, the value of shell’s residual height Hr after sustained loading and the height Hl after the rest period of 24 h. To guarantee the reliability of experimental results, the outer contour of the specimen was reinforced by a special ring (Figure 2) after it was moved from the loading device (Figure 1(b)) to the measuring device (Figure 1(c)). The objects of the investigation were two types of knitted fabrics: one outwear and the other interlining material (Table I). Figure 1. Principal schemes of textile shell parameter measurements: (a) parameters H and P of punching tests; (b) parameter H* ¼ f(t) of shell creep process; (c) initial H0 and residual Hr bending heights of tested fabrics and assemblies
IJCST 15,3/4
Two non-fused assemblies A (K+I and I+K) were made out of these fabrics with different position of their layers in respect to the punch. Fused assemblies F were also made, which were effected by the punch from the side of interlining. The size of the shell was evaluated on the basis of the parameter: 1¼
206
H 100 per cent; R
which to our mind characterises spatial deformation of knitted material in the most sufficient way, because it relates the initial size of the specimen R with the deflection height of the specimen (H0 or Hl). 3. Results and discussion The parameters of punching curves H-P of the investigated materials (Figure 3) have shown that interlining material I is twice weaker compared to the main material K. Additionally, its initial rigidity is two times lower (Table II). The shape of both curves is similar, while the shape of forcing through curves of systems I+K and K+I obtains a stepped shape. The stronger component determines the strength P of the whole system, while the weaker determines its deformation H. Now, when the interlining material I fails, the value of force P temporarily falls down and the decrease in value depends on the size and location of tear. Further resistance of the system to the punch loading depends on the location of system components in respect to the punch. In the case when the weaker component is located near the punch, e.g. I+K, its failure is arrested by
Figure 2. The ring to prepare specimens for creep tests
Tensile force (N/mm) Table I. Mechanical characteristics of tested materials
Tensile deformation (per cent)
Materials Wale direction Course direction Wale direction Course direction Thickness (mm) K I
5.65 2.83
1.67 1.63
106.8 97.4
372.3 138.9
1.0 0.66
Fused knitted systems stability
207
Figure 3. Typical punching curves of investigated materials (R ¼ 28.2 mm; r ¼ 14.25 mm; r/R ¼ 0.51)
Materials and systems K I I+K K+I
P (N)
H (mm)
tga
766 415 760 800
48.7 34.5 35.4 37.1
12,428 6,028 8,051 7,634
the stronger component, which lies above sustaining the multistaged character of tear propagation. On the contrary, when the stronger component of the system is located near the punch, the weaker component of K+I fails first and the punch quickly widens the tear zone until the moment when the underlying component is forced through. In the case of I+K system, the highest value of force P is obtained at the end of the punching process. Meanwhile, for the K+I system, two different cases of H-P curve ending can exist (Figure 3). For both systems I+K and K+I, the average value of two peaks of H-P curves P1+P2/2 usually obtains the similar value. In the case of investigated materials, it was 2788 N. The distance between the curve peaks D ¼ H 2 2 H 1 also has the similar values 214.6 and 13.9 mm (Figure 4). On the second stage of investigations, the performance stability of the investigated systems was evaluated on the basis of the parameters of long-term creep process. Experimental results in the form of 1 deformation are presented in Figure 5. The obtained results have proved that all three parameters 1*, 1r and 10 are independent of the ratio r/R and are higher for wet testing
Table II. Strength parameters of investigated materials in punch loading
IJCST 15,3/4
208
Figure 4. Resistance parameters of systems and its components in punch loading when r/R ¼ 0.51
Figure 5. The dependence between the relative values of shell height and the location of disc shaped systems components and testing conditions (r/R ¼ 0.85)
conditions. Only in one case of K material, i.e. r=R ¼ 0:85; the obtained result is contrary, and it can be explained by the swell of wet knitted fabric. The parameter 1r shows the residual height of the shell, which characterises material stability. It must be noted that the values of 1r parameter show the highest differences that reach even 10.5 per cent. Up to the value 1r , 10 per cent the shell visually looks low, but starting from the moment when it reaches 10 per cent it starts to look high. In the investigated loading cases, parameter 10 for separate materials and their systems varied between 5.3 per cent and 17.7 per cent in dry testing conditions and between 10.6 per cent and 21.2 per cent in wet testing conditions, while the parameter 1r varied between 10.6 per cent and 21.2 per cent and between 17.7 per cent and 30.1 per cent, respectively. The most unstable material was interlining fabric I, because its parameter 1r was varying between 17.7 per cent and 30.1 per cent, while the most stable was the non-fused assembly A – 10.6 per cent and
23.0 per cent, respectively. The fused system F in all cases of r/R and in both Fused knitted testing conditions – dry and wet – showed a slight decrease of stability systems stability expressed through the parameter 1r compared to the stability of the non-fused assembly A. The values of parameter 1* in the cases of interlining material I and nonfused assembly A were similar because interlining material in those cases was 209 touching the punch and accepted almost all loading. However, the value of 1r parameter for those two cases differed significantly because in creep process, the second component of the system – knitted material K – reduced the height of the shell. During the investigations, an interesting phenomenon was observed that parameter 1r for fused system F, almost in all cases (especially in wet testing conditions) obtained higher values than those for non-fused assembly A. It can be explained by the fact that the main component of the system was stretchable knitted fabric K, the characteristic feature of which was the significant deformation even at low values of punch loading, i.e. to form high shells H ø R. During the period of rest in dry testing conditions, it practically returned to the initial flat state (10 ¼ 11:5 per cent; 1r ¼ 12:4 per centÞ: The behaviour of the second component of the system – knitted interlining material I with a layer of adhesive dots – under punch loading was different. At the same level of punch loading it formed the shell, the height of which was twice lower than that of the main component K, and during the rest period it did not return to the initial flat state 1r . 10 . In the case of non-fused assembly A, the loading of shell formation is distributed among two components, the main part of which falls upon the interlining material I. Thus, the deformation of the whole non-fused assembly is more close to the deformation of the separate interlining fabric. After a certain period of rest, such assembly returned practically to the deformation level of elastic component. It should be noted that in the case of fused assembly, all three constituent parts of the deformation, i.e. 1*, 1r and 10 are higher because of the effect of adhesive layer, which bonds both components of the assembly into the bisystem. In the rest period, such component loses part of its elastic properties and is not able to return to the initial stage of assembly’s deformation (according to 1r). 4. Conclusions The performance behaviour of non-fused textile assemblies in punch deformation depends upon the location of its components in respect to the loading direction. In the case when the weaker component is located near the punch, the failure of the system is arrested by the stronger component. On the contrary, when the stronger component of the system is located near the punch, the weaker component fails first and the punch widens the tear zone
IJCST 15,3/4
210
until the moment when the underlying component is forced through. In the first case, the force P obtains its highest value at the end of punching process. Meanwhile, for the second system two different cases of H-P curve ending can exist. The investigations of non-fused assemblies and fused ones have proved that all three parameters 1*, 1r and 10 (max height of the shell after the 48 h period of loading, residual height of the shell after 25 h period of rest and initial height of the shell before loading, respectively) are independent of the ratio r/R and are higher for wet testing conditions. Besides, in the case of fused assembly all three constituent parts of the deformation 1*, 1r and 10 are higher because of the effect of the adhesive layer, which bonds both components of the assembly into the bisystem. In the rest period, such component loses part of its elastic properties and is not able to return to the initial stage. References Gutauskas, M. (1978), “The complex of techniques and technical devices for the evaluation of textile products performance stability”, Technology of Textile and Leather Products - VII Proceedings of Lithuanian HER, pp. 109-115 (in Lithuanian). Gutauskas, M. and Masteikaite, V. (1997), “Mechanical stability of fused textile systems”, International Journal of Clothing Science and Technology, Vol. 9 Nos 4/5, pp. 360-6. Kobliakov, A. (1973), Structure and Mechanical Properties of Knitted Materials, Lioghkaja Industrija, Moscow, Vol. 273 (in Russian). Strazdiene, E. and Gutauskas, M. (1999), “The behaviour of woven and knitted materials in uniaxial and biaxial tensile deformations”, Materials Science, Kaunas: Technologija, Vol. 10 No. 3, pp. 48-53. Strazdiene, E. and Gutauskas, M. (2000), “The evaluation of textile stability by the methods of biaxial loading”, The Textiles: Research in Design and Technology - 2000, Proceedings of International Conference, Kaunas: Technologija, pp. 142-7. Strazdiene, E. and Gutauskas, M. (2001), “The peculiarities of textile behaviour in biaxial punch deformation”, International Journal of Clothing Science and Technology, Vol. 13 Nos 3/4, pp. 176-85. Strazdiene, E., Gutauskas, M., Papreckiene, L. and Williams, J.T. (1997), “The behaviour of textile membranes in punch deformation”, Materials Science, Kaunas: Technologija, Vol. 5 No. 2, pp. 50-4.
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Engineering the drapability of textile fabrics George K. Stylios and Norman J. Powell School of Textiles, Heriot-Watt University, Scotland, UK
Engineering the drapability of textile fabrics 211
Keywords Fabrics, Drape Abstract The drape attributes of fabrics, number of folds, depth of folds and evenness of folds were measured together with the drape coefficient. The relationship between these measurements and the subjective evaluation of the fabric drape was modelled for each end-use on a neural network using back propagation, which can correctly predict the grades of 90 per cent of the samples. The relationship between the drape attributes and fabric bending, shear and weight was also modelled using neural networks. It was found that using the natural logarithm of the material property divided first by the weight of the fabric produced the most predictive model. Together, these models provide a powerful predictive tool to determine both the drape attributes and the drape grade from the mechanical properties of a fabric. The accuracy of the prediction of this system was found to be 83 per cent overall. Combining this with a novel feedback system, the drape grade or drape attributes of a fabric can be modified to fit the customer requirements and then the changes to the material properties required to achieve them can be determined.
1. Introduction Traditionally, fabric and garment designers have subjectively assessed the drapability in an informal manner. Chu et al. (1950, 1960) and later Cusick (1965, 1968) attempted to capture this quality by direct measurement using a disc of fabric draped over a circular support plate, which has resulted in the development of the drape meter. A quantity, known as the drape coefficient, was measured. This represented the coverage of unsupported fabric as a proportion of the area it would cover if it were rigid, when viewed from above. Low drape coefficients indicate a highly deformable fabric. Later researchers (Stylios and Zhu, 1997; Vangheluwe and Kiekens, 1993) have automated this measurement, using a camera above the drape meter and capturing the resulting image on a computer for image analysis, as well as investigating other factors associated with drape. Stylios and Zhu (1997) considered that the drape coefficient by itself did not capture the full aesthetic quality of the drape of a fabric. They defined a feature vector (Ballard and Brown, 1982), consisting of the average minima and average maxima fold lengths and the evenness of the folds. Drape grades were assigned according to the closest standard drape samples in the feature vector space. In this work, three additional drape attributes were added to the drape coefficient, namely: number, depth and evenness of folds. These were considered to be closer to the qualities that are unconsciously assessed by a consumer when considering the appearance of a draped apparel product.
International Journal of Clothing Science and Technology Vol. 15 No. 3/4, 2003 pp. 211-217 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310478305
IJCST 15,3/4
212
2. The principles of engineering fabric drape There are three types of fabric evaluation that relate to its drapability. (1) The objective measurements of each fabric’s material properties measured by the Kawabata evaluation system (KES) (Kawabata, 1975, 1982), specifically shear, bending and weight. (2) The attributes of drape measured objectively by the drape meter, namely drape coefficient, number, depth and evenness of folds. (3) The subjective evaluation of the drape of a fabric, relative to its end use. This is an attempt to capture the relationship between the material properties of the fabric and the aesthetics of the appearance of the draped fabric. 2.1 Measurement of fabric mechanical properties The material properties of the fabric are objectively measured by the KES. The material properties relating to the way a fabric drapes were obtained from the COMIT now Research Institute for Flexible Materials (RIFleX) database (Stylios, 1998). There were 312 fabrics with the required set of objective measurements and sufficient available fabric to construct a sample for the measurement of drape. 2.2 Measurement of drape attributes A 254 mm diameter disc of fabric was supported over a 127 mm diameter plate. The following measurements were then made: drape coefficient (coef), number of folds (folds), depth of folds (depth) and evenness of folds (even). The drape attributes were established because of the possibility of two draped fabrics with the same drape coefficient, but a different number of folds. These fabrics would be aesthetically assessed differently (Stylios and Zhu, 1997). It is important to include quantitative measurements that relate directly to the qualitative, cognitive assessment of aesthetic drape. These could then be related to the mechanical properties of the fabric. 2.3 Categorization of fabrics by end-use The fabrics have been divided into categories according to their end-use. This paper will concentrate on the results of the first two categories of fabric end-use for the prediction of drape grade, where there is a large quantity of samples for precise end-uses. However, the same principles and practices have been applied to the other end-use categories. 2.4 Measurement of subjective drape The drape grade is subjectively evaluated by comparing the aesthetic appearance of the fabric to other fabrics with the same end-use. The fabrics are observed in the same situation in which they are measured in the drape meter, i.e., a 254 mm diameter disc of fabric over a 127 mm diameter plate. The aesthetic quality of drape is extremely end-use specific and is also subjected
to variations in taste. For example, in 1950s, tastes tended towards stiffer fabrics producing broader fuller folds, whereas today tastes tend towards very flimsy fabrics that produce many smaller folds. Since it is not our aim to capture this transient aspect of subjective drape, the fabrics were graded according to their degree of drape and not to how much the drape was liked.
Engineering the drapability of textile fabrics
3. Implementation of the model This project aims to find the relationships between three types of fabric evaluation and hence objectify the aesthetic drape of fabric. The relationships work both in the forward direction, for prediction and classification of the drape from the material properties, and in the reverse direction, to obtain the required material properties in order to obtain a specific drape. Figure 1 shows these relationships. The relationship between the measured drape attributes and the subjective drape grades is a non-linear phenomenon. Consequently, it requires the use of modern artificial intelligence techniques to find the relationship. Therefore, a neural network (Nguyen and Widrow, 1990; Rumelhart et al., 1986) was “trained” with the appropriate data. The predictive power of this model was compared with models derived from more tradition techniques of linear regression (Kim and Kohout, 1975) and discriminant analysis (Klecka, 1975), and found to be more successful. The relationship between the measured drape attributes and the mechanical properties of the fabric was also established in the same way. The neural networks (Rumelhart et al., 1986) used have a single hidden layer of neurons between the inputs and the output neuron. The number of neurons used in the hidden layer were generally 2N 2 1 where N is the number of inputs, as generally recommended (Rumelhart et al., 1986). The hidden layer neurons use a bipolar sigmoid function to produce their output, whilst the output neuron is simply linear. The weights and biases are set by training the neural network using a back propagation algorithm (Rumelhart et al., 1986).
213
Figure 1. Schematic of drape relationship
IJCST 15,3/4
214
The least mean square error between the desired outputs and the actual outputs for the full set of training data is calculated and the weights and biases are adjusted to reduce the size of this error. This is employed in conjunction with a momentum term and a variable learning rate, in order to enhance the speed of training and avoid local minima (Rumelhart et al., 1986). The weights and biases were initialised using a method proposed by Nguyen and Widrow (1990) to reduce the training time. 3.1 Performance indicators In order to measure the success rate of the various models when compared with each other, performance indicators were required which could be applied across the different varieties of model. The quantities chosen were the percentage of samples correctly classified and the product-moment correlation coefficient: these measures are described in the following section. 3.1.1 Product-moment correlation coefficient. The product-moment correlation coefficient (Nie et al., 1975) is a measure of the linear relationship between two parameters. The coefficient takes values approaching 1 if there is a strong positive linear relationship, 21 if there is a strong negative linear relationship and 0 if there is no linear relationship. 3.2 Drape attributes to drape grade The first models generated were those relating the drape attributes measured by the drape meter to the subjectively assessed drape grade. Models using the drape coefficient alone were compared with models using all the drape attributes. Two methods of discriminant analysis were used, forced entry: where all the available parameters are used and iterative, where combinations of parameters used are iteratively tested. Two neural networks using the drape coefficient alone were constructed, one with a single hidden layer neuron and another with the same number of neurons as the neural network using all the drape attributes. This was done to distinguish between the effect of increasing the number of neurons from the effect of the additional information provided by the other drape attributes. In general, the neural networks outperform the discriminant analysis, which in turn outperforms linear regression. This suggests that the increasing nonlinearity or flexibility of the approach produces improvements in prediction. To correspond to the relative flexibility of the models involved, the single neuron neural network using the drape coefficient alone is outperformed by the discriminant analysis models using the drape coefficient alone. The concept of this work was to relate the aesthetic attributes in a predictive model that recognises people’s aesthetic appreciation of fabric drape. Furthermore, as it would be made clearer in the following sections of the paper, the incorporation of aesthetic attributes is important for the feedback part of the model. To that effect, we feel that the incorporation of aesthetic drape attributes is justified.
Figures 2 and 3 show a schematic diagram of the drape attribute and drape grade prediction models. 3.3 Model reversal (feedback) Together, the above models provide a powerful predictive tool to find both the drape attributes and the drape grade from the material properties. These models were combined with a novel feedback system (Stylios and Chang, n.d.) to allow the modification of the drape grade and the drape attributes of a fabric
Engineering the drapability of textile fabrics 215
Figure 2. Drape attribute prediction model
Figure 3. Drape grade prediction model
IJCST 15,3/4
216
to be more desirable, and then find the changes to the material properties required to achieve them. The feedback system is a modification of the back propagation algorithm (Rumelhart et al., 1986) that modifies the inputs to the network instead of the weights and biases. Throughout this process, drape attributes and drape grades produced from different sources are displayed together for comparison. 4. Conclusions This paper has proposed a powerful predictive tool to find both the drape attributes and the drape grade from the mechanical properties of a fabric. The accuracy of the prediction of this system was found to be 83 per cent overall. Combining this with a novel feedback system (Stylios and Cheng, n.d.), there is now the possibility to modify the drape grade or drape attributes of a fabric to be more desirable, and then find the changes to the material properties required to achieve them. This opens up the possibility of re-engineering fabrics to match the changing fashion and market requirements and possibly opens a new approach towards the “cloning” of aesthetic attributes of apparel products. The system consists of neural networks, which predict the drape attributes from the natural logarithm of the material properties divided by the weight, and neural networks that predict the subjective drape of a fabric from the drape attributes. References Ballard, D.H. and Brown, C.M. (1982), Computer Vision, Prentice-Hall, Englewood Cliffs, NJ, USA. Chu, C.C., Cummings, C.L. and Teixeira, N.A. (1950), “Mechanics of elastic performance of textile materials; Part V: a study of the factors affecting the drape of fabrics – the development of the drape meter”, Textile Research Journal, Vol. 20, pp. 539-48. Chu, C.C., Platt, M.M. and Hamburger, W.J. (1960), “Investigation of the factors affecting the drapability of fabric”, Textile Research Journal, Vol. 30, pp. 66-7. Cusick, G.E. (1965), “The dependence of fabric drape on bending and shear stiffness”, Journal of the Textile Institute, Vol. 56, pp. T596-T606. Cusick, G.E. (1968), “The measurement of fabric drape”, Journal of the Textile Institute, Vol. 59, pp. 253-60. Kawabata, S. (1975), Standard Hand Evaluation, Hand Evaluation and Standardization Committee, Japan. Kawabata, S. (1982), “The development of the objective measurement of fabric handle”, in Kawabata, S., Postle, R. and Niwa, M. (Ed.), Objective Specification Fabric Quality, Mechanical Properties and Performance, The Textile Machinery of Japan, pp. 31-59. Kim, J-O. and Kohout, F.J. (1975), “Multiple regression analysis: subprogram regression”, in Nie, N.H., Hull, C.H., Jenkins, J.G., Steinbrenner, K. and Bent, D.H. (Eds), Statistical Package for the Social Sciences, Chapter 20, 2nd ed., ISBN 0-07-046531-2, McGraw-Hill, NY, USA, pp. 320-67.
Klecka, W.R. (1975), “Discriminant analysis”, in Nie, N.H., Hull, C.H., Jenkins, J.G., Steinbrenner, K. and Bent, D. (Eds), Statistical Package for the Social Sciences, Chapter 23, 2nd ed., ISBN 0-07-046531-2, McGraw-Hill, NY, USA, pp. 434-67. Nguyen, D. and Widrow, B. (1990), “Improving learning speed of 2-layer neural networks by choosing initial values of the adaptive weights”, International Joint Conference of Neural Networks, Vol. 3, pp. 21-6. Nie, N.H., Hull, C.H., Jenkins, J.G., Steinbrenner, K. and Bent, D.H. (1975), “Bivariate correlation analysis: pearson correlation, rank order correlation, and scatter diagrams”, Statistical Package for the Social Sciences, Chapter 18, 2nd ed., ISBN 0-07-046531-2, McGraw-Hill, NY, USA, pp. 276-300. Rumelhart, D.E., Hinton, G.E. and Williams, R.J. (1986), “Learning internal representation by error propagation”, in Rumlhart, D.E. and McClelland, J. (Eds), Parallel Data Processing, Chapter 8. MIT Press, 1, pp. 318-62. Stylios, G. (1998), “Centre for objective measurement and innovation technologies (COMIT) for the textile clothing and retailing industries”, International Journal of Clothing Science and Technology, Vol. 10 No. 6, pp. 5-6. Stylios, G.K. and Cheng, L. (n.d.), “Engineering the tactile properties of fabrics for intelligent garment manufacture”, (in preparation). Stylios, G.K. and Zhu, R. (1997), “The characterisation of the static and dynamic drape of fabric”, Journal of the Textile Institute, Vol. 88 No. 4, pp. 465-75. Vangheluwe, L. and Kiekens, P. (1993), “Time dependence of the drape coefficient of fabrics”, International Journal of Clothing Science and Technology, Vol. 5 No. 5, pp. 5-8.
Engineering the drapability of textile fabrics 217
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
IJCST 15,3/4
218
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Dynamics of moisture vapour and liquid water transfer through composite textile structures Natalya Suprun Kiev National University of Technology and Design, Ukraine Keywords Protective clothing, Moisture, Films Abstract A combination of two semipermeable microporous hydrophilic polymer films, inserted between two different fabrics (internal and external), proposed for some kinds of protective clothing, when it is necessary to protect the room atmosphere against particles (causing illness) emitted from human body, and to ensure comfort in using.
International Journal of Clothing Science and Technology Vol. 15 No. 3/4, 2003 pp. 218-223 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310478314
1. Introduction Some kinds of protective clothing are used to minimise the influence of spread of the products of metabolism into the environment (e.g. clean rooms garments, some types of medical clothing). Such clothes are, as a rule, multilayered and could be presented (Figure 1) as a sum of air (l01, l12) and textile (l1, l2) layers with their own properties – temperature (T ), pressure (P ), humidity (w), heat transfer coefficient (l) etc. Textiles for upper layer of such kinds of clothes typically are woven with a multifilament yarn in a limited range of porosity and permeability. So they often have very unsatisfactory comfort, especially in the active case. At a given activity level, the temperature and humidity ranges in which the garment will be comfortable are dependent on the thermal insulation of the garment material and the rate at which the material will allow moisture vapour to pass through it. The human body regulates its temperature to a large degree by the evaporation of perspiration. If a moistured vapour in the underwear space is damp, the body cannot cool itself through latent heat of evaporation, and the wearer may feel hot and uncomfortable. The influence of buffering by hygroscopic fabric on removal of moisture from the clothing underwear and estimation of heat associated with it is the important practical and scientific problem, discussed, firstly, many years ago (Cassie, 1962). When the relative humidity and or temperature gradient changes, textile material absorbs or desorbs moisture from or to the adjacent air and, in the clothing microclimate, this acts as a buffer against climatic changes. On the other hand, the effectiveness of hygroscopic buffering seems to vary widely, depending on the differing test conditions, textile materials, films (semipermeable microporous membrane between textile material layers)
Dynamics of moisture vapour
219
Figure 1. Multilayered protective clothing
employed by different researchers (Barnes and Holcombe, 1996; Hong et al., 1988; Kim, 1999; Kim and Spivak, 1994; Li et al., 1992; Scheurell et al., 1985; Wang and Yasuda, 1991; Yasuda et al., 1992). The common, nearly trivial conclusion is that due to its high sorbing capacity, wool textile material removes moisture more effectively from the underwear space than does, for instance, polyester, and the wearer perceived a relatively more dry and pleasant microclimate in wool clothing. Various new breathable waterproof textiles have been introduced to minimise wearer heat stress through effective moisture vapour transfer while blocking external water penetration. It is known that when a fibre absorbs moisture, a certain amount of heat is liberated. Moreover, the changes in fibre dimension caused by moisture absorption can lead to changes in thickness and porosity. It is shown by Gretton et al. (1996) and Matsudaira and Kondo (1996), that water vapour and liquid transmission cannot be realistically examined in the absence of temperature. The desirable performance characteristics of fabrics are often achieved by inserting a layer of a semipermeable microporous
IJCST 15,3/4
220
membrane between fabric layers, laminating with microporous membranes or coating the fabric with hydrophilic substances (Gretton et al., 1996; Kim, 1999; Matsudaira and Kondo, 1996). The transport properties of these hydrophilic polymers depend also on the temperature and humidity on either sides of the polymer, so they do not offer a fixed resistance to the passage of water vapour and liquid (Farnworth et al., 1990; Goosey, 1985; Lomax, 1990; Osczevski, 1996; Osczevski and Dolhan, 1989; Stannett and Vasuda, 1965). The assessment of water transport through hydrophilic polymers is therefore highly influenced by the test conditions, which are usually adopted as isothermal ones. Only a few studies (Gibbson et al., 1995; Kim, 1999; Sullivan and Mekjavic, 1992) have examined the effect of semipermeable film characteristics in actual functional protective clothing assemblies of the type: fabric-polymer film-fabric (f-p-f). On our opinion, it is necessary to examine the effect of film characteristics (hydrophilicity, thickness, pore size and porosity of semipermeable membranes) on the surface temperature changes and moisture vapour-liquid transfer of a novel type of assemblies: fabric-double polymer-films construction-fabric (f-dp-f). We consider, that a combination of two hydrophilic polymer films, inserted between two different fabrics (internal and external), can be very useful in solving the problem of creation of textiles for some kinds of protective clothing, when it is necessary to protect the room atmosphere against particles (causing illness) emitted from human body. On the other hand, such textile must act as a barrier protecting the human body against dust particles ( possible carriers of micro-organisms) from atmosphere. In the active case, water vapour and liquid water (also called sensible perspiration) transfer simultaneously through the hygroscopic internal fabrics up to the non-hygroscopic (polyester type) external fabrics (ef ) - barriers. It is known that one of the defects of polyester fibres lies in their non-absorbency of water. Hence many devices have been tried to improve the above-named defect of polyester fibres. It is expected, for example, that warm and lightweight clothes will be achieved by making a grooved and/or a non-grooved hollow in a fibre. Unfortunately, it is obvious that most of these devices are not applicable in the clean room textile creation. We consider the possibility to incorporate the dp-construction between the if and ef layers to accumulate in this storage water, mainly in a liquid phase from the human body as well as (in a small quantity) from an atmosphere. The comparison of the proposed layered fabrics structure: if-dp-ef with the conventional usage of a semipermeable polymer film: if-p-ef- is represented in Figure 2 for multilayered protective clothes. Although the steady-state experiments for measuring the moisture transmission rate provide reliable heat and mass transfer data for non-active case, they cannot explain wetness or moisture related subjective sensations that determine human comfort. Consequently, dynamic moisture vapour transport through the textiles for protective clothes must receive special
Dynamics of moisture vapour
221 Figure 2. Structure of multilayered protective clothes. (a) Standard type (if-p-ef structure); (b) proposed type (if-dp-ep composite structure)
attention. In this problem, the estimation of real transient period (time range) which is required for filling water in the storage of an internal volume for the proposed dp-construction (Figure 2(b)) is very important. The duration of the transient behavior has to be strongly dependent on the moisture sorption capacity of both dp-construction sides as well as on the empty volume (at the start of measurement) between two semipermeable membranes. The development of theoretical prediction by a reasonable model and comparison with the experimental data is important as well. One can interpret the dp-construction (two closely located semipermeable polymer membranes, which provide the accumulation of water from both internal and external fabrics) and liquid water inside it as the composite structure. The adequate model has to combine the conservation of heat and mass with the sorption kinetics in the porous materials. With the application of appropriate boundary conditions, the model must be versatile enough to be applied to many different clothing situations. A detailed description of the derivation and application of the novel model for nearly non-active, stationary case has been published in the works of Rogankov and Suprun (1999) and Suprun and Rogankov (2000). Relations between physical properties, heat of sorption and latent heat of phase transition (condensation in the small pores), thermal conductivity, isobaric thermal expansion and isothermal compressibility, moisture concentration of the surrounding air have been accounted to solve the coupled heat and moisture transfer equation. The solution has been based on the standard analogy of transport processes with the equivalent electrical circuits (Figure 3). Here, we consider damp air underwear layers as conductors having some thermal resistance (l1/l1, l2/l2)E, and, on the contrary, the air layers to characterise “moistened resistance”: (l01/l01, l12/l12)M.
IJCST 15,3/4
222 Figure 3. Conditional circuit of the moisture (a) and heat (b) balance in multilayer protective clothing
Entering the time parameters tM and tE, and using a set of standard mathematical transformations, we received equations for characteristic time scales tE – for heat transferee, tM – for diffusion. The derivation of the appropriate relationships has been fully described elsewhere and the relevant, explicit equations (1) and (2) for the duration of the temporal characteristics have been represented (Rogankov and Suprun, 1999; Suprun and Rogankov, 2000). l1 l2 l 01 C P01 r01 l 12 C P12 r12 tE ¼ þ ð1Þ l1 l2 l 01 C P01 r01 þ l 12 C P12 r12 tM ¼
l 01 r01 l 12 r12 l 1 K T1 r1 l 2 K T2 r2 þ g01 g12 lK T1 r1 þ l 2 K T2 r2
ð2Þ
They connect dissipative (l,g) and fluctuational (CP, KT) characteristics and generalise a multilayed structure of expression for effective temperature and moisture transferee (equation (3)): tE ¼
l 2 rC P l
tM ¼
l 2KT r 2 g
ð3Þ
We propose the modification of our model (Rogankov and Suprun, 1999; Suprun and Rogankov, 2000) on the basis of the semiconductors insertion into the above-named equivalent electrical circuits. The time of blocking up in such a circuit is an equivalent transient period, which is necessary to achieve the (undesirable) sense of discomfort in the protective clothing textiles. References Barnes, J.C. and Holcombe, B.V. (1996), “Moisture sorption and transport in clothing during wear”, Text Res. J., Vol. 66, pp. 771-86. Cassie, A.B.D. (1962), “Fibers and fluids”, J. Textile Inst., Vol. 53, pp. 739-45.
Farnworth, B., Lotens, W.A. and Wittgen, P. (1990), “Variation of water vapour resistance of microporous and hydrophilic films and relative humidity”, Text. Res. J., Vol. 60, pp. 50-3. Gibbson, P., Kendrick, C., Rivin, D., Sicuranza, L. and Charmchi, M. (1995), “An automated water vapour diffusion test method for fabrics, laminates and films”, J. Coated Fabrics, Vol. 24, pp. 322-45. Goosey, M. (1985), Polymer Permeability, in Comyn, J. (Ed.), Chapter 8. Elsevier Applied Science Publishers, London. Gretton, J.C., Brook, D.B., Dyson, H.M. and Harlock, S.C. (1996), “Moisture vapour transport through waterproof breathable fabrics and clothing systems under a temperature gradient”, Text. Res. J., Vol. 68, pp. 936-94. Hong, K., Hollies, N.R.S. and Spivak, S.M. (1988), “Dynamic moisture vapour transfer through textile. I. Clothing hygrometry and the influence of fibre type”, Text. Res. J., Vol. 58, pp. 697-706. Kim, J.O. and Spivak, S.M. (1994), “Dynamic moisture vapour transfer through textiles. II. Further techniques for microclimate moisture and temperature measurement”, Text. Res. J., Vol. 64, pp. 112-21. Kim, J.O. (1999), “Dynamic Moisture Vapour Transfer Through Textiles III: Effect of film characteristics on Microclimate Moisture and Temperature Changes”, Text. Res, J., Vol. 69, pp. 193-202. Li, Y., Holcomb, B.V. and Apcar, F. (1992), “Moisture buffering behaviour of hygroscopic fabric during wear”, Text. Res. J., Vol. 62, pp. 619-27. Lomax, G.R. (1990), “Hydrophilic Polyurethane Coatings”, J. Coated Fabrics, Vol. 20, pp. 88-107. Matsudaira, M. and Kondo, Y. (1996), “The Effect of a grooved hollow in a fibre on fabric moisture- and heat-transport properties”, J. Text. Inst., Vol. 87, pp. 409-16. Osczevski, R.J. (1996), “Water vapour transfer through a hydrophilic film at subzero temperatures”, Text. Res. J., Vol. 66, pp. 24-9. Osczevski, R.J. and Dolhan, P.A. (1989), “Anomalous diffusion in a water vapour permeable waterproof coating”, J. Coated Fabrics, Vol. 18, pp. 255-8. Rogankov, V.B. and Suprun, N.P. (1999), “About some peculiarities and concepts of the overalls designing”, Problems of the Light and Textile Industry of Ukraine, Vol. N2, pp. 215-27. Scheurell, D.M., Spivak, S.M. and Hollies, N.R.S. (1985), “Dynamic surface wetness of fabrics in relation to clothing comfort”, Text. Res. J., Vol. 55, pp. 394-9. Stannett, V. and Vasuda, H. (1965), Testing of Polymers, in Schmiz, J.V. (Ed.), Chapter 13, Interscience Publishers, Vol. 4. Sullivan, P.T. and Mekjavic, I.B. (1992), “Temperature and humidity within the clothing microenvironment”, Avia Space Environ. Med., pp. 186-92. Suprun, N.P. and Rogankov, V.B. (2000), “About the comfort of protective garments”, 4-th Int. Symp. EL-TEX 2000, 26-27 October 2000, Lodz. Wang, J.H. and Yasuda, H. (1991), “Dynamic water vapour and heat transport through layered fabrics. I. Effect of surface modification”, Text. Res. J., Vol. 61, pp. 10-20. Yasuda, T., Miyama, M. and Yasuda, H. (1992), “Dynamic water vapour and heat transport through layered fabrics. II. Effect of chemical nature of fibres”, Text. Res. J., Vol. 62, pp. 227-35.
Dynamics of moisture vapour
223
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
IJCST 15,3/4
224
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Some aspects of medical clothing manufacturing Natalya Suprun, Victoria Vlasenko and Yulia Ostrovetchkhaya Kiev National University of Technologies and Design, Ukraine Keywords Clothing, Medical, Quality, Questionnaires Abstract On the basis of data of interrogation, the peculiarities of a dentist’s requirements of medical clothes were defined. The most significant quality parameters of textiles for such clothes were estimated.
International Journal of Clothing Science and Technology Vol. 15 No. 3/4, 2003 pp. 224-230 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310478323
1. Introduction One of the basic functions of medical clothes is to protect the medical personnel against a risk of biological infection by micro-organisms (viruses, microbes, fungi). It is important to maintain the hygiene of such clothes, as they are used for not less than 6 h a day. Besides, the medical clothes should have the whole complex of consumer and operational properties, and should meet the requirements of all standards and managing documents on the protection of work. The specified functions allow one to consider medical clothes as the original barrier between human’s body and environment. Furthermore, such clothes should rule out the possibility of an opportunity to forecast an occurrence of a cross infection. If special measures are not undertaken, these clothes can be the basic centres for infections in a hospital environment, and render a certain influence on the occurrence and distribution of infectious diseases transmitted by contact and by air. Such an approach determines the rigid requirements for designing medical clothes, choice of textiles for their manufacture, selection of package of clothes, maintenance of correct conditions for their operation and methods of cleaning. The correctly chosen medical clothes and accessories provide the effective way of protection for medical personnel and patients. In addition, they also create comfortable working conditions, by having a positive influence on the personnel. Designing clothes for a certain category of medical personnel must ensure the maximal degree of required properties. The creation of the barrier clothes for dentists, which are exposed to the first degree of risk for infection during their work, is an urgent task. The significant part of the requirements to clothes is satisfied by the rational choice of materials for them. The complexity of this problem lies in the difficulty of simultaneous maintenance of a group of properties, for example,
barrier, physical, hygienic and ergonomic. To avoid pollution, materials for Medical clothing medical clothes must not be accumulated, should easily be cleaned by usual manufacturing methods (washing or sterilisation). Besides, such materials should be stable against friction and preserve their properties during the entire operation. This task could become practically impossible if traditional textile materials are used. The textile materials of new generation with the given properties, if 225 properly selected, ensure a rational combination of required qualities of medical clothes. Currently, in the world market of modern medical clothes, textiles used are made from natural (cotton, linen) and synthetic (polyester, polyamide) fibres, and also from materials consisting of mixtures of natural and synthetic fibres. Until now, the traditional clothes from cotton fabrics are highly appreciated by the medical personnel for their comfort. However, cotton fabrics have the ability to accumulate pollution and micro-organisms from air, and with their high hygroscopicity they provide a favourable medium for the duplication of micro-organisms. Also, the clothes made up of cotton fabrics cannot protect the penetration of blood and other liquids. The ability for the survival of micro-organisms is defined by some characteristics of the textile material structure, their ability for electrisation, porosity, hygroscopicity, presence of moisture, dust and organic pollution. In view of the necessity to protect against cross infection, the modern line of manufacture of medical clothes uses synthetic and mixed textile materials. Their low hygroscopicity creates adverse conditions for the survival of microorganisms. The best fabrics are those made from a mixture of cotton and polyester fibres, combining advantages of natural and synthetic materials. The clothes from such fabrics do not have the ability for electrisation, can be easily cleaned by washing and sterilisation, quickly dry up and require minimal ironing. The big advantage of medical clothes over the synthetic and mixed fabrics is in their high resistance for friction, stability of appearance and low crumpling. Among the textile materials used for manufacturing medical clothes, it is necessary to select fabrics from ultra thin polyester and polyamide fibres having a complex of unique properties. They are water-repellent, but at the same time pass the outgoing moisture from the human body, thus providing comfort. They also create a barrier for the penetration of micro-organisms. Clothes made from such fabrics have good hygiene, as they are easily washed, disinfected and ironed. Their properties are kept practical during the entire operation (not less than 300 washings or not less than 50 cycles “washing sterilisation”). The barrier function of medical clothes helps in protecting personnel from bacteria, physiological liquids, various secretions of man, aggressive liquids (acids, alkalis, solvents, chemical and medicinal substances) and general
IJCST 15,3/4
226
pollution. A dentist uses medical clothes at an average of 6-8 h a day in a week. Hence, the medical clothes should not only give protection, but should also be comfortable to wear. When a dentist is at work, it is the top layer of his medical cloth that gets dirty due to biological particles, blood, saliva, chemical reagents and medicinal substances. However, he should always be dressed in ideal clean clothes. Therefore, such clothes must have the ability for easy and fast cleaning either by sterilisation, washing, or chemical cleaning, subsequently maintaining a big account of such cycles without the loss of functional properties. For the improvement of medical clothes, in order to satisfy both the industrial and consumer requirements, we carried out a sociological interrogation. Thirty eight dentists from various medical establishments of Kiev with a work experience of not less than 8 years took part in the interrogation. The main purpose of this interrogation was to understand the dentist’s working atmosphere, and to find the basic reasons for the dissatisfaction on the existing home market medical clothes. It was established that during work, doctors are exposed to a large physical and psychological stress. The size of a working zone (together with the equipment) makes up 5 m2. In most cases, temperature of an environment measures approximately 20-228C. The duration of the everyday use of medical clothes makes an average of about 6-8 h. The basic cloth care includes bleaching at periodic intervals, starching and ironing. The majority of interrogated personnel (65 per cent) have said they basically lacked these clothes. Besides, 25 per cent interrogated have remained dissatisfied by the constructive decision and 10 per cent by design. The following are some of the unsatisfactory properties of the medical clothes, which prevent the personnel from using them: . protective (concerning influence and penetration of blood, medicinal solutions, alkalis and acids); . ergonomic (low air permeability, rigidity); . operational (change of the linear sizes after washing and sterilisation, easy crushes, electrisation, ability for pills formation, change of colour during service). After data processing, it was revealed that the lady dentists (30-35 years old) with higher education and with an average financial prosperity, were the potential customers. As to the reasons for purchasing a new medical cloth, 100 per cent of the respondents said the older one was physically deteriorated. Some of them attributed the new purchase to the changes in their body sizes and also change of fashion style. Taking into account the wide colour range of materials intended for the manufacture of medical clothes, it was interesting to determine the colours
that the dentists prefer. The majority of respondents (45 per cent) preferred Medical clothing white coloured medical clothes, followed by turquoise (17 per cent) and pink manufacturing (8 per cent). The dark green colour, which is currently used for manufacturing some types of medical clothes is not very popular among dentists. The results of interrogation is shown in Figure 1. At present, the upper part of medical clothes are made mainly from the 227 synthetic and mixed materials. The underwear is usually produced from hygroscope (cotton, viscose) fibres. For the purpose of optimisation of the clothing system for dentists, it is necessary to define the influence of the type of clothes and their protective and hygienic properties on the packages. The hygienic requirements of medical clothes are directed on the maintenance of the normal mass (gauss, vapour) exchange between the underwear space and the environment to provide the normal level of temperature for body, skin and skin breath. These requirements are made possible by the application of textile materials with optimum air and vapour permeability, hygroscopicity, water sorption, capillarity and some other properties. For a concrete definition of a dentist’s requirements to the properties of medical clothes, some questions were proposed during interrogation. The degree of co-ordination of the ideas of 23 experts was determined as an essential element (concordation coefficient W ¼ 0:63Þ: After the analysis of the results of the questionnaire, 17 parameters pertaining to the quality of materials were allocated, which characterised the following groups of properties: protective, hygienic, convenience of usage and reliability. The example of the questionnaire is presented in Table I. After mathematical processing by the expert method, the most significant parameters of quality was estimated (Table II). The analysis of results of the carried out interrogation has allowed to reveal the peculiarities of a dentist’s working conditions, reasons of the discrepancies in using medical clothes, and to allocate parameters for the quality of textile materials for manufacturing such clothes.
Figure 1. Colours of medical cloth materials which dentists prefer
IJCST 15,3/4
Some properties of the textile materials, having various fibrous structures, interlacing, thickness and superficial density, which are intended for manufacturing medical clothes were investigated. Their structural characteristics are shown in Table III.
228 The name of the parameter, unit of measurements
Table I. Example of the questionnaire
1. Stability to acids, alkalis and organic solvents, per cent 2. Filtering ability, per cent 3. Bacteriological contamination and stability to action of micro-organisms, per cent 4. Dust permeability 5. Ability for dust accumulation, per cent 6. Air permeability, dm3/m2 s 7. Hygroscopicity, per cent 8. Capillarity, mm 9. Ability for electrisation, Ohm 10. Water sorption, per cent 11. Absence of necessity of bleaching, per cent 12. Rigidity at a bend, mN sm2 13. Resistance to crumpling, per cent 14. Change of the linear sizes after washing, per cent 15. Stability to washing, per cent 16. Stability to mechanical influences, per cent 17. Absence of pilling, sm2 1
The name of a parameter, unit of measurements
Table II. The most significant parameters of quality of materials for dentist’s medical clothes
1. Stability to acids, alkalis and organic solvents, per cent 2. Filtering ability, per cent 3. Ability for dust accumulation, per cent 4. Air permeability, dm3/m2s 5. Rigidity at a bend, mN sm2 6. Resistance to crumpling, per cent 7. Stability to washing, per cent 8. Ability for electrisation, Ohm 9. Capillarity, mm 10. Dust permeability, per cent
Conditional designation
Rank
X1 X2
1 2
X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13
14 11 3 4 5 10 8 13 12 6 7
X14 X15 X16 X17
15 9 17 16
Parameter significance
P
0.16 0.14 0.13 0.10 0.08 0.07 0.07 0.06 0.05 0.05 = 1.00
Structure elements (yarn or stitch) count (/10 cm) Sample
Composition
Surface density Thickness Twarp Interlacing (G/m2) (mm) (tex)
Tweft (tex)
Medical clothing manufacturing
229
Warp Weft
Fabrics Cotton 100 per cent Cotton 35 per cent; PE-65 per cent Cotton 35 per cent; PE-65 per cent Cotton 35 per cent; PE-65 per cent Cotton 35 per cent; PE-65 per cent Cotton 33 per cent; PE-67 per cent Cotton 20 per cent; viscose 40 per cent; PE-40 per cent PE 100 per cent PE 100 per cent Knitted textile Cotton 100 per cent Cotton 50 per cent; PE-50 per cent Viscose 50 per cent; PP-50 per cent
Plain
201
0.41
228
179
228
179
Plain
165
0.30
406
208
406
208
Twill 1/2
213
0.32
390
291
390
291
Twill 1/4
170
0.31
522
278
522
278
Twill 1/3
181
0.30
529
288
529
288
Twill 1/2
163
0.33
258
423
258
423
Plain Twill 1/2 Plain
156 129 169
0.40 0.30 0.32
451 407 410
181 239 209
451 407 410
181 239 209
151
0.50
150
180
311
0.60
97
137
340
0.80
95
136
As the medical clothes are basically worn atop the underwear, it was ideal to investigate the materials for this layer of clothes, with various fibrous structure and parameters of knitting (samples N10-12), and also their packages (Table IV). The air permeability of materials was determined at DP ¼ 50 Pa: Experimental data of air permeability of packages was compared with that calculated on the Kleyton formula and a good agreement was found. On the basis of the data collected, three types of cloth packages (the upper fabric +knitted underclothes material) were chosen (No. 2+ No. 11), (No. 8+No. 11) and (No. 8+No. 12). Also three types of sets of clothes were produced and given to dentists for experimental use.
Table III. Structural characteristics of textile materials for medical clothes
IJCST 15,3/4
Air permeability of packages (dm3/m2 s) Number of sample in underwear layer (experimental value)
230 Sample
Table IV. Air permeability of textile materials and their packages for medical clothes for dentists
1 2 3 4 5 6 7 8 9 10 11 12
Number of sample in underwear layer (calculated value)
Air permeability (dm3/m2 s)
10
11
12
10
11
12
107 48 20 50 50 48 136 150 46 165 625 456
66 37 18 40 36 36 80 80 36 – – –
90 42 20 50 45 43 100 120 42 – – –
80 40 18 45 40 40 70 100 40 – – –
64 37 18 38 38 34 74 80 36 – – –
90 44 20 46 47 40 112 123 43 – – –
85 43 20 45 45 40 104 113 42 – – –
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Investigations of the relation between fabric mechanical properties and behaviour Daniela Zavec Pavlinic´ and Jelka Gersˇak
Fabric mechanical properties 231
Department of Textiles, Faculty of Mechanical Engineering, University of Maribor, Smetanova, Maribor, Slovenia Keywords Clothing, Mechanical properties, Fabric, Manufacturing Abstract In transforming fabrics into garment it is necessary to know, besides the manner of processing, the behaviour of the fabric in particular manufacturing processes. It is necessary to define why and how fabrics behave in a particular way when exposed to various strains. The answers to these questions are obtained by investigating fabric mechanics, as non-linear mechanical fabric properties at lower strains, which is the case in transforming fabrics into garments. The area to be investigated is quite wide and the investigations presented here deal only with the most important elastic strains occurring in processing fabrics into garments, such as tensile, pressure, shear and bending, as each individual type of strain bears specific importance in studying fabric behaviour, as well as in garment quality control. Strains impacting the fabric, i.e. the reaction of the fabric to these strains, are presented through the parameters of mechanical properties. A relation is also explained between characteristic histeresis curves and fabric behaviour in real garment manufacturing processes, obtained through recording fabric behaviour in particular garment manufacturing processes. Results obtained through the investigations of mechanical properties of the fabrics analysed and their behaviour in garment manufacturing processes helped to determine the so-called critical, or border values for particular parameters of mechanical properties.
1. Introduction Knowledge of fabric properties and their behaviour in the process of transforming into an article of clothing is a valuable information for garment manufactures, which was unavailable until now. This is why studying mechanics, i.e. mechanical properties of the materials incorporated into articles of clothing – woven fabrics in particular – is a direction of development that should be paid special attention. Furthermore, understanding fabric behaviour from the point of view of mechanics is highly demanding, as fabrics are nonhomogeneous materials, and establishing rules of their behaviour is quite a complex task. However, analysis of a number of fabrics, both from the point of view of mechanics and regarding their behaviour in processing, gives a sound basis and directions for further development. Fabrics are exposed to various strains when processed into garments and they behave in different ways. Fabrics are stretched when spread, they are exposed to compression in cutting because of the vacuum present, and in sewing they are transformed from two-dimensional to three-dimensional articles of clothing. They are again stretched during ironing and shaped into
International Journal of Clothing Science and Technology Vol. 15 No. 3/4, 2003 pp. 231-240 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310478332
IJCST 15,3/4
232
the required three-dimensional form, to ensure a high-quality appearance of the garment manufactured. The influences fabrics are exposed to are called strains, and their nature should be seeked in the field of solid body mechanics, to be applied to textiles (woven fabrics), which are non-homogeneous, anisotropic materials. General rule for such materials is that their properties are considerably changed under moderate strains. Detailed determination of particular parameters of mechanical properties in the investigations presented is based on the parameters obtained by the KESFB measuring system. The purpose of the investigations is to find a relationship between fabric description from the point of view of its mechanics on one hand and its behaviour in transforming it into garments on the other. The relation was investigated on 300 woven fabrics analysed, intended for ladies’ outerwear. 2. Fabric mechanical properties 2.1 Tensile and shear properties Woven fabric, as non-homogeneous materials, behaves like mesh models in investigating tensile properties. It means that in the situation when exposed to no particular strain they consist of two equally spaced yarns, crossing at a right angle (Kilby, 1963). This is especially true for plain (cloth) weave, where the mesh is simple and the interlacing points of warp and weft yarns are completely uniform. If such a mesh is exposed to homogeneous strains, its tension components can be expressed as a linear function of the co-ordinates. Straight lines in the mesh without strain remain straight in the distorted mesh as well. However, although the yarns are still parallel in any given direction, the mesh itself can stretch or shrink, whatever the case may be. An example of a mesh is shown in Figure 1(a) and the mesh that is exposed to normal strain
Figure 1. Simple tensile (b, c) and shear (d, e) strains
sx, x and sy, y, are shown in Figure 1(b) and (c), respectively. Figure 1(d) and (e) shows the mesh that is exposed to simple shearing strain ty, x and tx, y, respectively (Kilby, 1963). In the case of simple shearing strain, homogeneous materials exhibit linear shear histereses. However, different forms of shear histereses are encountered with textile fabrics (Figure 2), depending upon construction parameters of the fabric in question (weave, yarn density, cover factor). Shear histeresis is defined as the force of friction occurring among interlacing points of the system of the warp and weft yarn when they are moving over each other, having their origin in the forces of stretching/shrinking, since the system of warp and weft yarns stretches/shrinks under strain. In the case of closely woven fabrics, there is not much slippage between warp and weft yarns under shearing strain, the result being just a higher friction between individual yarns. More loosely woven fabrics, with lower cover factor, exhibit lower friction between warp and weft yarns. Shear histeresis for such fabrics assumes the state of the letter S, which means that shearing strain grows quite rapidly in the beginning, which causes slippage between warp and weft yarns, then remains stable, and again grows considerably by the end of the straining action (Lindberg et al., 1961). There are also cases of shear histeresis that can be described as convex or concave, explained by the index that defines the ratio of the energy used in shearing strain and energy calculated through the shear angle, under the condition that the shear histeresis is a linear one. Higher ratio means convexity, lower concavity, while the value of 1 means that the histeresis is linear (Lindberg et al., 1961).
Fabric mechanical properties 233
Figure 2. Shear histeresis
IJCST 15,3/4
234
2.2 Bending properties The first definitions of bending properties were offered by Peirce (1930). He defined bending length c as the length of the part of the fabric that bends to a particular degree under its own weight. Consequently, firmer (more rigid) fabrics have a higher bending length, and the value is a measure of the fabric property that determines its drape. The first definition of flexural rigidity is the fabric resistance to bending, calculated for mass per surface area unit. When bending length c and fabric mass W are known, bending rigidity can be calculated as follows (Cooper, 1960; Peirce, 1930): B ¼ Wc 3
ð1Þ
Fabric thickness, together with bending length and fabric mass, influences flexural rigidity. Doubled thickness of a fabric can result in a eight times higher bending rigidity. Investigations of textile fabric bending properties are based on the concept of bending solid bodies. It can be illustrated with the example of a stick which is exposed to bending forces in one direction, while the system of internal forces is reduced to only two forces, both acting in the x axis direction, as can be seen in Figure 3 (Bona, 1994). When bending is realised outside the axis, which is the only possibility in the above case, the cross-section of the bending surface follows one of the basic inertia axes of the section investigated. While rotating around the axis, the effect of flexural forces is pronounced in each section of the stick in the direction perpendicular to the x axis. The area above the symmetry surface, where neutral forces act (neutral zone), is exposed to the forces of compression, while its front part is exposed to tensile forces (neutral zone is not exposed to forces Figure 4). Testing bending properties of textile fabrics employing the KES-FB measuring system is also based on the bending concept described, since testing sample bends following a circle, as can be seen in Figure 5 (Kawabata, 1980). Woven fabric bending is definitely not linear, since fabrics are made of yarns and fibres, which have a high degree of freedom of movement in relation to each other within the fabric structure. Forces acting in fabric bending are considerably lower than those in paper bending, since textile fabric flexural rigidity is considerably lower, due to a higher fabric flexibility. Potential fabric flexibility, as well as inner friction among fibres, directly associated with movements, is of crucial importance for non-linear bending behaviour of fabrics. Consequently, each fabric has its own and specific bending histeresis.
Figure 3. Basic concept of bending
A characteristic fabric bending histeresis can be seen in Figure 6 (Zhou and Ghosh, 1998). In bending a fabric, fibres are exposed to pressure strain and cannot slip in contact with each other without causing resistance to friction. As flexural strains are quite high, resistance to friction results in the actual fabric bending. After pressure strain has ceased, a histeresis of the fabric returning to its initial state (not bent) follows. Initially, flexural strain causes linear growth of the
Fabric mechanical properties 235
Figure 4. Bending effect
Figure 5. Bending circle, with radius of 1 cm
Figure 6. Characteristic woven fabric bending histereses
IJCST 15,3/4
236
moment-bending curve, while at higher loads a change of the linear part into a curve can be observed. Numerous investigations have shown that the bending properties are decisive in determining the ability of a fabric to be shaped to its handling, creasing behaviour, care and crease recovery (Peirce, 1930). Two key parameters that characterise fabric flexural behaviour are its flexural and folding rigidity and adequate histereses, which are, as a matter of fact, the measure of fabric relaxation. It can be rightly supposed for most materials that their bending behaviour is linear, while flexural rigidity is defined as a constant of the ratio between the moment applied DM and curvature DK according to the following expression (Peirce, 1930): B ¼ tg a ¼
DM DK
ð2Þ
3. Measuring and results Fabrics are exposed to various strains in garment manufacture and they react in various ways. Fabric behaviour can be more precisely predicted if we know their mechanical properties. This paper presents the investigations of fabric mechanical parameter properties, and their impact on fabric behaviour in garment manufacture. The behaviour of 300 fabrics analysed was recorded in the course of particular garment manufacturing processes. It was the basis for determining measuring values of the parameters of individual mechanical properties. The results obtained, dealing with the parameters of tensile, shear and flexural properties, are presented in the form of characteristic histeresis curves (Figures 7-9), while the results of measuring shear and flexural properties are presented in Tables I and II. Characteristic histeresis curves for particular groups of fabrics analysed are shown for different fabric raw material content: wool fabrics (a), blends of wool and man-made fibres (b), blends of wool and
Figure 7. Characteristic tensile histereses
Fabric mechanical properties 237 Figure 8. Characteristic shear histereses
Figure 9. Characteristic bending histereses
man-made fibres with elastane (c), cotton fabrics and blends of cotton with man-made fibres (d), acetate fabrics and blends of acetate with other man-made fibres (e), polyester fabrics and blends of polyester with other man-made fibres (f), polyamide fabrics and blends of polyamide, elastane and other man-made fibres (g). 4. Discussion The analysis of the results obtained indicates, for all the fabrics analysed regardless of the raw material content, that higher elongation results in fabric distortion. However, when the ability to relax is reduced with the reduction of tensile strain, there are still occasional deviations from the rule, caused by construction parameters of the fabrics analysed, as well as by various raw material contents in blends. Most easily noticeable deviations of tensile property parameters can be seen in the relaxation ability, especially with the fabrics of high yarn density and those made of boucle yarn. Higher deviations can also be noticed with fabrics in leno weave, as these fabrics remain partly distorted even after the source of strain has been eliminated.
IJCST 15,3/4
Measuring values
Meaning
G ¼ 0.6-0.9
Shear rigidity G
No potential problems in garment manufacture
G , 0.6
Low shear rigidity G
Distortions in laying: creasing, straight and oblique distortions Significant distortions of cutting parts in cutting (left – right parts) Considerable fabric instability in cutting Considerable seam puckering Problems with slippage, guiding aligning in positioning in sewing Poor appearance of the inset sleeve
0.9 , G , 1.5
High shear rigidity G
Poor sticking together of the sewn parts, creasing of one or both parts at the seam Poor shaping ability, e.g. inability to be shaped by ironing
G.2
Very high shear rigidity G Extremely poor sticking together of the sewn parts, creasing of one or both parts at the seam Poor shaping ability, e.g. inability to obtain fullness of shape
238
Table I. Measuring values of shear properties
Problems expected in garment manufacture
It is also true, for the direction of the warp and weft alike under a shearing strain, that the level of shear histeresis rises with higher shear rigidity. There are deviations of this rule, with the fabrics containing elastane fibres, especially regarding the direction of shear distortion. Fabric weave has a considerable influence on shearing properties, as well as the number of interlacing points of warp and weft yarns. More interlacing points mean higher value of shearing property parameters. Specific shapes of shear histeresis can also be noticed with the fabrics with polyurethane coating, as well as with hairy fabrics, where shear histereses are noticeably broader (Figure 8). Higher bending rigidity results in higher bending histeresis. Specific deviations can be observed with thicker fabrics, corduroy fabrics and also fabrics made of boucle yarns. Lower mass and high yarn density (cloth and reps weave) are the factors that raise the fabric bending rigidity. However, even the fabrics with higher mass and smaller number of interlacing points can exhibit higher values of bending property parameters. They can also be noticed with the fabrics containing a polyurethane coating and plastified fabrics, which are quite often used for fashion garments today. Based on the results of recording the behaviour of particular fabrics in garment manufacturing processes, the parameters of shear (G) and bending (B) rigidity are presented in Tables I and II, given in the form of border, e.g. critical values. Each specific area is shaped on the basis of analysing the relation of mechanical properties and fabric behaviour in garment manufacturing
Measuring values
Meaning
Problems expected in garment manufacture
B ¼ 0.04-0.1
Bending rigidity – B
No potential problems in garment manufacturing
B , 0.04
Low bending rigidity – B
Additional alignment in laying necessary at certain points Threads sticking from the parts cut Yarns drawn out from the fabric Seam puckering Seams produced are not smooth
0.1 , B , 0.2
High bending rigidity – B
Cutting layers unstable Fabric unstable in cutting Stitching holes appearing and stitches of uneven length Problems in sticking sewing parts together, creasing of one or both components at the seam Poor ability to stick to the contour in guiding curved lines Poor shaping ability, e.g. inability to be shaped by ironing Sleeve puckering
0.2 , B , 0.4
High bending rigidity – B
Considerable cutting layer instability Stitching holes appearing and stitches of uneven length Problems in sticking sewing parts together, creasing of one or both components at the seam Poor ability to stick to the contour in guiding curved lines Poor shaping ability, e.g. inability to be shaped by ironing
B . 0.4
High bending rigidity – B
Extreme cutting layer instability Stitching holes appearing and stitches of uneven length Considerable problems in sticking sewing parts together, creasing of one or both components at the seam Extremely poor ability to stick to the contour in guiding curved lines Poor shaping ability, e.g. inability to be shaped by ironing Poor sleeve matching
processes, recorded following the elements and criteria explained above. For both the parameters of mechanical properties, an area is determined where no specific problems are encountered in garment manufacture, as well as the areas of excessively high and low values, which, on the other hand, result in specific
Fabric mechanical properties 239
Table II. Measuring values of bending properties
IJCST 15,3/4
240
fabric behaviour. Critical areas for the parameters of elongation, relaxation ability, friction coefficient, geometric roughness and dimensional stability are determined in a similar manner (Zavec, 2001; Zavec and Gersˇak, 2001). 5. Conclusion Individual mechanical property parameters are unavoidable in describing a fabric and predicting its behaviour. Detailed knowledge of fabric mechanical properties offers both the information on its properties and potential behaviour, particularly when it is exposed to tensile, shear and flexural strains. Knowledge of the fabrics from this point of view is not only important for solving some problems in garment manufacturing processes, but also for making accurate engineering predictions in manufacturing high-quality articles of clothing. This is the only way to ensure the visual appearance of the garment that has been planned. References Bona, M. (1994), Textile Quality, Physical Methods of Product and Process Control, Textilia, Istituto per la Tradizione e la Technologia Tessile, Biella. Cooper, D.N.E. (1960), “The stifness of woven textiles”, Journal of the Textile Institute, Vol. 51 No. 5, pp. 317-35, ISSN 0400-5000. Kawabata, S. (1980), The Standardization and Analysis of Hand Evaluation, 2nd ed., Osaka, July 1980, The Hand Evaluation and Standardization Comittee, The Textile Machinery Society of Japan. Kilby, W.F. (1963), “Planar stress-strain relationship in woven fabrics”, Journal of the Textile Institute, Vol. 54, pp. 9-27, ISSN 0400-5000. Lindberg, J., Behre, B. and Dahlberg, B. (1961), “Shearing and buckling of various commercial fabrics, Part III”, Textile Research Journal, Vol. 31 No. 2, pp. 99-122, ISSN 0040-5175. Peirce, F.T. (1930), “The handle of cloth as a measureable quantity”, Journal of the Textile Institute, Vol. 21, pp. T377-T416, ISSN 0400-5000. Zavec, D. (2001), “Prediction of fabric behaviour in garment manufacturing processes”, Master thesis, Faculty of Mechanical Engineering, University of Maribor, Maribor March 2001 (in Slovene). Zavec, D. and Gersˇak, J. (2001), “Prediction of fabric behaviour as an input information for garment manufacturing process”, Tekstilec, Vol. 44 Nos 9-10, pp. 271-9, ISSN 0351-3386 (in Slovene). Zhou, N. and Ghosh, T.K. (1998), “Communication, on-line measurement of fabric-bending behavior: background, need and potential solutions”, International Journal of Clothing Science and Technology, Vol. 10 No. 2, pp. 143-56, ISSN 0955-6222.
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Determination of the sewing thread friction coefficient Darja Zˇunicˇ-Lojen and Jelka Gersˇak Department of Textiles, Faculty of Mechanical Engineering, University of Maribor, Smetanova, Slovenia
Sewing thread friction coefficient 241
Keywords Friction, Friction coefficient, Sewing Abstract The quality of sewing threads, as defined by their mechanical and physical properties, is connected with seam quality and seam strength. Seam quality and seam appearance depend mainly on the following sewing thread properties: bending properties, dimensional stability, thread twist and twist direction, fineness, regularity and surface properties on used fibres. The friction on the guide elements of the sewing machine depends on the surface treatment of the thread and affects the tensile force during the stitch formation process. It is important therefore, to know the friction coefficient when choosing the appropriate sewing thread. This contribution presents the influence of the sewing thread movement velocity over the guide element, the influence of the contact angle between the thread and guide element and the influence of the guide element material on the friction coefficient. The results of the research show that on increasing the thread velocity over the guide element, the friction coefficient slightly increases, whilst with the increase of the contact angle between the thread and guide element, the friction force exceedingly decreases. Furthermore, the results show that the friction coefficients using the steel guide are lower when compared with the ceramic guide. The behaviour of the sewing thread during the sewing process, can be concluded, based on these statements.
1. Introduction Friction also presents a very complex problem during a stitch formation process, because it is difficult to determine exactly which component or type of friction actually occurs. Therefore, the general characteristics and mathematical definitions of friction are presented, which can also be considered in sewing process. In friction, the following components can act (Helouvry, 1991): . static friction, which is the force necessary to initiate motion from rest; . kinetic friction, which is independent of the magnitude of the velocity; . viscous friction, which is a friction component proportional to velocity. Furthermore, other phenomena also appear: . break-away, this is the transition from static to kinetic friction; . break-away force is required to overcome the static friction; . break-away distance, this is the distance during break-away; . Dahl effect, this is the friction phenomenon that arises from the elastic deformation of bonding sites between two surfaces, which are locked in static friction;
International Journal of Clothing Science and Technology Vol. 15 No. 3/4, 2003 pp. 241-249 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310478341
IJCST 15,3/4
.
.
242
negative viscous friction, this is a phenomenon where friction decreases with increasing velocity; Stribeck effect, this is a case of negative viscous friction, which arises from the use of fluid lubrication. A characteristic of this effect is that at low velocity, friction decreases with increasing velocity.
Mechanism for the friction of threads is similar to that of other materials, which means the friction is dependent on the force needed to shear the junctions (Morton and Hearle, 1993). The friction coefficient varies in dependency for different experimental parameters such as load, speed, contact area, contact geometry, humidity and surface state. Therefore, only typical values for the friction coefficients obtained in a particular experiment can be presented. In general, the values of the friction coefficients for fibres are between 0.1 and 0.8, although they can also be much higher or lower (Morton and Hearle, 1993). 2. Fibre and sewing thread friction Generally, the coefficient of friction m is defined as the “dimensionless ratio of the friction force between two bodies to the normal force pressing these bodies together” (ASM International Handbook Committee, 1992):
m¼
F N
ð1Þ
where F is the friction force and N is the normal force. This depends on parameters like: material composition, surface finish, sliding velocity, temperature, lubrication, humidity, contamination, and oxide films. Investigations on fibres and threads showed that the ratio of frictional force to normal force decreases as the load increases. This means Coulomb’s law (1) is invalid for fibres and threads. Hence, it also follows that Eytelwein’s law (2) for calculating friction by sliding threads over a cylindrical guide, which is derived from Coulomb’s law for threads, does not exactly hold (Howell et al., 1959): F2 ¼ e ma F1
ð2Þ
where F1 is the incoming force, F2 is the leaving force, m is the friction coefficient, and a is the angle of contact, rad. The friction coefficient in Eytelwein’s equation depends on the ratio between the leaving and incoming forces and the angle of contact and should be of a constant value. From the experiments it can be seen that this ratio is not constant but depends on the leaving force (Howell, 1953). For explaining of this fact, different equations were used to fit experimental data. Some empirical equations, but without physical interpretation, were given. One of the
mathematical expressions, which successfully explains the experimental results, is: F ¼ aN n
ð3Þ
where F is the frictional force, N is the normal load on area A, and n, a are the constants. The value of parameter n generally lies between 0.67 and 1, depending on the material. For two perfectly elastic solids in contact n is 0.67, and for two surfaces having an area of contact determined by purely plastic conditions n ¼ 1: If n ¼ 1; equation (3) is equal to equation (1). The value of a is connected to n and it decreases almost linearly with n. In regard to equation (2), equation (3) gets the form (Morton and Hearle, 1993): F ð12nÞ ¼ F ð12nÞ þ ð1 2 nÞ aar ð12nÞ 2 1
ð4Þ
where r is the radius of the cylinder. Another expression given by Howell (1953) is: r ð12naÞ F2 a F 1 ¼e F1
ð5Þ
The relationship between the frictional force and normal load also depends on the system geometry: diameter of the crossed fibre and the diameter of the cylinder and contact area (Morton and Hearle, 1993). When the thread slides over the cylinder, the friction, first, depends on the sewing threads’ properties such as surface finish, stiffness, and thread lubrication, second, on guide properties like the smoothness and hardness of the surface, and finally on the thread velocity. The friction force can be reduced with appropriate lubrication as well as the difference between static and kinetic friction coefficients (Howell et al., 1959). Compared to the lubrication effect on metals, lubrication has a small effect on the friction of fibres and usually does not reduce the value of the friction coefficient to below 0.2. This behaviour is usually explained as boundary lubrication in which the layer of lubricant does not cover the asperities on the surface. Under these conditions the role of a good lubricant is to form monolayers on the surface at the contact points and to reduce the contact area between the materials. In this case, most of the friction force results from the force needed to shear the lubricant film itself (Morton and Hearle, 1993). When the lubricant is thermally unresistant, it results in an increase in the friction force. In the sewing process, heat occurs because of the friction between the sewing needle and the fabric, as well as friction between the sewing thread and the fabric (Gersˇak, 1988).
Sewing thread friction coefficient 243
IJCST 15,3/4
244
The twist in the yarn also has an influence on the reduction of the friction reducing the contact area (Morton and Hearle, 1993; Rudolf and Gersˇak, 1997). Reduction in the contact area can also be attained by reduction of the cylinder radius. Measurements on an apparatus, which considers Eytelwein’s equation (Wegener and Schuler, n.d.) showed that with an increasing velocity of thread running over the cylinder, the friction coefficient rises. The path of the curve depends on the cylinder diameter, thread material, thread twist, and incoming force. With a small cylinder diameter the curve is less steep as for cylinders with greater diameters. As the velocity influences the friction, it can also be concluded that viscose friction is present. Temperature and relative humidity are the other influential parameters, because with decreasing temperature and increasing relative humidity at similar absolute humidity, the friction coefficient increases. In the sewing process, the friction force reflects as the friction between thread and fabric, the thread on the guide elements and mechanisms of the sewing machine, and as the friction between needle thread and looper thread. The intensity of friction differs at different places and depends on the main shaft turn and related phases of the stitch formation process (Gersˇak, 1987). During the sewing process, the sewing thread slides over several guide elements between tension regulator plates and through the needle eye, which effects the friction force between the thread and guide surfaces. On some guides, the angle of contact is unchangeable, whilst on others the angle changes. During the phase when the bobbin catches the loop of the needle thread, the friction occurs between the thread and the bin of hook. When determining friction forces, authors investigating the forces acting on sewing thread moving over or through the guides on the sewing machine usually used Eytelwein’s equation. The force for the sliding thread over the n guide elements when leaving the last guide element is defined as (Fischer and Mollard, 1966): F n ¼ F 1 emða1 þa2 þ...þan Þ
ð6Þ
where ai is the angle of contact, rad, i ¼ 1; 2,. . ., n, and n is the number of guide elements. Equation (6) does not consider the dynamics of the process, in which the sewing thread on different guide elements does not move in the same direction during particular phases of the stitch formation process and so the friction forces act in different directions (Zˇunicˇ-Lojen, 2001). When considering the friction forces and inertia forces after the last guide, at the moment of stitch stretching, the tensile force FS is defined as (Gersˇak, 1991):
7 P F S ¼ Fe
i¼1
mi ai
FN mR ðe m4 a4 þ 1Þe þ 2
7 P
mi ai
i¼5
ð7Þ
þ F in
where F – force needed to draw off the thread from the bobbin, FN – pressure force, which acts on the thread over the tension regulator plates, mi – friction coefficient on ith guide element, mR – friction coefficient between the tension regulator plates, ai – angle of contact, formed between thread and ith guide element, in rad, m4 – friction coefficient on the tension regulator axle, a4 – angle of contact, where the thread embraces the tension regulator axle, in rad, and Fin – inertia force, which is acting in the area of the ith guide element.
Sewing thread friction coefficient 245
3. Methodology The investigation of the friction coefficient, and the incoming and outgoing forces of the thread sliding over the friction cylinder was carried on core PES thread with different linear densities (Table I). Analyses were carried out depending on the thread velocity over the cylinder, angle of contact, and the material of the friction cylinder. The friction coefficient, as well as the friction forces, were measured using a F-METER R-1188 instrument, with R-1084, and R-1073 thread winders and corresponding hardware and software. The friction coefficients were measured as thread-to-ceramic/steel cylinder friction and thread-to-thread friction. All measurements were carried out under standard testing conditions. 4. Results The results of the friction coefficient measurement thread/ceramic cylinder and measurement of the outgoing force F2 are given depending on the velocity
Thread designation G1 G2 G3
Nominal linear density Tt (tex)
Real linear density Tt (tex)
Real twist Tm (m2 1)
10 tex £ 2 12.5 tex £ 2 15 tex £ 2
21.72 27.77 36.32
999 1,033 914
Table I. Properties of the applied threads
IJCST 15,3/4
246
(Table II) and depending on the angle of contact (Table III). The dependency of the friction coefficient for the cylinder material for thread G1 is given in Figure 1. The results are shown at constant angle of contact and at constant velocity. The results of the thread-to-thread friction are shown in Figure 2. Table IV shows the tensile forces of the thread measured in the stitch formation process at the time when the take-up lever stretches the stitch in the fabric and the thread is pulled over the tension regulator on the sewing machine. The ratio are also given between the outgoing force F2 and the incoming force F1, and the calculated value ema of the increased force using Eytelwein’s equation and friction coefficient measured on the F-meter. At that moment, the thread moves from the left side of the take-up lever eye to the right side, considered by the definition of incoming and outgoing force.
5. Conclusions The analyses of the used threads’ friction coefficients show that with the increasing velocity of the sliding threads over the cylinder, the friction coefficients increases slightly on the steel as well as on the ceramic cylinders
G1 Velocity v (m min2 1) Table II. The friction coefficient m and outgoing force F2 for threads G1, G2, and G3 depending on the velocity of the thread
Table III. The friction coefficient m and outgoing force F2 for threads G1, G2, and G3 depending on the angle of contact of the thread
G2 F2(cN)
m
1 0.33 35.74 10 0.33 36.50 20 0.35 42.24 30 0.37 42.92 60 0.40 38.72 Note: Over the ceramic friction cylinder at , 12 cN)
m
F2(cN)
m
G2 F2(cN)
45 0.99 26.52 60 0.84 29.36 90 0.54 28.18 180 0.33 36.50 Note: Over the ceramic cylinder at a velocity
F2(cN)
m
0.26 35.14 0.39 41.22 0.28 39.48 0.40 42.08 0.31 45.44 0.41 43.88 0.34 46.92 0.41 44.42 0.34 47.74 0.43 46.56 the angle of contact a¼ 1808 (incoming force is
G1 Angle a (8)
G3
m
G3 F2(cN)
m
F2(cN)
0.99 26.22 0.88 24.68 0.81 27.62 0.78 27.38 0.53 28.44 0.53 27.94 0.28 39.48 0.40 42.08 of 10 m min2 1 (incoming force is , 12 cN)
Sewing thread friction coefficient 247
Figure 1. Comparison of the friction coefficient, and outgoing force F2 for thread G1 depending on the cylinder material. (a) Constant angle of the contact (a¼ 1808); (b) constant velocity (v ¼ 10 m min2 1); s – steel guide, c – ceramic guide
Figure 2. Results of the friction coefficient m and outgoing force F2 for threads G1, G2, and G3 depending on the velocity of sliding thread over thread
IJCST 15,3/4
248
Table IV. Measured tensile forces of the threads before and after the take-up lever on the sewing machine
for all analysed threads. With increasing velocity, the outgoing force at equal incoming force (,12 cN) also increases. This shows the dependency of the friction coefficient on the velocity. Furthermore analyses of the influence of the angle between the thread and cylinder show a large decrease in the friction coefficient with increasing angle (Table III). These results do not conform to the fact that with the increasing contact area, the friction increases. The values fall rapidly in the areas of smaller contact angles. At about an equal value of incoming force F1, the outgoing force F2 increases with any increase of the angle, however, the increase does not follow the increase of the contact angle. Comparison of the friction coefficients when sliding over the steel or over the ceramic cylinders shows that on a steel cylinder the friction coefficients are lower when investigating the velocity influence as well as the contact angle (Figure 1). When two thread slides over each other the results indicate that thread G2 has the highest value of friction coefficient and the thread G1, the lowest (Figure 2). The outgoing forces in this test also increases slightly with increasing velocity for all analysed threads. The highest forces were recorded for thread G2 and the lowest for thread G1 (Figure 2). On the basis of the comparison between the friction coefficients of thread sliding over ceramic, steel cylinder, and thread, it can be seen, that the friction coefficient is higher for thread sliding over the cylinder (around 12-16 per cent) than that for the thread. Comparison between the outgoing forces also shows higher values when sliding over the cylinder. It can be seen in Table IV that for the ratio between the outgoing and incoming forces, respectively, the increase of the outgoing force to the incoming force when the thread slides through the eye of the take-up lever is lower than the calculated values from Eytelwein’s equation (2), which concerns the measured friction coefficient on F-meter and the contact angle on the guide element on the sewing machine. We can conclude from these results that measured values of friction coefficient can often serve as a comparison between different threads and help to choose the suitable thread. They cannot, however, be directly used for calculation of the forces that act on the thread during the stitch formation process.
Thread
F1(N)
F2(N)
F2/F1
ms,60 m min2 1, 1808
ema
G1 G3
0.95 1.39
1.44 1.85
1.52 1.33
0.32 0.33
2.92 3.02
Note: At stitch velocity n ¼ 1; 000 rpm and calculated values ema, a ¼ 3:35 rad
References ASM International Handbook Committee (1992), Friction, Lubrication, and Wear Technology, ASM Handbook, Vol. 18, ISBN 0-87170-380-7. Fischer, V. and Mollard, C.E. (1966), Entwicklung von Verfahren zur Untersuchung der Stichbildung bei Industrie Schnellna¨hern, Dissertation, Fakulta¨t fu¨r Maschinenwesen der Technischen Hochschule Carolo-Wilhelmina zu Braunschweig. Gersˇak, J. (1987), “Analiza obremenitve sukanca med procesom oblikovanja vboda”, Tekstil, Vol. 36 No. 9, pp. 481-9, ISSN 0492-5882. Gersˇak, J. (1988), “Untersuchungen u¨ber die Gro¨ße der Belastung des Na¨hfadens wa¨hrend des Stichbildungsprozesses”, Bekleidung und Wa¨sche, Vol. 40 No. 16, pp. 37-41, ISSN 0005-8270. Gersˇak, J. (1991), “Dinamicˇko naprezanje konca kao posljedica tehnolosˇki uvjetovanih sila u procesu oblikovanja uboda”, Tekstil, Vol. 40 No. 5, pp. 213-22, ISSN 0492-5882. Helouvry, A.V. (1991), Control of Machines with Friction, ISBN 0-7923-9133-0, Kluwer Academic Publishers, Boston. Howell, H.G. (1953), “The general case of friction of a string round a cylinder”, Journal of the Textile Institute, Vol. 44, p. T359. Howell, H.G., Mieszkis, K.W. and Tabor, D. (1959), Friction in Textiles, Butterworths Scientific Publications, London. Morton, W.E. and Hearle, J.W.S. (1993), Physical Properties of Textile Fibres, ISBN 1 870812 41 7, The Textile Institute, Manchester. Rudolf, A. and Gersˇak, J. (1997), “Influence of the twist on elastic properties of a thread”, in Gersˇak, J., Zˇunicˇ-Lojen, D. and Stjepanovicˇ, Z. (Eds), 2nd International Conference IMCEP ’97, pp. 207-212, ISBN 86-435-0202-2, Maribor, Slovenia, October 1997, Faculty of Mechanical Engineering, Institute of Textile and Garment Manufacture Processes, Maribor. Wegener, W. and Schuler, B. (n.d.), Beitrag zur Grundlagenermittlung des Reibungskoeffizienten von Fa¨den, Sonderdruck aus Zeitschrift fu¨r die gesamte Textilindustrie, p. 18. ˇZunicˇ-Lojen, D. (2001), Interakcija mehanizmov sˇivalnega stroja in sukanca v procesu oblikovanja vboda, Doktorska disertacija, Univerza v Mariboru, Fakulteta za strojnisˇtvo, Maribor.
Sewing thread friction coefficient 249
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
IJCST 15,3/4
250
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Testing electrostatic properties of polyurethane coated textiles used for protective clothing Nada Hainsˇ, Vera Frisˇcˇic´ and Dubravka Gordosˇ Faculty of Textile Technology Studies, University of Zagreb, Varazˇdin, Croatia Keywords Protective clothing, Testing, Textiles Abstract The objective of the study was to test the electrostatic properties of textiles used for protective clothing worn in flammable and potentially explosive environment. The protective clothing is intended for multi-use and, apart from wear and stretching during use, it is also exposed to impacts of various cleaning agents. The testing was carried out in accordance with EN 1149-1; 1995 specifying requirements for electrostatic properties and testing methods for protective clothing to be worn in specific situations. The objective was to test whether there was a likelihood of fire due to the electric discharge. EN 1149-1 requires measuring surface resistance at 25 ^ 2 per cent relative humidity and temperature of 23 ^ 18C. Measurements were taken on knitting of varying materials coated with polyurethane. The knittings used were made of polyamide, polyester and cotton. All materials were coated with the same polyurethane coating. All samples were tested for surface resistance before and after five dry-cleanings and machine washings. In addition to the required European standard, surface resistance was also measured at the following relative humidity: 35, 45, 55 and 65 per cent. The results were processed by the correlation analysis and shown graphically.
International Journal of Clothing Science and Technology Vol. 15 No. 3/4, 2003 pp. 250-257 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310478350
1. Introduction Textile materials, under certain conditions, get charged with static electricity. This can be caused by friction with other textiles or solid bodies, by touching or detaching from such surfaces, because of air currents in closed spaces, or UV rays. Except causing interferences in production and finishing of textiles, the tendency of getting charged with static electricity has an impact on comfort of wearing clothes, and in certain areas can even be a very dangerous occurrence. This especially concerns protective clothing used in explosive surroundings, e.g. gas stations or chemical industry, where there is a danger of explosion resulting from a spark. Danger of explosion can be determined by measuring surface and transitory resistance. With protective clothing the surface resistivity has to be less than 5 £ 1010 V (EN 1149-1, 1995). If the measurement results are below that value, explosion can be reached only in very isolated cases (Schmeer - Lioe, 2000). There is a correlation between charging and unloading with static electricity and electrical resistance. The active charged time caused by friction is basically proportional to the surface resistivity (Schwager and Nestler, 1984).
The primary switching of charged state especially depends on the materials in contact with one another, on their position in electrostatic sequence, as well as on geometrical and physical state of surfaces (pressure, friction intensity, fuzziness) (Schwager and Nestler, 1984). 2. Methods DIN 53435/T-1, DIN 54345-5, AATCC 76, AATCC 84 and EN 1149-1 define determination of electrical resistance of textiles (Cˇunko, 1988; DIN 54345 Teil 1, 1985; DIN 54345 Teil 5, 1985). Electrical resistance is much dependent on the humidity of textiles. To encompass even the toughest conditions, so that materials could be better differentiated, relative humidity range of 24-40 per cent is prescribed. DIN 53435/T-1 regulates determination of electrical resistance of threads, yarn and flat textiles. Surface and conditioned resistance is measured in V using preconditioned samples under relative air humidity of 25 per cent and the temperature of 238C. Measures are taken with a circling electrode. DIN 54345-5 defines determination of electrical resistance of conductive flat products; standard threads or/and consisting of special conductive threads. Measurement is carried out under relative air humidity of 25 per cent and temperature of 238C on work clothes and some technical fabrics. The AATCC 76 method pinpoints the procedure of determining electrical resistance of textiles for flat products. Measurement is carried out using two electrodes which are placed on the surface of the flat product textile. Electrode dimensions are not specified like they are in DIN. Measurement is carried out under relative air humidity of 40 per cent and temperature of 248C. AATCC TM 84 regulates the determination of electrical resistance using electric current which pass through a strand of yarn or parallel small bushels of such strands. Relative air humidity is 40 per cent and the temperature is 248C. EN 1149-1 is a built-in component in a series of methods for testing protective clothing. It is not applicable for use on textiles with conductive threads because it has been determined that measuring surface resistance on such materials gives physically illogical results. Therefore, methods for the third part of EN 1149 are being prepared, which will be able to test on such textiles as well. For coated materials, such as textiles coated with polyurethanes, the measured surface resistivity on at least one side must be under 5 £ 1010 V: The resistance of the electric current passing between the two electrodes is measured. Measurement of surface resistance is carried out with an Ohmmeter on a sample placed on an isolated background which rests on a grounded conductive surface, e.g. metal plate, using a circling electrode consisting of a cylindrical and a ring electrode concentrically placed one around the another. The norm prescribes two types of electrodes. The type A is a non-corrosive steel electrode, and type B a solid brass electrode. Electrode of type A was used for this work. Five cycles of cleaning and washing should be done on materials
Testing electrostatic properties 251
IJCST 15,3/4
which are not for one use only. The prescribed relative air humidity is 25 per cent and the temperature is 23 ^ 18C: Surface resistivity is calculated using the following equation: p ¼ kR
252
ð1Þ
where p is the calculated surface resistivity (V), R is the calculated resistance (V), and k is the geometrical factor, which is 19.8 for electrode type A and 5.7 for electrode type B. Measurements can be carried out on different parts of the clothing item without ruining the item, which is especially important. Oscillations of charge thickness on surfaces charged with static electricity cannot be determined with measuring resistance. 3. Testing Measurement of surface resistance was carried out pursuant to EN 1149-1 using an Ohmmeter of type DMB-6 made by MAHLO with a type A electrode under air humidity of 25, 35, 45, 55 and 65 per cent and temperatures of 23 ^ 18C: Preliminary measurements determined that the optimal conditioning time ranges from 72 to 96 h. Relative humidity of 24, 35 and 45 per cent are reached using sulphuric acids of convenient concentrations. Tests were done on knitting samples made up of various raw materials and coated with polyurethanes meant for protective clothing. An antistatic was added in same percentage to this polyurethane coating on all samples. For the sake of comparison, a sample of PA 6.6 knitting coated with polyurethane, without antistatic was tested as well. Samples were also tested after five chemical cleanings and five machine washes. Machine washing was done according to DIN 53920 - 1978, and the chemical cleaning using perchlorethylene according to HRN F: S3. 226 - 1983. Testing was done on the front (polyurethane coating) and back of the samples. The characteristics of the knitting are shown in Table I. Five samples were tested: . Sample 1. Front PU, back PA 6.6, white without antistatic; . Sample 2. Front PU, back PA 6.6, white with antistatic; Density Raw material
Table I. The characteristics of the knitting
PA 6.6 white PA 6.6 blue PA 6 Cotton
Strength (daN)
Stretching (per cent)
No. of No. of Surface mass rows (cm) stitches (cm) Length Breadth Length Breadth (g/m2) 103.3 106.0 96.8 139
12.5 12 12 15
17 17 15 19
29.2 31.0 29.7 18.7
19.9 17.2 19.0 12.0
98.5 102.0 107.0 74.8
174.0 178.4 225.0 148.7
. . .
Testing electrostatic properties
Sample 3. Front PU, back PA 6.6, blue with antistatic; Sample 4. Front PU, back PA 6 black with antistatic; Sample 5. Front PU, back cotton with antistatic.
Samples 2-5 were tested before and after five chemical cleanings and machine washes. Chemically cleaned samples were additionally marked with P and those machine washed with W. The results of testing under the relative air humidity of 25 per cent and the temperature of 23 ^ 18C are shown in Tables II-IV. Averages from five measures are shown. Surface resistivity measured on the front and the back of sample 1 without antistatic is 2:61 £ 1012 V: Samples 2-4 with antistatic have a surface resistivity measured on the front (polyurethane coating) less than 5 £ 1010 V: Sample 5 has a surface resistivity of 5:44 £ 1010 V; so a little more than 5 £ 1010 V: It is evident that blue PA 6.6 has a greater surface resistivity than white PA 6.6. After chemical cleaning the surface resistivity measured on each samples front is less than 5 £ 1010 V; except sample 5, which is 6:8 £ 1010 V: After washing, the surface resistivity, on all samples, front and back, is greater than 5 £ 1010 V: Therefore for products care chemical cleaning using perchlorethylene and disallo washing care are recommended. The results of testing on samples 1-5 are statistically analysed using correlation and regressive analysis (Figures 1-5). The correlation between the relative humidity and the surface resistivity, measured on the fronts of samples is very good (correlation coefficients greater
Sample Surface resistivity £ 1010 V
Sample Surface resistivity £ 1010 V
Sample Surface resistivity £ 1010 V
1 2 3 4 5 Front Back Front Back Front Back Front Back Front Back 261
261
2.06
2P Front Back 3.47
256
2W Front Back 194
273
44.6
3.09
3P Front Back 1.01
517
3W Front Back 329
530
510
2.90
379
4P Front Back 3.04
337
4W Front Back 262
575
5.44
54.2
5P Front Back 6.80
21.5
5W Front Back 347
416
253
Table II. Results of measuring surface resistivity of samples 1-5
Table III. Results of measuring surface resistivity of samples 2-5 after chemical cleaning
Table IV. Results of measuring surface resistivity of samples 2-5 after machine washing
IJCST 15,3/4
254
Figure 1. Dependence of surface resistivity on relative air humidity of sample 1. (a) Front; (b) back
Figure 2. Dependence of surface resistivity on relative air humidity of sample 2. (a) Front; (b) back
Testing electrostatic properties 255
Figure 3. Dependence of surface resistivity on relative air humidity of sample 3. (a) Front; (b) back
Figure 4. Dependence of surface resistivity on relative air humidity of sample 4. (a) Front; (b) back
IJCST 15,3/4
256
Figure 5. Dependence of surface resistivity on relative air humidity of sample 5. (a) Front; (b) back
than 0.9, except for sample 5). The correlation between the relative humidity and the surface resistivity measured on the backs of the samples is not good on blue PA 6.6 and cotton knitting. 4. Conclusion After adding antistatic, samples 2-4 are satisfactory according to EN 1149-1. Sample 5 has a little greater surface resistivity than the mentioned European norm prescribes. Based on the good correlation between the surface resistivity, measured on front of samples and the relative humidity we could conclude that testing resistance should not always be done under prescribed conditions of low relative humidity, which is in most laboratories harder to achieve than 55-65 per cent humidity range. From the individual measuring diagram for each product it is possible to consider the surface resistivity for the prescribed relative humidity, which is especially important for the quality control of products.
References Cˇunko, R. (1988), “Staticˇki elektricitet u tekstilu - nastajanje, mjerenje i uklanjanje”, Tekstil, Vol. 37 No. 12, p. 703. DIN 54345 Teil 1 (1985), Pruefung von Textilien; elektrostatisches Verhalten; Bestimmung elektrischer Widerstandgroessen. DIN 54345 Teil 5 (1985), Pruefung von Textilien; elektrostatisches Verhalten; Bestimmung des dielektrischen Widerstandes an Streifen aus textilen Flaechengebilden. EN 1149-1 (1995), Protective clothing-electrostatic properties, Part 1: surface resistivity test methods and requirements. Schmeer - Lioe, G., (2000), Anwendungsorientierte Prufung des elektrostatischen Verhaltens von Textilien, August 2000, Vol. 43, p. 182. Schwager, I. and Nestler R. (1984), Einfluss der elektrostatischen Aufladung beim Lagenlegeund Zuschnittprozess in der Konfektion, Textiltechnik Vol. 34 No. 10, p. 540.
Testing electrostatic properties 257
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
IJCST 15,3/4
258
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Surface roughness of heat protective clothing textiles Jirˇ´ı Militky´ and Vladimı´r Bajzı´k Department of Textile Materials, Textile Faculty, Technical University of Liberec, Liberec, Czech Republic Keywords Surface roughness, Measurement, Fractals Abstract The surface roughness is one of the main parts of hand prediction. Classical method of surface roughness measurements is based on the surface profile measurement. Characteristic of roughness is then variation coefficient of surface profile (surface height variation). The main aim of this work is to estimate the surface profile complexity by using variogram (structure function). The surface profile variation is classified to the group according to short- and long-range dependence. The concept of fractal dimension is proposed especially for long-term correlation cases. The applicability of the proposed approach is demonstrated on the typical heat protective clothing fabrics and compared with the results of surface roughness evaluated by the KES system.
International Journal of Clothing Science and Technology Vol. 15 No. 3/4, 2003 pp. 258-267 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310478369
1. Introduction Roughness of engineering surfaces has been traditionally measured by the stylus profiling method creation of surface profile called surface height variation (SHV) trace (Vandenberg and Osborne, 1992). This profile characterizes thickness (height) variation in selected direction. Modern methods are based on the image processing of surface images (Zhang and Gopalakrishnan, 1996). Surface irregularity of plain textiles has been identified by friction (Ajayi, 1992), contact blade (Ajayi, 1994; Kawabata, 1980), lateral air flow (Ajayi, 1988), step thickness meter (Militky´ and Bajzı´k, 2000) or subjective assessment (Stockbridge et al., 1957). Standard methods of surface profile evaluation are based on the relative variability characterized by the variation coefficient (analogy with evaluation of yarn’s mass unevenness) (Meloun et al., 1992) or simply by the standard deviation. This approach is used in Shirley software for evaluation of results for step thickness meter (Operation Manual, 1999). Characterization of roughness based on the mean absolute deviation (MAD) is the classical descriptive statistical approach. This statistical characteristic is useful for random SHV traces, where elements of SHV trace are statistically independent of each other. The SHV profile of a lot of fabrics has been identified as irregular and more structured. The descriptive statistical approach based on the assumptions of independence and normality leads to biased estimators if the SHV has short- or long-range correlations (Meloun et al., 1992). Therefore, it is necessary to distinguish between standard white Gauss noise and more complex models. For description of short range correlations, the models based on the autoregressive This work was supported by project LN 00 B090 of Czech Ministry of Education.
moving average are useful (Quinn and Hannan, 2001). The long-range correlations are characterized by the fractal models (Constantine and Hall, 1994; Mandelbrot and Van Ness, 1968). The deterministic chaos type models are useful for revealing chaotic dynamic in deterministic processes, where variation appears to be random, but in fact they are predictable (Ott et al., 1994). For the selection among the above-mentioned models, the power spectral density (PSD) curve evaluated from experimental SHV can be applied ( Eke et al., 2000). Especially, the fractal models are widely used for rough surface description (Whitehouse, 2001). For these models the dependence of log (PSD) on the log (frequency) should be linear. Slope of this plot is proportional to fractal dimension and intercept to the so-called topothesy. For, white noise has dependence of log(PSD) on the log(frequency) nearly horizontal plateau for all frequencies (the ordinates of PSD are independent and exponentially distributed with common variance (Ott et al., 1994)). More complicated rough surfaces as a result of grinding can be modelled by the Markov type processes (Sacerdotti et al., 2000). For these models the dependence of log(PSD) on the log(frequency) has plateau at small frequencies than bent down and are nearly linear at high frequencies. The fractal type models were criticized by Whitehouse (2001), who concluded that the benefits are more virtual than real. On the other hand, the deeper analysis of rough surface should use a more complex model than the classical descriptive statistics. Greenwood (1984) proposed a technique based on the definition of local maxims (peaks) and derivation of peaks height distribution. A lot of recent works is based on the assumption that the stochastic process (Brownian motion) can describe thickness variation (Nayak, 1971). This work is devoted to the analysis of load required to move the blade on fabric surface R(d ) obtained from new accessory to tensile testing machine. 2. Surface profile evaluation Kawabata (1980) constructed a measuring device for registering the SHV trace. The main part of this device is the contactor in the form of wire (diameter 0.5 mm). This contactor is moved at a constant rate 0.1 cm /s and SHV is registered on the paper sheet. The sample length, L ¼ 2 cm is used. Characterization of roughness is based on the MAD (the classical descriptive statistical approach). Similar approach is based on the measurement of R(d ) by Shirley step meter with replacement of measuring head by blade (Militky´ and Bajzı´k, 2001). We have constructed the simple accessory to the tensile testing machine. The principle is registration of the force F(d ) needed for tracking the blade on the textile surface. Roughly speaking, the F(d ) should be inversely proportional to the R(d ). In reality, the F(d ) profile is different due to small surface; deformation caused by the tracked blade. Output from measurements is sequence of loads F(di). Variation of thickness R(di) or loads F(di) can be
Surface roughness
259
IJCST 15,3/4
260
generally assumed as combination of random fluctuations (uneven threads, spacing between yarns, non-uniformity of production etc.) and periodic fluctuations caused by the repeated patterns (twill, cord, rib etc.) created by weft and warp yarns. For the description of roughness the characteristics computed from R(d ) or F(d ) in places 0 , d , T (T is maximum investigated sample length and M is number of places) are used. Especially, for weaves it is necessary to identify periodic component in R(d ) or F(d ) as well. For this purpose, the spectral analysis can be useful. 3. Surface roughness description From the SHV or SFV trace it is possible to evaluate a lot of roughness parameters. Let us define roughness characteristics for SHV (the same equations are also valid for SFV). Classical roughness parameters are based on the set of points R(dj), j ¼ 1 . . . M defined in the sample length interval L. The measurement points dj are obviously selected as equidistant and then R(dj) can be replaced by the variable Rj. For the identification of positions in length scale, it is sufficient to know sampling distance d s ¼ d j 2 dj21 ¼ L=M for j . 1: The standard roughness parameters used frequently in practice are (Wu, 2000): MAD. This parameter is equal to the mean absolute difference of surface heights from average value (Ra). For a surface profile, this is given by: 1 X ð1Þ jRj 2 Ra j MAD ¼ M j This parameter is often useful for quality control. However, it does not distinguish between profiles of different shapes. Its properties are known for the case when Rj’s are independent identically distributed (iid) random variables. Standard deviation (root mean square) value (SD). This is given by: sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1 X SD ¼ ð2Þ ðRj 2 Ra Þ2 M j Its properties are known for the case when Rj’s are iid random variables. One advantage of SD over MAD is that for normally distributed data is simplicity of computation of confidence interval and realization of statistical tests. SD is always higher than MAD and for normal data SD ¼ 1:25 MAD: It does not distinguish between profiles of different shapes as well. The parameter SD is less suitable than MAD for monitoring certain surfaces having large deviations (corresponding distribution has heavy tail). Mean height of peaks (MP). This is calculated as the average of the profile deviations above the reference value R (often R ¼ Ra ). It is given as mean value of peaks Pi, i ¼ Np where: P i ¼ Ri 2 R for Ri 2 R . 0 and P i ¼ 0 elsewhere
Mean height of valleys (MV). This is calculated as the average of the profile deviations below the reference value R (often R ¼ Ra ). It is given as mean value of valleys Vi, i ¼ N v where:
Surface roughness
V i ¼ R 2 Ri for Ri 2 R , 0 and V i ¼ 0 elsewhere The parameters MP and MV give information on the profile complexity. Exceptional peaks or valleys are not considered, but are useful in tribological applications. The SD of profile slope (PS). This is given by: vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi u u 1 X dRðxÞ 2 t PS ¼ ð3Þ M j dx j The SD of profile curvature (PC). This quantity often called as waviness is defined by the similar way: vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi u u 1 X d2 RðxÞ 2 t PC ¼ ð4Þ M j dx 2 j The slope and curvature are characteristics of a profile shape. The PS parameter is useful in tribological applications. The lower the slope the smaller the friction and wear. Also, the reflectance property of a surface increases in the case of small PS or PD. Mean slope of the profile (MS). This is given by: 1 X dRðxÞ MS ¼ ð5Þ M j dx j Mean slope is an important parameter in several applications such as in the estimation of sliding friction and in the study of the reflectance of light from surfaces. Ten point average (TP). This characteristic is defined as the average difference between the five highest peaks and five deepest valleys within a surface profile. The parameter TP is sensitive to the presence of high peaks or deep scratches in the surface and is preferred for quality control purposes. These parameters are useful in the case of functional surfaces or for characterizing surface bearing and fluid retention and other relevant properties. For, the characterization of hand will probably be the best to use waviness PC. The characteristics of slope and curvature can be computed for the case of fractal surfaces from power spectral density, autocorrelation function or variogram. A set of parameters for profile and surface characterization are collected in (Nayak, 1971).
261
IJCST 15,3/4
262
There exist a vast number of empirical profile or surface roughness characteristics suitable often in very special situations. Some of them are closely connected with characteristics computed from fractal models (fractal dimension and topothesy). Greenwood (1984) proposed a general theory for description of surface roughness based on the distribution of heights. The most common way to separate roughness and waviness is spectral analysis. This analysis is based on the Fourier transformation from space domain d to the frequency domain v ¼ 2p=d: For computation of the above-mentioned characteristics, the program DRSNOST in MATLAB has been created. The following characteristics are computed: (1) Mean absolute deviation MAD; (2) Mean profile slope MS; (3) Standard deviation of profile slope PS; (4) Standard deviation of profile curvature PC; (5) Ten point average TP; (6) Variation coefficient CV ¼ SD=Ra ; (7) Mean fractal dimension DF; (8) Initial fractal dimension DFp. The computation of fractal dimensions is described in chapter 7. 4. Statistical analysis A basic statistical feature of R(d ) is autocorrelation between distances. Autocorrelation depends on the lag h (i.e. selected distances between places of thickness evaluation). The main characteristics of autocorrelation is covariance function C(h) CðhÞ ¼ covðRðdÞ; Rðd þ hÞÞ ¼ EððRðdÞ 2 EðRðdÞÞÞ ðRðd þ hÞ 2 EðRðdÞÞÞÞ ð6Þ and autocorrelation function ACF(h) defined as normalized version of C(h): ACFðhÞ ¼
CðhÞ Cð0Þ
ð7Þ
The E(x) denotes expected value of x. ACF is one of the main characteristics for the detection of short- and long-range dependencies in dynamic (time) series. It could be used for the preliminary inspection of data. The computation of sample autocorrelation directly from definition for large data is tedious. The technique of ACF creation based on the FFT is contained in the signal processing toolbox of MATLAB ( procedure xcorr.m) [18] (Bloomfield, 2000). In spatial statistics, more frequent variogram (called often as structure function) is defined as one half variance of differences ðRðdÞ 2 Rðd þ hÞÞ
GðhÞ ¼ 0:5 D½RðdÞ 2 Rðd þ hÞ
ð8Þ
Symbol D(x) denotes variance of x. For stationary random process mean value is independent on lag h i.e. EðRðhÞÞ ¼ m and then GðhÞ ¼ 0:5 EðRðdÞ 2 Rðd þ hÞÞ2
ð9Þ
The variogram is relatively simpler to calculate and assumes a weaker model of statistical stationarity, than the power spectrum. Several estimators have been suggested for the variogram. The traditional estimator is GðhÞ ¼
ðhÞ X 1 M ðRðd j Þ 2 Rðd jþh Þ2 2M ðhÞ j¼1
ð10Þ
where M(h) is the number of pairs of observations separated by lag h. Problems of bias in this estimate when the stationarity hypothesis becomes locally invalid have led to the proposal of more robust estimators. 5. Fractal dimension Benoit Mandelbrot has coined the term fractal in the 1970s (Mandelbrot and Van Ness, 1968). Fractals have two interesting characteristics. First of all, fractals are self-similar on multiple scales, in that a small portion of a fractal will often look similar to the whole object. Second, fractals have a fractional dimension, as opposite to integer dimension of regular geometrical objects. The fractional (fractal) dimension D can be evaluated by the following way: The number N(d) of line segments of length d needed to cover the whole curve in plane is measured. The length of curve is estimated as LðdÞ ¼ NðdÞd: In the limit d ! 0 the estimator L(d) becomes asymptotically equal to the length of the curve, L, independently on d. The Hausdorf-Besicovitch dimension D (fractal dimension) of this curve is the critical dimension for which the measure M d ðdÞ defined as: M d ðdÞ ¼ N ðdÞd d
ð11Þ
changes from zero to infinity (Feder, 1988). The value of M d ðdÞ for d ¼ D is often finite and therefore for sufficiently small d: NðdÞ < d D or LðdÞ < d 12D
ð12Þ
The fractal dimension is then computed as: D ¼12
log LðdÞ log d
Surface roughness
ð13Þ
For, random fractal is simpler to use power spectral density or related functions. Some techniques for fractal dimension computations are
263
IJCST 15,3/4
264
summarized, e.g. in Mannelqvist and Groth (2001). The methods for computation or Hurst coefficient is described in Wu (1999). In measurement of the surface profile (thickness variation R(h)), the data are available through one-dimensional line transect surface. Such data represent curve in plane. The fractal dimension DF is then number between 1 (for smooth curve) and 2 (for rough curve). Fractals are characterized by power type dependence of variogram and power spectral density. For a power law variogram: H
GðhÞ < cjhj
ð14Þ
where c is a constant. The Hurst exponent H, lies in the interval (0, 1). Where H ¼ 0 this denotes a curve of extreme irregularity and H ¼ 1 denotes a smooth curve. Exponents H and fractal dimension D are in fact related: DF ¼ 2 2 H
ð15Þ
Fractal dimension is conventionally obtained through estimating the parameter from a LSE linear regression of the log-log transformation of equation (14). In practice, its behavior is expressed by equation (14) valid near origin. In general, DF computed from this relation is denoted as an effective fractal dimension. Based on these equations the program DRSNOST in MATLAB for estimation of fractal dimension from variogram has been constructed. From the first 12 points (excluding three points near origin) the initial fractal dimension DFp and from all points the mean fractal dimension DF are computed. 6. Experimental part The 54 flame retardant barrier textiles have been selected for investigation. They covered flame retardant finished cotton fabrics (satin, linen and twill patterns), fabrics created from heat resistant fibers (Nomex, FR Viscose and modacrylic fibers) and combinations of heat resistant fibers with flame retardant finished cotton. The F(d ) traces have been obtained by means of the above described accessory. The R(d ) traces have been obtained from KES device, and Kawabata mean roughness (MAD) was computed. The subjective hand SH was evaluated from judgment of 30 persons. They rated the fabrics to the 11-point scale. The subjective hand SH was computed as median of ratings divided by 11. 7. Results and discussion For the investigation of mutual relations among subjective hand, classical characteristics of roughness (outputs 1-6 from DRSNOST program) and fractal characteristics of roughness (outputs 7-8 from DRSNOST program) the correlation map has been created. This map is shown in Figure 1(a). In the first
Surface roughness
265
Figure 1. (a) Correlation map of characteristics (first variable is SH); (b) relation between initial fractal dimension and subjective hand SH; (c) dependence between roughness from SFV (DFp ) and Kawabata SHV (MAD)
IJCST 15,3/4
266
column of this map are correlations of subjective hand with roughness characteristics. It is clear, that the correlations are not so high (black denotes no correlation and white denotes perfect linear relation). Maximum correlation is between subjective hand and fractal dimensions. There are correlations between some roughness characteristics as well. The dependence between subjective hand and initial fractal dimension DFp is shown in the Figure 1(b). It can be said that for these materials the roughness has a little influence on hand. The deeper analysis of the correlation map and partial relations between roughness characteristics lead to the following conclusions: MAD highly correlates with other roughness characteristics; MAD correlates with fractal dimensions as well, but some no linearity appears. Comparison of DFp calculated from SFV and Kawabata MAD from SHV is shown in Figure 1(c). Moderate correlation in Figure 1(c) indicates the differences between these two methods. One reason is the filtration of some frequencies realized automatically by the KES device. 8. Conclusion The initial fractal dimension is probably most suitable for the complexity of roughness characterization. The analysis of SFV based on the DRSNOST program is more complex in reality. The more classical roughness characteristics and topothesy are computed as well and many other techniques of fractal dimension calculation are included. References Ajayi, J.O. (1988), “Some studies of frictional properties of fabrics”, Doctoral thesis, University of Strathclyde, Glasgow. Ajayi, J.O. (1992), “Fabric smoothness, friction and handle”, Textile Research Journal, Vol. 62, pp. 87-93. Ajayi, J.O. (1994), “An attachment to the constant rate of elongation tester for estimating surface irregularity of fabric”, Text. Res. J., Vol. 64, pp. 475-6. Bloomfield, P. (2000), Fourier Analysis of Time Series, Wiley, NY. Constantine, A.G. and Hall, P. (1994), J. Roy. Stat. Soc., Vol. B56, p. 97. Eke, A. et al., (2000), Eur. J. Physiol., Vol. 439, p. 403. Feder, J. (1988), Fractals, Plenum Press, NY. Greenwood, J.A. (1984), “A unified theory of surface roughness”, Proc. Roy. Soc. London, Vol. A393, p. 133. Kawabata, S. (1980), “The standardization and analysis of hand evaluation”, Text. Mach. Soc. Japan. Mandelbrot, B.B. and Van Ness, J.W. (1968), “Fractional Brownian motion, fractional noises and applications”, SIAM Review, Vol. 10, p. 442. Mannelqvist, A. and Groth, M.R. (2001), Appl. Phys., Vol. A73, pp. 347-56. Meloun, M., Militky, J. and Forina, M. (1992), “Chemometrics for analytic chemistry Vol. I”, Statistical Data Analysis, Ellis Horwood, Chichester.
Militky´, J. and Bajzı´k, V. (2000), “Description of thickness variation by fractal dimension”, Proc. Conf. STRUTEX 2000, December Liberec (2000) (in Czech). Militky´, J. and Bajzı´k, V. (2001), “Characterization of textiles surface roughness”, Proc. 7th Int. Asian Textile Conference, August 2001, Hong Kong. Nayak, P.R. (1971), Trans. ASME: J. Lub. Tech., Vol. 93F, p. 398. Operation Manual (1999), Shirley. Ott, E., Sauer, T. and Yorke, J.A. (Eds) (1994), Copying with chaos, Wiley, NY, USA. Quinn, B.G. and Hannan, E.J. (2001), The Estimation and Tracking of Frequency, Cambridge University Press, Cambridge. Sacerdotti, F., Griffiths, B.J., Butler, C. and Benati, F. (2000), Proc. Inst. Mech. Engfs., Vol. 214B, p. 811. Stockbridge, H.C. et al., (1957), “The subjective assessment of the roughness of fabrics”, Journal of the Textile Institute, Vol. 48, pp. T26-34. Vandenberg, S. and Osborne, C.F. (1992), Wear, Vol. 159, pp. 17-30. Whitehouse, D.J. (2001), Wear, Vol. 249, p. 345. Wu, J. (1999), Wear, Vol. 230, p. 194. Wu, J. (2000), Wear, Vol. 239, p. 36. Zhang, C. and Gopalakrishnan, S. (1996), “Fractal geometry applied to on line monitoring of surface finish”, Int. J. Mach. Tools Manufact., Vol. 36, pp. 1137-50.
Surface roughness
267
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
IJCST 15,3/4
268
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Influence of the sterilisation process on the tactile feeling of surgical gowns Flora Philippe Laboratoire de Physique et Me´canique Textile, Ecole Nationale Supe´rieure des Industries Textiles de Mulhouse, Mulhouse Cedex, France
Maria Jose´ Abreu Departamento de Engenharia Textil, Universidade do Minho, Guimara˜es, Portugal
Laurence Schacher and Dominique C. Adolphe Laboratoire de Physique et Me´canique Textile, Ecole Nationale ´ Superieure des Industries Textiles de Mulhouse, Mulhouse Cedex, France
Maria Elisabete Cabec¸o Silva Departamento de Engenharia Textil, Universidade do Minho, Guimara˜es, Portugal Keywords Medical, Garments, Evaluation, Perception Abstract More and more disposable goods are available in surgical rooms. Rules and standards have been proposed in order to prevent infection from patient to surgical team and vice versa. A proposed mandatory European standard prEn 13795 “Surgical drapes, gowns and clear air suits used as medical devices, for patients, clinical staff and equipment”, is being developed by the Committee of European Normalisation and specifies the basic performance requirements and test methods for single-use and reusable materials after sterilisation process. Therefore, the performances of the surgical gowns demand a balance between barrier and comfort properties. In comfort evaluation, tactile feeling is one of the most primary and important aspects with regard to the grading of the products. Therefore, the influence of the sterilisation process on the tactile perception is important to be evaluated. Subsequently, the final aim of this paper is to contribute to the knowledge of influence of sterilisation treatment on the tactile perception.
International Journal of Clothing Science and Technology Vol. 15 No. 3/4, 2003 pp. 268-275 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310478378
1. Introduction The goals to be pursued by this presentation are to show the influence of the sterilisation methods over non-active medical devices used for medical and surgical applications. Essentially, the impact of ionising radiation (gamma and beta radiation) considered low temperature sterilisation, for single use materials, like non-woven based surgical gowns. Gamma and beta radiation are often used for the sterilisation of polymeric based disposable medical devices. In order to investigate the influence of the radiation on the tactile properties of non-woven based surgical gowns, three types of material (non-woven, laminate and polyethylene) and two radiations
were considered in this study. Nevertheless, this sterilisation can induce some changes in their tactile behaviour and could restrict their use for the defined applications. So, this article intends to relate the results of the evaluation of tactical properties of surgical gowns after irradiating at several doses in a range from 0 to a maximum dose of 160 kGy. This dose was determinated by DSC and TGA techniques. The materials considered in our study do not present high levels of degradation, for dose values lower than 100 kGy in the case of the non-woven, 130 kGy for the polyethylene and 200 kGy for the laminate. One of the basic ideas is that the performance of the surgical gowns demands a balance between barrier and comfort properties (Sishoo, 1981). Subsequently, the final aim of this paper is to contribute to the knowledge and clarification of the impact of the different sterilisation methods over the comfort, through tactical sensory analysis of the materials used in surgical gowns. Furthermore, the materials used to manufacture this type of non-active medical device – non-woven based surgical gowns – are not only textile materials, but also laminates and polymers, where a combination of specific properties is required and the interaction of these materials is still unknown. The microbiological environment, the type of material and their related properties and interaction with the lethal agent are the parameters that always have to be considered. 2. Directives and standards According to the Medical Device Directive 93/42/EEC, surgical clothing, drapes and air suits are considered to be the medical devices, regardless of whether they are reusable or disposable. The CEN (2001) (European Committee of Standardisation) working group (TC 205/WG 14) proposed a mandatory European standard prEN 13795 surgical drapes, gowns and clean air suits, used as medical devices, for patients, clinical staff and equipments. This standard specifies the basic performance requirements (properties) and test methods for single-use and reusable materials used to protect the patient, surgical staff and surgical facility. These requirements must also be fulfilled after the sterilisation process, underlying the importance of this study for the good development of this European standard. The first part of the European standard concerning general requirements for manufacturers, processors and products has been published and gives a general guidance on the characteristics of single-use and reusable surgical gowns (Table I), surgical drapes and clean air suits used as medical devices for patients, clinical staff and equipment and intends to prevent the transmission of infective agents between patients and clinical staff during surgical and other invasive procedures.
Influence of the sterilisation process 269
IJCST 15,3/4
270 Table I. Requirements for surgical gowns
Requirements for surgical gowns Resistance to microbial penetration - dry Resistance to microbial penetration - wet Cleanliness - microbial Cleanliness - particulate matter Linting Resistance to liquid penetration Bursting strength - dry Bursting strength - wet Tensile Strength - dry Tensile Strength - wet
3. Tactile evaluation The standards previously presented mainly take into account the mechanical and physical parameters, in order to validate the protective functions. The comfort aspect is, from this point of view, not clearly envisaged. In our study, this aspect has been developed through the gowns and textile medical goods tactile feeling evaluation. These evaluations have been performed, thanks to the sensory analysis in order to highlight the influence of the sterilisation process on the comfort and tactile feeling. 3.1 How do we evaluate the tactile perception of the textile goods? In textile field, the instrumental tests are widely used and many standards have been designed in accordance with equipment development and the required characteristics. These tools are usually fast, repeatable and well understood; however, they may not precisely represent the textile goods in use. Moreover, the tactile evaluation cannot be extracted from these sets of numerous data even some and evaluations have been done in this way through “Hand evaluation of Fabrics” (Kawabata, 1980). In order to perform this evaluation, the sensory methods using the human being as a measurement tool have been envisaged. These methods are composed of two main parts, the hedonic analysis and the sensory analysis (AFNOR, 1999). The latter is defined as the examination of sensory attributes by the sense organs, whatever the end-use domain may be. It is based on the work of a trained panel. A sensory profile is drawn out from this methodology. The hedonic analysis is issued from the consumer’s enquiries. It takes the enduse applications into account and the “like” or “don’t like” comments of the consumers. These two approaches can then be correlated in order to link the obtained sensory profile and the consumers’ perception. A preference mapping, which displays the consumer’s expectations and the product description, can be obtained (Depledt, 1998). One disadvantage of these sensory methodologies is the time consumption due to the panel training and the validations of each step of the method.
3.2 Tactile panel presentation As explained previously, the evaluation is performed with the help of a panel composed of trained assessors, who describe, quantify and grade – thanks to attributes and scales – the tactile perception of textile goods. The used attributes are chosen through strict procedures in order to obtain “consensus”, full understanding and statistical reliability of the panel assessors. To reach this aim, numerous training sessions have to be performed using a huge range of textile surface. During the evaluation, facilities have been chosen and evaluation protocol have been established to each selected attribute (Plate 1). To interpret the obtained data, statistical tools such as analysis of variance (ANOVA) and principal component analysis (PCA) are used (Kawabata, 1980). The ANOVA computation is required to verify the homogeneity of the group that is to say if the attributes have been understood in the same way, it also allows assessor repeatability to be validated. The PCA is a technique used to simplify complex data interpretations. By considering the correlation between large number of variables (attributes), a PCA will seek out factors or components in which the variables have a great deal in common and allows the differences and similarities of the products in the map products to be visualised.
Influence of the sterilisation process 271
4. Experimental part 4.1 Tested products For experimental part, the products, issued of different kind of gowns, presented in Table II have been evaluated. The nature of each layer is displayed in Table III.
Plate 1. Assessor under test conditions
Material Gown
Outer layer
Inner layer
Type 1 Type 2 Type 3
Non-woven PE Non-woven
Non-woven Non-woven Laminate
Table II. Composition of tested products
IJCST 15,3/4
4.2 Sterilisation procedure used Different kinds of sterilisation processes have been tested for several doses of irradiation. For this presentation, the sterilisation processes and doses displayed in Table IV will be presented.
272
5. Results and discussion The reliability of the panel is controlled after each session and the pertinence of the attributes is issued from ANOVA, factorial and ranking test analysis. In this case, nine attributes are detected as pertinent by the statistical analysis and by the panel. The obtained results can be displayed under specific visualisations. The socalled “profile” is the simplest and the most commonly used. It represents, for a specific product or a group of products, the relevant attributes versus their intensity. Figure 1 shows the results obtained for the non-irradiated non-woven Material
Table III. Nature of each layer of tested product
Non-woven
45 per cent PET - 55 per cent Cellulose Spunlace
Laminated
LDPE - 20 g/m2 + Non-woven 70 per cent viscose – 30 per cent PET
Radiation Table IV. Kind of radiation and doses used
Figure 1. Non-woven profile
Nature
Dose
b-radiation
25, 80 and 150 kGy
g-radiation
25, 80 and 150 kGy
profile compared with b and g irradiated material for different doses. Figure 2 shows the results obtained for the non-irradiated laminate compared with b and g irradiated material for 150 kGy doses. The ANOVA tests highlight the fact that tactile felling is not affected by the radiation treatment, however, a PCA analysis gives us more information in terms of the predominant trends for the tactile feeling as shown in Figure 3. On the product map (Figure 3), two axes are used. One axe is composed of a set of three attributes “pileux” (hairy), “nerveux” (nervous) and “granuleux”
Influence of the sterilisation process 273
Figure 2. Laminate profile
Figure 3. PCA analysis for the sensorial tactile evaluation
IJCST 15,3/4
274
(granulous) and a set of five attributes “e´pais” (thick), “lourd” (heavy), “froissable” (crumple), “glissant” (slippy) and “doux” (soft). The second axe represents “collant” (sticky) attribute. More than 80 per cent of the whole information is kept in this map, which is a satisfactory result in terms of PCA analyses. Two groups of materials emerged from the graph above: (1) laminate group, and (2) non-woven group. In the case of non-woven group, it presents following characteristics: . “glissant” (slippy), . “doux” (soft), . “lourd” (heavy), . “froissable” (crumple), and . “e´pais” (thick). The laminate group can be condidered as . “nerveux” (nervous), . “granuleux” (granulous), and . “pileux” (hairy). In terms of the radiation treatment effect on the laminate products, the sticky attribute highlights significant differences between b and g treatments. b-treated product is more sticky than the g-one. For non-woven products, the obtained results are complex and are in accordance to mechanical and physical behaviour, which has shown a contradictory phenomenon between cellulose and polymeric components after being irradiated. In fact, for low irradiation doses, cellulose degrades as reticulations are observed for polymer and therefore, complete behaviour of the composite has to be done considering the percentage of the non-woven blend. 6. Conclusion The effect of sterilisation on the tactile feeling for surgical gowns has been studied. Thanks to the new tool called sensory analysis which gives us some interesting results. For the tested product and in range of process and doses (including the standard proposed dose), the tactile modification has been described and detected by the panel of assessors. The results show that there are differences between products slight , which do not greatly affect the final “touch” of the tested products. Other tests have simultaneously been carried out in the fields of mechanical, thermal and physical properties in order to understand the influence of sterilisation process over the micro and the macro structure of the composite. A link between instrumental approach will permit to bind panel evaluation and conventional objective fabric evaluation.
References AFNOR XP V 09-501 (1999), Sensory analysis-general guidance for sensory evaluationdescription, differentiation and hedonic measurement. CEN/TC 205/WG14 – pr EN 13795 (2001), Surgical drapes gowns and clean air suits, used as medical devices, for patient, clinical staff and equipment – Part 1: General requirements for manufacturers, processors and products. Depledt, F. (1998), Socie´te´ Scientifique d’Hygie`ne Alimentaire (SSHA): Evaluation Sensorielle Manuel me´thodologique, 2nd ed., Lavoisier TEC&DOC, France. Kawabata, S. (1980), The Standardisation and Analysis of Hand Evaluation, 2nd ed., Textile Machinery Society of Japan, Osaka, Japan. Sishoo, R.L. (1981), “Technology for comfort”, Textile Asia.
Influence of the sterilisation process 275
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
IJCST 15,3/4
276
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Numerical simulation of textile flexibility testing Zˇeljko Sˇomodi, Anica Hursa and Dubravko Rogale Department of Clothing Technology, Faculty of Textile Technology, University of Zagreb, Zagreb, Croatia Keywords Clothing, Numerical simulation, Testing Abstract A non-linear numerical simulation of a standard procedure for textile flexibility testing is performed using discretised beam bending model. Geometric non-linearity due to large deflections is traced using incremental method. Linear moment-curvature response is assumed, as well as constant curvature of a finite element of the beam. Numerical procedure is incorporated into a PC programme producing graphical results for the deformed shape of the specimen, nonlinear load-deflection diagrams and internal force distributions in deformed state. Finally, the method is applied to compute the flexural stiffness of textile materials from the data produced by the standard procedure for flexibility testing.
International Journal of Clothing Science and Technology Vol. 15 No. 3/4, 2003 pp. 276-283 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310478387
1. Introduction Numerical modelling constitutes an important area in the field of methods for engineering analysis. In the last few decades it has become a strong counterpart to analytical methods of function analysis, which remain suitable only for the cases of extremely simple geometry. In particular, the finite element method (FEM), as a mean of transposing a boundary value problem of the field theory to the solution methods of discrete algebra, has received a rapidly growing attention and popularity in applied mechanics and other fields of engineering (Zienkiewicz, 1977). It was developing pretty much along with the growing numerical and graphic power of computers, and at present contemporary software based on finite element analysis is becoming a part of the standard engineering analysis tools. In contrast to solid body mechanics, which is applied in mechanical and civil engineering, soil mechanics etc., numerical modelling is relatively recent in textile and clothing engineering. The main reason for this fact is existence of severe non-linearities in the mechanical behaviour of textile. Non-linearities – material, geometric or especially a combination of the two – require sophistication in mathematical description and often iterative solution procedures (Crisfield, 1997; Potluri et al., 1996). In this paper, we present our starting effort in introducing numerical modelling of textile mechanics in the research field studied at the Department of Clothing Technology of the Faculty of Textile Technology, University of Zagreb. Aiming at the problems involving flexural deformation (free or forced formation in draping, introduction of textile in auxiliary devices for sewing), our initial goal is to establish the relationship between a possible simple description of moment-curvature response and a standard testing procedure for
textile flexibility. A similar study was published by Fridrichova and Mevald Numerical (2000), and in the model proposed in this paper certain improvements are simulation of offered. textile flexibility 2. Discrete model of geometrically non-linear beam bending First, let us briefly describe the standard procedure for textile flexibility testing. The specimen – band of the given textile material – is held in the position of a cantilever beam (one end fixed and the rest free of any support) of a gradually increasing length. The movement of the specimen stops when its free end touches the inclined plane coming from the point of the fixed end at an angle of 41.58 with respect to the horizontal. The free length of the specimen at that moment is taken as the measure of the flexural rigidity of the material (Cˇunko, 1995; DIN 53362, 1970). In FAST measurement system for objective evaluation of mechanical properties of fabrics, a beam of light is used in the place of the inclined plane. The specimen is at the end of the described test obviously well bent, by primarily in the vicinity of the fixed end, and its geometry is far from the small displacement supposition usual in the elementary strength of materials. Assuming constant flexural stiffness in moment-curvature response of the beam element 1 M ¼ r cfl
ð1Þ
where r is the radius of curvature, 1/r is the curvature of beam element, M is the bending moment and cfl the flexural stiffness. Non-linear structural load-deflection response can clearly be expected at large deflections. Loading by self-weight of the beam can be described as follower-conservative, in the sense that it is attached to the element of the beam and remains vertically downward regardless of the deflection. In this work, the simple incremental tracing of non-linear response is adapted, as preferred to an iterative procedure (Hinton and Owen, 1980), due to its simplicity and consistency with the programming possibilities. In this approach, the final deformed shape is achieved by the small changes of a parameter responsible for deformation, producing the growth of deflection from zero to the final state. Here, we can identify two such parameters: (1) uniformly distributed load per unit length q resulting from the weight of the beam, (2) flexural stiffness cfl. In order to compute the bending moment distribution and geometry in the deformed shape, the length of the beam is discretised into finite number of short segments (Figure 1).
277
IJCST 15,3/4
278
Figure 1. Discretisation of the beam into finite segments
Bending moment in each segment is computed at its mid-section as the moment resulting from the weight of the right hand part of the beam: Mi ¼
X qDl 2i qDl j ðxj 2 xi Þ þ cos wi ; 8 j.i
i ¼ 1; . . .; n
ð2Þ
In this expression, xi represents the coordinate of the mid-section of the ith segment, while xj represents the coordinate of the centroid of the jth segment. The last term in equation (2) results from the weight of the right half of the ith segment. Denoting by fi, the angle of inclination at the right hand side of the ith segment, and assuming constant curvature along each finite segment, the coordinates of the centre of curvature for each segment are computed as (see Figure 1) RS xCi ¼ xi21 2 ri sin wi21 ;
RS zCi ¼ zi21 þ ri cos wi21
ð3Þ
where the coordinates of the right hand side of each segment are xiRS ¼ xCi þ ri sin wi ;
ziRS ¼ zCi 2 ri cos wi
ð4Þ
Obviously, at the left hand side of the beam ði ¼ 1Þ we have the boundary conditions of the fixed end x0RS ¼ z0RS ¼ w0 ¼ 0
ð5Þ
The small central angle of a curved segment is equal to the ratio of its length over its radius of curvature
Dwi ¼
Dl i ri
ð6Þ
and the angle of inclination at the right hand side of ith segment is obtained by accumulation: X Dwj wi ¼ ð7Þ j#i
With this algorithm at hand, we can perform the incremental tracing of the load-deflection (q-w) curve starting from the undeformed geometry, adapting the initial/incremental load Dq corresponding to the prescribed small deflection (e. g. 1 ¼ 0:01) at the free end w0 ¼
Dql 4 8cfl 1 ¼ l · 1 ) Dq ¼ 3 l 8cfl
ð8Þ
and computing the deformed geometry at the load level q from the bending moment distribution in the previous step M(q – Dq). While the procedure described above computes the deflection for increasing load q with constant flexural stiffness cfl and beam length l, the similar procedure can be set for the case of the standard flexibility test: with the given load q (mass per unit area of material is known) and length l (result of flexibility test), incremental computations starting from the small initial/incremental flexibility 1/cfl are terminated once the condition wðlÞ=l . tan 41:58 is met. Flexural stiffness at that step cfl is then stored as the final result of the computation. 3. Computer implementation The described numerical procedure has been built into a PC programme written in MS Visual BASIC. Its programming possibilities (Perry, 1998) are used to organise simple data input and graphic representation of both deformed beam and non-linear load-deflection curve. In fact, three different simulations are programmed; the first of which produces the q-w curve and corresponding pictures of beam deflection for increasing load q with constant l and cfl. Additionally, as the previous deformed shape can be erased, the distributions of internal axial forces and bending moments are shown as diagrams along the deformed line of the beam. The second simulation is based on the data from the standard flexibility test as described in the final paragraph of Section 2. Finally, the data cfl acquired by this second simulation is used in the third simulation in which an animated sequence of flexibility test with slowly increasing beam length is reproduced. The figures showing the described simulations follow with the examples in the next section.
Numerical simulation of textile flexibility 279
IJCST 15,3/4
280
4. Examples 4.1 Analysis of discretisation sensitivity In the first example, let us consider the influence of discretisation upon the computational results. Two parameters concerning discretisation are varied in the range from coarse to fine as follows. In the first case, number of finite segments ¼ 5 (coarse), 10 (medium), 20 (fine) with constant segment length scaling factor of 1.1 and constant initial/incremental load step factor 1 ¼ 0:05 (equation (8)). The results – dimensionless deflection at the free end w at the two load levels q ¼ q0 and q ¼ 2q0 ; are listed in Table I. Reference load q0 would correspond to the case when the deflection at the free end given by the formula for small deflections equals the length l: q0 ¼
8cfl l3
ð9Þ
In the second case, load increment factor 1 ¼ 0:1 (coarse), 0.05 (medium), 0.02 (fine) with ten segments scaled with factor 1.1. The results are given in Table II. The results given in Tables I and II indicate that computations based on 20 segments and load step factor 0.02 can be considered as reasonably accurate. 4.2 Load-deflection response for increasing load In this example, the beam is discretised into ten segments scaled with the scaling factor 1.1 (Figure 2(a)). The load step factor is 0.05. The curve showing the non-linear load-deflection relationship is produced by the programme and shown in Figure 2(b). Here, we also demonstrate the distribution of internal axial forces and bending moments. In Figure 2(c), the internal axial force distribution is given for the load q ¼ q0 (equation (9)) in the form of a diagram following deformed shape of the beam. Similarly, the internal bending moment distribution is given in Figure 2(d) for the load q ¼ 2q0 : Table I. Computed deflections for three levels of geometric discretisation
Table II. Computed deflections for three values of initial/incremental load
Number of segments w/l at q ¼ q0 w/l at q¼2q0
Load step factor 1 w/l at q ¼ q0 w/l at q¼2q0
5
10
20
0.638647 0.790997
0.644874 0.797885
0.6461175 0.799128
0.1
0.05
0.02
0.650938 0.800266
0.644874 0.797885
0.6413065 0.796444
Numerical simulation of textile flexibility 281
Figure 2. (a) Discretisation of the beam, (b) load-deflection curve, (c) deformed shape with axial forces at q ¼ q0 ; (d) deformed shape with bending moments at q ¼ 2q0
4.3 Simulation of flexibility test In this final example, the flexural stiffness, of textile is computed for the parameters obtained by measurements. With the given load (mass per unit area, g/m2) and result of standard flexibility test (length of the specimen at which the inclined line at 41.58 is first touched), the non-linear deformation is traced by gradually increasing the flexibility 1/cfl (see final paragraph of section 2). The computed cfl results for the typical range of input parameters are listed in Table III. To obtain cfl in (Nm2/m), multiply the values in Table III by 102 6. Figure 3 shows a typical screen picture produced by the programme for simulation of flexibility testing. In Table III, a high degree of linearity between cfl and mass per unit area is observed (for constant length of the specimen, flexural stiffness is proportional to the mass per unit area of textile).
IJCST 15,3/4
282
5. Concluding remarks Numerical modelling of mechanical behaviour of textile is related to considerable non-linearities. In this paper, the geometric non-linearity of bending deformation is traced using simple incremental approach, and it seems promising for possible future analyses of more complex situations as well. Numerical discretisation of geometry and load increment is reasonably balanced between demands for accuracy and simplicity. In our experience, computer time consumption in these simple analyses was of no real concern, but in some cases of very fine discretisation, an error was reported by Visual BASIC interpreter (“the program has performed an illegal operation”). This problem has limited our analysis of discretisation sensitivity to some extent.
Mass per unit area (g/m2) Length (mm) Table III. Computed flexural stiffnesses
Figure 3. Simulation of the flexibility test
50 100 150 200
50
100
150
200
250
300
7.66 61.3 207 491
15.3 123 414 981
23.0 184 621 1,472
30.7 245 828 1,962
38.3 307 1,035 2,453
46.0 368 1,242 2,943
As far as the textile flexibility is concerned, a clear relationship between Numerical flexural stiffness in simple materially linear model (equation (1)) and data simulation of acquired by the standard procedure for flexibility testing is established, as textile flexibility shown in Table III. For the case of a specific value of specimen length other than those in Table III, an approximate result can be obtained by linear interpolation between the two neighbouring values.
283
References Crisfield, M.A. (1997), Non-linear Finite Element Analysis of Solids and Structures, ISBN 0-47197059-X, Wiley, New York. ˇCunko, R. (1995), Ispitivanje tekstila, Sveucˇilisˇte u Zagrebu, ISBN 86-329-0180-X, Zagreb. DIN 53362 (1970), “Testing of plastics films and coated fabrics, manufactured using plastics; determination of stiffness in bending”, pp. 56-7. Fridrichova, L. and Mevald, J. (2000), “Bending rigidity of textiles”, in Militky, J. (Ed.), Proceedings of 4th International Conference TEXSCI 2000, 112-115, ISBN 80-7083-409-9, June 2000, Technical University Liberec, Liberec. Hinton, E. and Owen, D.R.J. (1980), Finite Elements in Plasticity: Theory and Practice, ISBN 0-906674-05-2, Pineridge Press, Swansea. Perry, G. (1998), SAMS Tech Yourself Visual Basic 6 in 21 Days, ISBN 0-627-31310-3, SAMS – A Division of Macmillan Computer Publishing, Indianapolis. Potluri, P., Atkinson, J. and Porat, I. (1996), “Large deformation modelling of flexible materials”, The Journal of the Textile Institute, Vol. 87 No. 1, ISSN 0400-5000, pp. 129-51. Zienkiewicz, O.C. (1977), The Finite Element Method, ISBN 0-07-084072-5, McGraw-Hill, New York.
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
IJCST 15,3/4
284
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
The physical properties of the surface of apparel made from flax and polyester fibres Małgorzata Zimniewska Institute of Natural Fibres, Poznan, Poland
Marina Michalak and Izabella Krucin´ska Department of Textile Technology, Technical University of Lodz, Lodz, Poland
Bogusław Wie˛cek Institute of Electronics of Technical University of Lodz, Poland Keywords Synthetic fibres, Clothing, Physical properties Abstract In this paper, the clothes made of synthetic and natural fibres were tested. The characteristics of selected physical parameters such as temperature, electrical resistance, thermal resistance of fabrics used for tested clothes have been presented. The electrostatical charge and temperature distribution of clothes were investigated on human body. The temperature distribution and the coefficient of heat transmission were measured by a new thermovision method.
International Journal of Clothing Science and Technology Vol. 15 No. 3/4, 2003 pp. 284-294 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310478396
1. Introduction Clothing comfort is a complex and hazy subject that is hard to define in a few simple words. Most of all, this is a result of comfort being set of personal individual feelings (Figure 1). However, briefly, we can call comfort as all physiological human body reactions to the conditions related with environment-clothes system. Covering most parts of our body for most of the time in our daily life, clothes contact most parts of the skin dynamically and frequently. This produces various mechanical, thermal, chemical or electrical stimuli. Human skin is the interface between the human body and its environment. It is richly innervated and contains specialized sensory receptors to detect various external stimuli. Responding to these stimuli, the skin receptors produce the sensation of touch, warmth or cold and pain (Figure 2). Sensations generated from clothing depend on various combinations of human activities and environmental conditions experienced during day-to-day living. Researches have identified commonly recognized attributes of clothing related to comfort, involving thermal, moisture, tactile, hand and aesthetic experiences. But the aim of our research was not regarding he comfort itself, but the changes on human body caused by the cloth. The last research has been conducted on the effect of clothes made of synthetic and natural fibres,
The physical properties of the surface 285 Figure 1. The fabric interacts with the skin
Figure 2. Neutral pathways from the skin to the brain
temporarily covering the muscles of a forearm, on the action of motor units of those muscles in rest and exercise conditions. Additionally, the conductivity of motor fibres in nerve branches connecting the tested muscles was investigated. The electromiographic parameters of muscles were tested by EMG Neurorapid RunTime 10/20 apparatus, using surface electrodes. The temporary covering of the tested forearm muscles with a synthetic fabric changes the pattern of the motor unit activity. It is showed by a low frequency spontaneous activity of muscle fibres during rest or by a reduced high frequency activity during exercise. The presence of these phenomena is
IJCST 15,3/4
286
correlated with a slight alternation in nerve fibre conductivity of branches connecting muscles mentioned earlier. It is not, however, a pathology. Covering of arm with a natural fabric causes no similar effects. The changes observed are the cause of increased tendency of getting tired when wearing clothes made of synthetic fibres. The cause of such test results are most probably due to the following phenomena accompanying the polyester wear: . collecting electrostatic charges on the surface, . low air permeability, . low hygroscopicity, . increased sweating, and . higher body temperature. 2. Materials The clothes made of synthetic and natural fibres were tested. The linen represented the natural fibres. The synthetic fibres were represented by polyester, because it is the most commonly used fibre in clothing industry. The fabrics were tested in the form of men’s shirts made of fibres mentioned above. The shirts were identically designed, with long sleeves, had the same geometry and were of suitable size for the users (Table I). The main work in this study was focussed on investigating the electrostatic and thermal properties of clothes, since these properties are presumably responsible for previously observed changes in the EMG curves. The timeconstant and electrostatic charge potential were tested on the surface of clothes, differentiated only by the applied end-use finish: . washed clothes with no special anti-electrostatic end-use finish, . washed clothes with special anti-electrostatic end-use finish, and . unwashed factory-new clothes. 3. Methods 3.1 The ability of collecting electrostatic charges The man moves constantly performing day-to-day works; he moves his hands rubbing the trunk tens of times with the sleeves. In this way, the electrostatic
Hygroscopicity Table I. Selected physical parameters of tested clothes without antistatic finish
Material 100 per cent linen 100 per cent PES
At 65 per cent humidity of air
At 100 per cent humidity of air
Surface resistance [V]
Heat resistance [Km2/W]
9.3 1.0
17.1 1.3
1.5 £ 109 6.5 £ 1011
14.8 5.4
charges are collected on the surface of clothes. During the study, an The physical electrostatic potential was registered on the surface of clothes, before and after properties of the 20 cycles of rubbing shirt with a sample of cloth coming from the same fabric. surface The initial measurement of the potential in all tested cases was 0 kV. The tests for the degree of electrostatic charges collected were done by measuring the potential difference of an electrostatic field at the fixed distance 287 from the shirt. The magnitude and mark of the electrostatic field depends on the mark and magnitude of a charge forming this field, i.e. charges collected on the surface of the shirt and a distance from the source of the field. The measurements were done by a rotation meter of HAUG Company. The detector of this meter has a rotating disk with segment of holes. The charges generate alternated electric signal, which is directed to the electronic unit of the apparatus. The meter gives readings of values of electrostatic field in the position of the detector. The meter can measure the potential difference from 100 V to 100 kV in several measurement ranges. In the case of electrified surface, the value determining the electrostatic field is the surface density of electrostatic charges. To avoid the error connected with the geometry of the tested object and position of a meter in electrostatic field, no density of electrostatic charges was measured. Instead, a size and mark of potential difference of electrostatic field was determined. In the measurements of electrostatic field the distance between the detector and the object is very important. The change of a distance causes the change of an electrostatic potential difference value at the same density of the charge. The measurements were done at the 15 mm distance between the detector and shirt surface. To assure the stability of measurement conditions, the meter was fixed on a mount.
3.2 Testing of temperature distribution The temperature measurements were done by a thermovision method. In this method, the distribution of the temperature is visualized by a non-contact method on the surface of a tested object by measuring the infra red radiation. The thermovision method is an observation and recording of radiation distribution and transforming this radiation into visible light range. The infrared radiation is an electromagnetic radiation with the wave range from 0.78 to 1,000 mm. Practically, the method uses detectors measuring two main wavelength ranges: 2-5 mm (so-called medium wave infrared radiation) and 8-4 mm long wave infrared radiation. Sometimes, it uses detectors measuring the wavelengths of 14-30 mm. The thermovision investigations are not quite new in textile science (Berardi and Cucurullo, 1994; Okamoto et al., 1998; Tournerie et al., 1991; Vavilov et al., 1998). In this paper, preliminary results for thermal parameters determination by thermal wave method is presented. This method is new and is applied to textiles for the first time.
IJCST 15,3/4
288
Figure 3. Block diagram and system setup
3.2.1 System architecture. The system used in the survey, consists of analogue and digital parts as shown in Figure 3. In the analogue one, a 8-bit A/D converter is used to convert thermal images. A/D flash converter for CCD camera is equipped with four-input analogue multiplexer. It makes possible to capture visual images one-by-one, with the delay of one frame (20 ms), but in parallel with thermal image. In the digital part 1 MB buffer VRAM memory is used to store images and transfer them to the computer. A dual-port memory is suitable for fast data acquisition, and it easily allows synchronizing the data capturing with its transfer to the operational memory. The entire control system is integrated in high density field programmable gate array (FPGA), providing high flexibility of the design and possibility of system reconfiguration for different cameras. Actually, our system is prepared for various thermal cameras with both analogue and digital outputs. The implemented memory buffer uses fast dual-port memories (VRAM) and was designed as a first-in-first-out (FIFO) memory. Input data coming from A/D converters is loaded into the memory through serial register (SAM). The corresponding line counter for locating the memory row, where the pixels are
placed in, is implemented in FPGA. This control block also contains similar The physical counters for pixels and lines to read data from the memory and transfer them to properties of the the computer. Software for processing thermal and video images and their surface sequences was written on Delphi and C++ language as 32-bit application. It is Win’95w and WinNTw compatible. This software is dedicated to thermal, visual and radiological image processing in parallel. There are various 289 functions available to measure the temperature easily and precisely. 3.2.2 The measurement stand. The measurements were conducted on a measurement stand in which a source of infrared radiation (a 3 kW lamp) was located on one side of a sample, while a thermovision camera on the other side of the sample. Thus, a distribution of temperature on the nonirradiated sample could be measured. The sample was in a perpendicular position to the optical axis of the thermovision camera. The distance between the source of radiation and the sample and from the sample to the camera was the same (40 cm). A block diagram of the measuring position is shown in Figure 4. The surface of the sample was divided into two areas. One was irradiated, while the other was covered with a special screen of low heat volume and low reflection coefficient. A scheme is shown in Figure 5. The surface “A” shows irradiated part, while part “B” shows non-irradiated part of the sample. The irradiation was conducted periodically at the frequency of 0.5 Hz by a light wave method. The dynamics of the temperature distribution on the surface B was registered in both the cases. The data management was carried out in off-line mode. The 100 per cent linen and 100 per cent PES fabrics were tested after washing. No anti-electrostatic end-use finish was applied. 4. Results 4.1 Time-constant and the ability of collecting electrostatic charges The time-constant is a measured time, in ms, in which the electrostatic charges collected on the surface of the fabric were discharged by 67 per cent. The limits
Figure 4. Block diagram of a measuring position with a sample placed between the source and the camera
Figure 5. The example of sample
IJCST 15,3/4
290
Figure 6. The time-constant of disappearance of electrostatic charges for 100 per cent linen and 100 per cent PES fabrics
Figure 7. The electrostatic potential on the surface of tested clothes after different methods of finishing and after 20 cycles of rubbing
of time-constant measurement: the upper value of potential difference of timeconstant is 150 V, the lower is 50 V (Figure 6). The low value of time-constant of electrostatic charges discharge for the linen fabrics shows that linen, unlike polyester, does not collect electrostatic charges on the surface. A person wearing polyester clothes is permanently exposed to the effect of electrostatic field and flashovers when he/she approaches conducting objects. This is caused by a high ability of PES to collect electrostatic charges. The results of the ability of collecting electrostatic charges are shown in Figure 7. Regardless to the end-use finish of tested fabric,
the man wearing polyester clothes is permanently exposed to an electrostatic The physical field. Household anti-static products only partially improve this situation. properties of the Conditions of testing: air humidity 40 per cent, temperature 208C. surface 4.2 Thermography The Figure 8(a) and (b) shows two single thermograms obtained from a linen fabric (a) and polyester fabric (b) irradiated by a lamp with a frequency of 0.5 Hz. The thermograms were registered after 14 s from initializing the irradiation at the maximum radiation of the source. The colour pattern on the thermograms is designed in such way that the highest temperature is 32.68C. The pink colour represents the temperature of the background, which is 22.68C. The conditions of the experiment were selected in the way to visualize the border between discussed areas, which is characterized by higher temperature (white colour on the thermogram). The irradiated part of the fabric warms up as an effect of radiation. The heat then spreads to the non-irradiated area. This area was characterized by the lower temperature as compared to the irradiated one. As mentioned before, the distribution of temperature on the surface opposite to the source of radiation was recorded. The comparison of thermograms shows that the irradiated part of the polyester fabric is much warmer than linen one ( polyester – white colour, linen – red). Additionally, the area of higher temperature on the non-irradiated part of the linen fabric is much wider than that of polyester fabric. These phenomena show that polyester and linen fabrics are characterized by different coefficients of heat take-up. Further, figures show thermograms for tested fabrics registered in a whole period of irradiation. The process of heat wave propagation is clearly visible. The Figure 9(a) and (c) refers to linen fabric. The Figure 9(b) and (d) refers
291
Figure 8. (a) The thermogram for linen fabric. (b) The thermogram for polyester fabric
IJCST 15,3/4
292
Figure 9. The temperature distribution (a and b) and isotherms along the TT line (c and d)
polyester fabric. Temperature distribution on thermograms along randomly chosen TT line is presented later (Figure 8(a) and (b)). In this case, a colour set is different from the one showed in Figure 8(a) and (b). The highest temperature is represented by the darkest area of the thermogram. To make the analysis easier, the isotherms were applied on Figure 9(c) and (d). The differences of heat wave propagation between polyester and linen fabrics are easily visible on compared isotherms. The Figure 9(a) and (c) and (b) and (d) shows periodical responses of non-irradiated area of tested fabrics to heat input with given area of irradiation. The vertical axis shows the temperature distribution patterns along the TT line (Figure 8(a) and (b)). The horizontal axis is a time axis. The time scales on Figure 9(a) and (b) differ. The same remarks refer to Figure 9(c) and (d), respectively. The lines parallel to the horizontal axis specify the distance of irradiated and non-irradiated area from the borderline. To make an example, the “00” line is marked on all figures. The area irradiated with a lamp is not included in discussed figures. The responses to heat input have periodical character consistent with the period of input. Each subsequent cycle of radiation in the non-irradiated area is characterized by widening the range of heat wave propagation and increasing temperature. In the case of linen fabric, the area of higher (periodically variable) temperature is wider (Figure 9(a) and (c)) than in the case of polyester fabric
(Figure 9(b) and (d)). The temperature on the lines equally distant from the The physical border for linen fabric is higher than for polyester fabric. The apparent properties of the differences in heat wave propagation in linen and polyester fabrics are visible surface on compared thermograms. A mathematical interpretation of results, using the method of heat wave presented in this study, offers the possibility of quantitative assessment of 293 characteristic thermal parameters. The work is still in progress. It should be emphasized that the heat wave method has been used for the first time in textile research. 5. Conclusion (1) The application of thermovision method for presented studies showed a possibility of using thermal wave method in textile testing. (2) The results on the study of time-constant of electrostatic charge distribution show that on the surface of tested polyester clothes the charges stay much longer than on linen clothes. This phenomenon is especially apparent when fabric-new, unwashed clothes and clothes washed with no addition of anti-electrostatic preparations are tested. (3) The polyester fabrics show high variability regarding ability of gathering of electrostatic charges on the surface, regardless to the end-use finish applied. (4) Based on the analysis of thermograms, it can be found that the heat exchange between the skin and environment is much easier in the case of linen than in polyester clothes. The body temperature under the polyester fabric may rise. (5) Both, electrostatic field from charges collected on the surface of the polyester clothes and increased temperature of skin, have a permanent effect on the human body and may cause changes in the electromiographic records of muscle tension, which prove desynchronicity of movement units, resulting in a higher tendency to get tired. References Berardi, P.G. and Cucurullo, G. (1994), “A procedure to measure thermal conductivities of anisotropic laminates by infrared thermography”, Quantative InfraRed Thermography, QIRT 94, 23-26 August 1994, Sorrento (NA) Italy, pp. 81-5. Okamoto, Y., Kamoi, A. and Ishii, T. (1998), “Thermal analysis on internal and surface flaws by means of an infrared radiometer”, Quantative InfraRed Thermography, QIRT 98, 7-10 September 1998, Lodz, Poland, pp. 71-6. Tournerie, B., Reungoat, D. and Frene, J. (1991), “Temperature measurements by infrared thermography in the interface of a radial face seal”, Journal of Tribology, Vol. 113, pp. 571-6.
IJCST 15,3/4
294
Vavilov, V.P., Almond, D.P., Busse, G., Grinzato, G., Krapez, J-C., Maldague, X., Marinetti, S., Peng, W., Shirayev, V. and Wu, D. (1998), Quantative InfraRed Thermography, QIRT 98, 7-10 September 1998, Lodz, Poland, pp. 43-52. Further reading Zimniewska, M., Michalak, M., Krucin´ska, I. and Wie˛cek, B. (2002a), “The comparison of the physical of the surface of clothing made from flax and polyester fibres”, Proceedings of the 2nd International NETECOFLAX Workshop, 25-27 January 2002, Covilha, Portugal. Zimniewska, M., Huber, J., Krucin´ska, I., Torlin´ska, T. and Kozłowski, R. (2002b), “The influence of natural and synthetic fibres on activity of the motor units in chosen muscles of forearm”, Fibres and Textiles in Eastern Europe, No. 4.
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Evolutionary algorithms aided textile design Darko Grundler and Tomislav Rolich Department of Clothing Technology, Faculty of Textile Technology, University of Zagreb, Zagreb, Croatia
Evolutionary algorithms aided textile design 295
Keywords Algorithms, Textiles, Design Abstract Fabric design has always been a matter of inspiration, depending mostly upon man’s ingenuity and creativity. The authors suggest that creation of fabric patterns using evolution algorithms, which can not only help in improving design created by man, but also make the procedure semi-automatic, meaning much less dependent upon the designer himself/herself. Evolution algorithm based software offers a wide range of fabric patterns and is also able to create new ones based on the user’s choice. The procedure can also be of considerable help to professional designers, as it can offer patterns they would not or could not create themselves. The system described is inexpensive and can be used on IBM compatible personal computers. It is user-friendly and can be implemented with no previous preparation on the part of the user. The results of preliminary investigations suggest a practical applicability of the software.
1. Introduction The process of designing aesthetically acceptable fabrics is a creative work, entirely dependent upon the abilities of the pattern designer. Although there are certain rules to help in the process and even formal education to make it easier, it is still primarily a creative artistic work. The designer is not often able to describe the manner in which he made a successful pattern, nor can he/she be sure to be able to do it in the next try. Although there are some computer software packages intended to aid the process of designing fabric patterns, they are of no help in the key, creative step of the process (Weaving Software, 2002). These packages most often help the designer to select the colour of yarns to be used, the type of weave and can offer visual presentation of the future fabric. They cannot offer help in matching the weave and colours of the fabric to be produced. This stage entirely depends upon the abilities and talent of the designer. The only software known to the authors that can help in the creative stage of the process is based on the expert knowledge of skillful designers (Weavemaker, 2002). As opposed to the above, the software described here is not based on any previous knowledge of aesthetic merits of fabrics. The authors wish to help in the creative stage of the process, and have developed computer software that, using evolution algorithm, can assist in creating aesthetically acceptable fabric pattern. The aim of doing this was two-fold. On one hand, it was to help the designer to push him/her into new areas of creation, and help him/her in the creative phase of the designing process. On the other, it was to give assistance to the users who have neither
International Journal of Clothing Science and Technology Vol. 15 No. 3/4, 2003 pp. 295-304 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310478404
IJCST 15,3/4
296
knowledge nor abilities necessary to design a fabric pattern in creating an aesthetically acceptable fabric pattern. Basic mechanism in creating fabric patterns described here is evolution algorithm. Various fabric patterns are the individuals, which try to “survive” the iterative procedure in an “environment” representing aesthetic appeal of the user. Analogous to natural evolution, the more a pattern appeals to the user, the better it is adapted to the environment and higher its chance to survive. The process of mutation is used to alter pattern characteristics, while selection eliminates the samples that do not appeal to the user. Repeating the procedure results in patterns that are well adapted to their environment, meaning appealing to the user. 2. Evolution algorithms Evolution algorithms are the procedures of optimising, learning and modelling based on the principles of natural evolution. These formal systems tend to be isomorphic with natural evolution. Evolution algorithms have been created for two purposes: to understand natural evolution better and to help to apply the principles of natural evolution for solving various tasks. They have been created on the basis of the idea to apply biological principles in solving technical problems using computers. The first idea of applying the principles of natural evolution in solving technical problems appeared in 1950s. Since the computers of that time were rather limited in their abilities, they did not attract much attention. Three independent evolution algorithms were developed in 1960s: evolutionary programming (EP) developed by L. J. Fogel, Owens and Walsh, evolution strategies (ES) developed by I. Rechenberg and H.-P. Schwefel and genetic algorithms (GA) developed by J. H. Holand (Ba¨ck, 1996; Fogel, 1998; Fogel et al., 1966; Holland, 1975; Koza, 1994). Evolution algorithms are robust, relative resistant to interference, able to search wide and multidimensional spaces and are relatively simple to use (Cercone and McCalla, 1994; Grundler et al., 1999). A variant of evolution algorithm is used in the investigations described here, known as evolution strategy (Grundler et al., 1999). ES are iterative procedures starting with initial population. New individuals (offspring) are created from the initial population, employing variation, while the best of them are selected to be the parents of the following generation (iteration). The procedures of variation and selection are repeated until the criterion for completing the algorithm is fulfilled (Bogunovic´ and Rolich, 2000; Grundler, 1996; Rolich, 2001; Rolich and Grundler, 2000). 3. Fabric patterning using weave and yarn colour A number of factors influence the appearance of a fabric: the manner of interlacing yarns (weave), yarn colour, yarn structure (e.g. fineness, twist direction, raw material content, construction and finishing) etc. This paper deals with the impact of yarn colour on fabric appearance only. Woven fabric is
a two-dimensional textile product, formed by perpendicularly interlacing two Evolutionary sets of yarns (Franulic´ Sˇaric´, 2000; Kovacˇevic´ and Franulic´ Sˇaric´, 1999). The set algorithms aided of yarn following the length of fabric is called warp, while the set perpendicular textile design to it is called weft. These two sets of yarns are crossed in the process of weaving on a weaving machine. Various effects and patterns on a fabric can be achieved by combining different yarn colours in both yarn systems, or in only 297 one, as well as by using different types of interlacing warp and weft. Patterning possibilities are more diversified if both warp and weft are multicolour, if more colours are used and if colour repeat is longer, although it is enough to have a single multicolour system of yarns to get weave and colour matching. The paper presented describes the situation in which both yarn systems can be multi-colour. 4. Software for creating fabric patterns based on evolution strategy From the point of view of the user, the software for creating fabric samples based on the evolution strategy should be as simple as possible, and the user should not be distracted by the evolution strategy parameters that he/she does not understand. This is why the software is designed so that the user gets only the information that is relevant for him and he is supposed to answer only to quite simple questions that are easy to understand. The following windows are opened after starting the software. . A window showing the selection of weaves and asking the user to select one of the offers. The user can select one of the three weaves offered, with the weave unit of 16 £ 16: . A window showing ranges of colours, asking the user to select one of the ranges to be used in selecting the colour of warp and weft. The user is offered the selection of 12 ranges, each consisting of 64 different colours. . A window asking the user to state the number of colours to be used for warp and the number to be used for weft. . A window with nine fabric patterns drawn using the weave selected and different colours of warp and weft. It is the initial population of randomly created patterns, based on the predetermined starting parameters. The user is expected to select three patterns which he prefers. Six new patterns are created on the basis of his choice, resulting in a population of nine patterns (three preferred from the previous generation and six newly created). The user should again select three patterns which he prefers most. The procedure is repeated for a predetermined number of times (five is the number selected in this paper). Each newly created set of patterns is based on the user’s selection and applied evolution algorithm. . When the procedure is completed, the user is offered a window with three fabric samples selected as his last choice. Fabric samples are represented in a few forms, to give the user as vivid a presentation of the fabric as
IJCST 15,3/4
possible, while the data on the weave and colours for warp and weft are stored on the hard disc. 4.1 Colour range Colour range is represented by a single-row matrix of the following form:
298
P ¼ ½1 2 . . . pmax
ð1Þ
where pmax is the biggest possible number of colours in the range. Each number represents a single colour. A range consisting of 64 colours is used in this paper ðpmax ¼ 64Þ: 4.2 Sequence of colours in the warp It has already been said that the user can select the number of colours from the range offered that he wishes in warp. If ca is the number of different colours in the warp, then the set Pa a subset of the set P is created containing the ca number of randomly selected elements of the set P : P a ¼ ½d1 d2 . . . d ca
ð2Þ
where P a , P is the subset of the set P ; di [ P is the randomly selected colours from the range (out of the set P ); i ¼ 1; 2; . . .; ca is the matrix element index (the designation of randomly selected colours to be used for creating the colours of the warp); 1 # ca # n is the number of different colours in the warp; n is the size of weave unit warp wise and for all the samples n ¼ 16: Sequence of colours in the warp is represented by a single-row matrix of the following form: A ¼ ½a1 a2 . . . an
ð3Þ
where aj [ P a is the colour of warp yarn from the subset Pa of the colour range; j ¼ 1; 2; . . .; n is the matrix element index (the designation of the ordinal number of the warp yarn); n is the size of weave unit warp wise and for all the samples n ¼ 16: Sequence of colours in the weft is created in the same manner. 4.3 Weave A matrix in the following form represents the weave: 3 2 vm1 vm2 . . . vmn 6 . .. 7 .. .. 7 6 . 6 . . 7 . . 7 6 V ¼6 7 6 v21 v22 . . . v2n 7 5 4 v11 v12 . . . v1n
ð4Þ
where vij [ {0; 1} is the warp crossing point ðvij ¼ 1Þ or weft crossing point Evolutionary ðvij ¼ 0Þ in the weave; i ¼ 1; 2; . . .; m is the matrix index denoting the ordinal algorithms aided number of the weft yarn in the weave; j ¼ 1; 2; . . .; n is the matrix index textile design denoting the ordinal number of the warp yarn in the weave. Unusual indexing noted in this case should be matched to the conventional way of marking warp and weft in the weaving industry. In the investigations described, the weaves 299 with m ¼ 16 and n ¼ 16 are used, meaning that the weave is always represented by a 16 £ 16 matrix. 4.4 Individual A matrix in the following form represents individual: 2
km1 6 . 6 . 6 . K¼6 6 6 k21 4 k11
km2 .. .
... .. .
k22
...
k12
...
3 kmn .. 7 7 . 7 7 7 k2n 7 5 k1n
ð5Þ
where ( kij ¼
aj
if
vij ¼ 1
bi
if
vij ¼ 0
;
i ¼ 1; 2; . . .; m
is the matrix index denoting the ordinal number of the weft yarn in the weave; j ¼ 1; 2; . . .; n is the matrix index denoting the ordinal number of the warp yarn in the weave. The matrix K is a single individual of the initial population. All the other individuals of the population are created in the same manner. 4.5 Population Evolution strategy (m+l)2 ES is applied where m represents the number of parents, while l represents the number of offspring. This form of the evolution strategy considers both parents and offspring as possible parents of the next generation. The evolution strategy (3+6) 2 ES is used in this paper, meaning that the population consists of 3 parents and 6 offsprings, i.e. 9 individuals. Initial population is created randomly, employing even distribution. Each subsequent generation is created so that 3 individuals are selected from the previous population (parents of the new population) and 6 new individuals are created from the three, employing the process of variation described in the following chapter (offspring of the new population). A population of 9 individuals is thus created, from which 3 most appealing are again selected and the procedure is repeated until the completion of the algorithm.
IJCST 15,3/4
4.6 Mutation In the evolution strategy applied here new individual is created by mutation of the existing individual genotype (Back et al., 1997). Mutation results in changing the colour of warp and weft yarn. For example, the warp of a new individual is created by the following procedure of mutation:
300
A0 ¼ A þ M a
ð6Þ
where A0 ¼ ½a01 a02 . . . a0n is the newly created individual (offspring); A ¼ ½a1 a2 . . . an is the individual from which the new one is created (parent); M a ¼ ½m1 m2 . . . mca is the vector of random numbers, the creation of which is described later. To fulfill the condition that yarns of the same colour remain such, i.e. the yarns that were of the same colour before mutation remain as such after mutation, the additional condition for the change of individual genotype should be fulfilled: a0j ¼ aj þ m1
if
aj ¼ d1
a0j ¼ aj þ m2
if
aj ¼ d2 ð7Þ
.. . a0j ¼ aj þ mca
if
aj ¼ d ca
where j ¼ 1; 2; . . .; n is the matrix index denoting the ordinal number of the warp yarn; n is the size of weave unit warp wise ðn ¼ 16Þ: Mutated weft is created in the same manner. The colours are represented by whole numbers in the range from 1 to pmax, and adding a random whole number can represent colour change. Random numbers are generated according to a normal distribution, following the customary procedure for the evolution algorithm (Back et al., 1997; Rolich, 2001). The parameter s of the normal distribution (standard deviation) is changed in the course of the algorithm according to the expression:
sg ¼ smax 2
smax 2 smin ðG 2 1Þ g21
ð8Þ
where sg is the step size in a generation (iteration) g; smax is the step size at the beginning of the algorithm; smin is the step size at the completion of the algorithm; g ¼ 1; 2; . . .; G is generation (iteration); G is the total number of generations. Varying step size is selected with the aim to obtain significant changes of colour at the beginning of the algorithm, to be gradually reduced towards its completion. The idea is to offer the user quite diversified patterns at the beginning, so that he/she could select one that he/she likes most. We can suppose that towards the end, the user has selected patterns of his liking, so
colour changes can be subtle and the pattern is being adapted to the user’s taste Evolutionary (fine tuning of colours in the yarns that constitute the pattern). It is necessary to algorithms aided avoid the situation in which adding a random number would result in values textile design higher than pmax or lower than 1. Consequently, colours are changed according to the following expression for warps: a0j ¼ a0j þ pmax
if
a0j , 1
a0j ¼ a0j 2 pmax
if
a0j . pmax
301 ð9Þ
where j ¼ 1; 2; . . .; n is the matrix index denoting the ordinal number of the warp yarn. The procedure is identical for wefts. A circle in which the colours from 1 to pmax are arranged can represent the process. Mutation means a shift on the circle in one or the other direction for a particular randomly selected whole number. 4.7 Selection Selection is the procedure of choosing individuals that will take part in creating new individuals (selecting the parents). The criterion is the adaptation of the individual to its environment and is expressed as fitness. There is no mathematical fitness in the case described (fitness is not defined by a function, number or in any other exact way). It depends on the user’s choice of the individual based on a personal aesthetic criterion. In each generation of nine individuals, the user selects three patterns which he likes most and they become the parents of the next generation. The algorithm applied is an elitist one, i.e. the individual with highest fitness always survives. It should be noted that the environment, i.e. aesthetic impression of the user, varies and depends upon some psychological factors (user’s mood, the influence of the other people around him etc.). It is quite possible that in the course of the algorithm, the user changes his mind while comparing two identical patterns, changing his preferences to the one he refused in the previous selection. 4.8 Condition for completing the algorithm The condition for completing the algorithm is the number of generations (iterations), for the purposes of this investigation five generations are considered. It means that the procedure is stopped after five generations (five iterations). After five generations the user finds it tedious and looses interest in the process. For a smaller number of steps (iterations), the patterns are still too diversified and the one that the user would feel attractive enough cannot be created. The software for automatic matching weave and colour is written in the MATLAB program ver. 6 R12 (MATLAB and Simulink, 2002).
IJCST 15,3/4
302
5. Checking the practical applicability of the software To test the practical applicability of the software, it was given to the users without any previous explanation. After each of the users used the software for five times, they were asked to answer anonymously to a questionnaire. Testing was implemented on 23 students from the Faculty of Textile Technology, University of Zagreb, Croatia, who can be considered as user, at least partially skilled in designing fabrics. The students were selected from all the years of study, from freshmen to senior, and from all the departments and branches. Each was given ten questions to answer. The first was concerned with the number of patterns presented to the user (size of the population). Seventy-one per cent of the users answered that the size of the population was adequate, 22 per cent considered the size too small, while 9 per cent thought it was too big. It can be concluded that it would be reasonable to increase the number of individuals (patterns) in the population, but not too much. To the question about the number of patterns selected as most appealing (three patterns in the case investigated), the users answered that the number is adequate (74 per cent), too big (22 per cent) or too small (4 per cent). It can be concluded that it would be reasonable to ask the user to select only one or two patterns. Regarding the question on pattern diversity, the users answered as follows: diversity is too small, i.e. the patterns are too much alike (43 per cent) and it is adequate (35 per cent) or too big (22 per cent). Different tastes of the users can be seen here, thus the assessments of diversity are rather uniform. The user himself can influence pattern diversity, if he selects a number of colours for warp and weft. This is why these answers cannot be used as a sound basis for altering the software. Next question concerned pattern appeal, as obtained by the end of the procedure. The users were required to grade pattern appeal using grades from 1 (worst) to 5 (best). Grading distribution is as follows: 1-0, 2-0, 3-30, 4-48 and 5-22 per cent. It is quite a promising result, since even at this stage it can be stated that the users are satisfied with the results obtained. The next question was about usability of the programme concerning pattern designers, and the answers were as follows: it would be of no use (0 per cent), it could help partially (30 per cent), it could help quite a lot (48 per cent), and it is essential for creating new patterns (22 per cent). It can be concluded that the users consider the software rather useful in creating new patterns. Asked about the number of iterations, the users answered as follows: the number of steps should be more than five (13 per cent), should be less than five (35 per cent), and five is the exact number (52 per cent). It is obvious that most users consider more iteration unnecessary, probably due to fatigue, loss of interest and concentration. The last question concerned the appeal of the software itself. The users answered as follows: interesting (91 per cent), funny (35 per cent), boring (0 per cent) and tiring (4 per cent). The users could select more than one answer here, thus the total percentage is higher than 100 per cent. This result is of special importance, as it is essential that the software is not psychologically
tiring to the user. It should be supposed that the persons who found Evolutionary the software tiring were the same who considered the number of iterations algorithms aided too big. textile design 6. Conclusion and suggestions for further work A few facts should be taken into account when drawing conclusions. First of all, it should be noted that only the basic idea is investigated here, thus a number of software improvements and additional possibilities were omitted on purpose. The purpose was to find out whether the idea was soundly based and whether it was reasonable to continue with its development. The other fact worth mentioning is that the groups of testing subjects were relatively small, came from the same institution and of approximately the same age (19-26 years of age). Both of these factors impart certain insecurity in reaching proper conclusions. However, general applicability of the idea can be assessed nevertheless. This investigation of the software applicability offers the following conclusions: . the software proposed can be of help in creating new fabric patterns, . the software is interesting and amusing for the users, which is important for its acceptability, and . the suggestions given by the users are not of such nature as to make the idea impractical and valueless. The idea is soundly based and it is quite reasonable to continue developing the system, accepting the suggestions and organizing further investigations under in-plant conditions. References Ba¨ck, T. (1996), Evolutionary Algorithms in Theory and Practice: Evolution Strategies, Evolutionary Programming, Genetic Algorithms, Oxford University Press, Oxford. Ba¨ck, T. et al., (1997), “Evolutionary computation: comments on the history and current state”, IEEE Transactions on Evolutionary Computation, Vol. 1 No. 1, pp. 3-17, 1089-778X. Bogunovic´, N. and Rolich, T. (2000), “An analysis of recombination techniques in evolution programs”, MIPRO 2000, 23rd International Convention, May 2000, Opatija, Croatia, pp. 1-4. Cercone, N. and McCalla, G. (1994), “The years of computational intelligence”, Computational Intelligence, Vol. 10 No. 4, pp. i-vi, ISSN 0824-7935. Fogel, D. (1998), Evolutionary Computation: The Fossil Record, IEEE Press, New York. Fogel, L.J. et al. (1966), Artificial Intelligence Through Simulated Evolution, Wiley Publishing, New York. Franulic´ Sˇaric´, D., (2000), Optimiranje tehnolosˇkih parametara tkanine primjenom racˇunala, magistarski rad, Sveucˇilisˇte u Zagrebu, Tekstilno tehnolosˇki fakultet. Grundler, D. (1996), Geneticˇkim algoritmom optimirano neizrazito visˇerazinsko vodenje procesa, doktorska disertacija, Sveucˇilisˇte u Zagrebu, Fakultet elektrotehnike i racˇunarstva.
303
IJCST 15,3/4
304
Grundler, D. and Rolich, T. (1999), “Primjena inteligentnih algoritama u odjevnom i tekstilnom inzˇenjerstvu,” Tekstil – cˇasopis za tekstilnu tehnologiju i konfekciju, Vol. 48 No. 7, pp. 331-8, ISSN 0492-5882. Holland, J.H. (1975), Adoption in Natural and Artificial Systems, The University of Michigan Press, Ann Arbor, USA. Kovacˇevic´, S. and Franulic´ Sˇaric´, D. (1999), “Efekti na tkanini nastali skladom veza i boja”, Tekstil – cˇasopis za tekstilnu tehnologiju i konfekciju, Vol. 48 No. 7, pp. 339-42, ISSN 04925882. Koza, J.R. (1994), Genetic Programming II: Automatic Discovery of Reusable Programs, MIT Press, Cambridge. MATLAB and Simulink, The MathWorks, Inc., 3 Apple Hill Drive, Natick, MA 01760-2098, United States, E-mail:
[email protected], http://wwwmathworks.com Accessed: 2002-04-15 Rolich, T. (2001), Vrednovanje primjene evolucijskog algoritma pri ostvarenju optimalnog vodenja, magistarski rad, Sveucˇilisˇte u Zagrebu, Fakultet elektrotehnike i racˇunarstva. Rolich, T. and Grundler, D. (2000), “Efficiency of recombination operator in evolution strategies”, The 11th International DAAAM Symposium on Intelligent Manufacturing and Automation: Man-Machine-Nature, October 2000, Opatija, Croatia, pp. 407-8. Weavemaker, Designer Software LLC, PO Box 6351, Syracuse, NY 13217-6351, United States, E-mail:
[email protected], http://wwwweavemaker.com/ Accessed: 2002-04-15 Weaving software – a survey, http://www.geocities.com/Eureka/4613/engindex.html Accessed: 2002-04-15
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
IJCST 15,5
320
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
A study on clothing pressure for men’s suit comfort evaluation Chen Dongsheng
Received January 2003 Accepted May 2003
Changchun University of Technology, Changchun, People’s Republic of China
Zhao Qing Tottori University, Tottori, Japan Keywords Clothing, Pressure, Men Abstract In order to pursue men’s suit wear comfort, the basic data on an accurate men’s suit comfort analysis with clothing pressure is required. Therefore, in such investigation, it is difficult to find out ideal persons as test subjects. In the measuring experiments, we used dummies designed for resuscitation practice to obtain the clothing pressure with both normal standing posture and movement patterns and compared these data with subjects’ to give the relationship of clothing pressure between the dummy and subject. The correlation between the measurements with dummies and the subjects’ are shown. It turns out that the dummy D1 and D6 are mainly as intended in pressure measurement with normal standing posture. But the dummy D2 and D3 are not. It turns out that the dummy D1 can imitate the human shoulder’s movement patterns well, and the dummy D4 and D6 are mainly as intended. But the dummy D2 is not. It turns out that all the dummies have a certain limitation to be placed in clothing pressure measurements. It also shows that instead of subject to use dummy to investigate clothing pressure with both normal standing posture and movement pattern, not only dummy’s features such as compression hardness, form and size, but measuring postures are also needed to take into consideration.
International Journal of Clothing Science and Technology Vol. 15 No. 5, 2003 pp. 320-334 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310492598
1. Introduction The men’s suit, shirt, and necktie have been established as fundamental clothes for salaried worker, and about 90 per cent or more of the adult man wear suit as business wear. Among those, ready-made suit user is about 81 per cent (Yuko and Haruko, 1998). It is necessary to take into consideration the relation between the form of men’s suit and of the upper body, and the relation between the amount of room of men’s suit and the clothing pressure furthermore for the comfortable men’s suit manufacturing. In the earlier researches on body shape of adult man, there are the study on the trunk part given by Hasebe et al. (1976) and Nobuko and Yayoi (1993), on the whole body can be found in Ume and Hasebe (1980, 1982), and on the neck or the shoulder part developed by Haruko and Sachiko (1983) and Sachiko and Haruko (1983). But these researches remained in the examination of body The authors wish to thank all the volunteers who participated in this study and students at Tottori University for their assistance in the measuring experiments.
shape’s characteristics. In the research on the design of men’s suit, Masako and A study on Kimiko (1985), Masako et al. (1984, 1985) showed the measurement value of clothing pressure overfeed and reported that the mechanical properties have the suitable value for good appearance in the clothing design. Moreover, Yuko and Haruko (1998a, b) reported the relation between the body characteristics of the standard upper body shape of adult man and the suit pattern, and the conformity of the 321 actual situation. In order to pursue comfortable men’s suit, we aim at the establishment of the index evaluation of clothing pressure. By the wearing experiment with subject and trial dummy, the clothing pressure values measured with normal standing posture and the posture flexing upper limbs 908 were gathered. To relate the pressure from trial dummy experiments with which from the subjects, and to obtain the base data for the clothing pressure evaluation of men’s suit, Grey related analysis theory (Deng, 1987; Sifeng and Yi, 1998; Wang, 1993) was used. In this paper, we investigate and clarify the relation between the subjects and the trial dummies. 2. Theory Grey system theory was proposed by Deng (1982) of China, and it is broadly applied also in Japan and China (Deng, 1993; Dongsheng and Zhao, 1996; Dongsheng et al., 1997; Momoko and Mitsuo, 2000; Ruquan and Mitsuo, 1998). Grey systems theory mainly studies problems with uncertainty due to the small samples involved. It is obviously different from the statistical methods such as the regression analysis, variance analysis and principal component analysis which have the pitfalls that a large amount of data are required, otherwise it would be difficult to draw statistical conclusions with reasonable confidence. Grey incidence analysis can be applied to the cases of various sample sizes and distribution patterns with small amount of computation that applying statistical methods can hardly achieve much useful conclusions. The fundamental idea of the grey incidence analysis is that the closeness of a relation is judged based on the similarity level of the geometrical patterns. Based on the change degree or the tendency between the system characteristic behaviours and related factors, some elementary analysis can be conducted. The degree of incidence is called the degree of grey incidence. Aiming at the ordering of the degrees of incidences between the system characteristic behaviours and each relevant factor’s behavioural sequence, the comparison and evaluation of system factors’ behaviours become possible. Denote the set of system characteristic behaviours as Y 1 ðY 1 [ Y Þ; and the relevant factors denoted as X i ðX i [ XÞ: That is to say, Y1is a sequence of the system characteristic behaviours Y 1 ¼ {Y 1 ð1Þ; Y 1 ð2Þ; . . .; Y 1 ðkÞ; . . .; Y 1 ðN Þ} and
ð1Þ
IJCST 15,5
X 1 ¼ {X 1 ð1Þ; X 1 ð2Þ; . . .; X 1 ðkÞ; . . .; X 1 ðN Þ} ...
...
X i ¼ {X i ð1Þ; X i ð2Þ; . . .; X i ðkÞ; . . .; X i ðN Þ} ...
322
ð2Þ
...
X m ¼ {X m ð1Þ; X m ð2Þ; . . .; X m ðkÞ; . . .; X m ðN Þ} are the sequences of relevant factors. With i ¼ 1; 2; . . .; m; k ¼ 1; 2; . . .; N ; k ^ 3; M is the data number of relevant factors, N stands for the number of data and k is the data order. Based on the grey incidence analysis, the computation of the degree of incidence coefficient j1i can be accomplished with the geometrical difference between the system characteristic behaviours and the relevant factors. Denoting j1i (k) as the incidence coefficient of the kth relevant factor XI with respect to the system characteristic behaviour Y1, then min minjY 1 ðkÞ 2 X i ðKÞjh max maxjY 1 ðkÞ 2 X i ðkÞj
j1i ðkÞ ¼
i
k
i
k
jY 1 ðkÞ 2 X i ðkÞj þ h max maxjY 1 ðKÞ 2 X i ðkÞj i
ð3Þ
k
where min minjY 1 ðkÞ 2 X i ðkÞj i
k
and max maxjY 1 ðkÞ 2 X i ðKÞj i
k
stand for the minimum and maximum differences. h [ ½0:5; 1 is called the distinguishing coefficient. In general, we can take h ¼ 0:5 when we are more interested in the relationship between the system characteristic behaviour Y1 and the relevant factors. Denoting g1i as the degree of grey incidence of system characteristic behaviour Y1 to the relevant factor Xi, then
g1i ¼
n 1X j1i ðkÞ n k¼1
ð4Þ
That is to say, the degree of grey incidence g1i {g1i } ¼ {g11 ; g12 ; . . .; g1i ; . . .; g1m }
ð5Þ
and the grey incidence analysis can be accomplished based on the value of the A study on degree of grey incidence g11, g12 ; . . .; g1m : clothing pressure If there are more than one system characteristic behaviours, denoted as Y j ðY j [ Y ; j ¼ 1; 2; . . .; nÞ; assume that X i ðX i [ X; i ¼ 1; 2; . . .; mÞ; then the degree of grey incidence can be rewritten as 8 9 g11 g12 · · · g1m > 323 > > > > > > > > > > < g21 g22 · · · g2m > = {gji } ¼ ð6Þ .. .. .. > > . · · · . . > > > > > > > > > : gn1 gn2 · · · gnm > ; and with this matrix the incidence analysis between Yj and Xi can be conducted. 3. Experimental 3.1 Suits for the experiments The men’s suits used in the experiments are fixed to the mean shape finger which is up to the JIS standard. Five kinds of materials were used in the experiments, and the sizes of suit are based on the type 92A5 (chest 92 cm, waist 80 cm and height 170 cm), the smaller sizes are up to the type 92A4 and the larger sizes are in conformity with the type 92A6. The wearing experiments were totally carried out in 15 ways. Considering the discrepancies caused by design or the arrangement during the suit manufacturing process, the manufacturer made a united effort to keep the suit patterns almost identical from a technical standpoint. The composition fibers and the sizes of men’s suit used for the experiments are shown in Tables I and II. 3.2 Subjects Ten subjects aged from 20 to 35 who usually wear the suit with the size matching the type 92A5 were selected. The body characteristics of the subject’s upper half are measured in accordance with the measuring method of JIS. After the subject’s Rohrer index was computed, respectively, those subjects who are neither fat nor thin with the good Rohrer index from 125 to 135 were chosen as the further experiments included chest, waist, and height. Finally, the wearing experiments were conducted through the chosen three best subjects. The characteristics of the three subjects’ body are shown in Table III. 3.3 Test dummy To evaluate the clothing pressure of men’s suit, instead of the subjects, the six kinds (D1-D6) of movable body dummies with the movable upper-limb parts approximated to 92A5 were used to measure the clothing pressure. Among those, the dummy D1, D4, and D5 have the same upper-limb parts, the
IJCST 15,5
324
Thickness (mm)
Weight (g/m2)
33.0
0.378
151.0
28.8
24.4
0.580
191.3
Wool 100 per cent
31.3
26.8
0.499
200.9
Wool 100 per cent
41.3
25.3
0.785
251.7
Polyurethane 55 per cent Wool 100 per cent
28.0
22.5
0.347
133.7
Fibers
No. 1-S M L
Wool 100 per cent
38.1
No. 2-S M L
Wool 100 per cent
No. 3-S M L No. 4-S M L* No. 5-S M Table I. Details of men’s suit
Yarn density (ends/cm) (picks/cm)
Samples
L* Notes: S: 90A4; M: 92A5; L: 94A6; L*: 95A6
Circumference Shoulder Dress length Circumference of abdominal length (cm) (cm) of chest (cm) No. Samples (cm)
Right sleeve length (cm)
Weight (g)
1
S M L
72.5 74.5 76.5
104.0 105.0 110.0
95.0 96.4 99.0
45.8 46.5 47.5
57.5 59.0 60.5
57.5 59.0 60.5
537.2 551.5 559.9
2
S M L
72.5 74.0 76.5
105.0 106.0 109.0
94.0 96.0 98.0
45.5 46.2 46.8
57.5 59.0 60.7
57.5 59.0 60.5
611.6 624.3 642.6
3
S M L
72.5 74.0 76.5
104.0 106.0 108.0
94.6 96.2 98.0
46.0 46.5 48.0
57.5 59.0 60.5
57.5 59.0 60.5
621.7 637.1 656.3
4
S M L*
71.0 72.5 75.0
106.8 108.0 110.0
96.3 98.2 100.8
45.8 46.0 47.7
57.2 57.0 60.2
57.0 58.2 59.7
738.6 760.8 791.9
105.2 96.0 109.4 98.5 110.0 101.1 L: 94A6; L*: 95A6
45.5 46.2 47.3
56.7 58.3 60.2
56.7 58.3 60.2
551.8 563.4 580.5
5
Table II. Dimensions of men’s suit
Left sleeve length (cm)
S 71.0 M 73.0 L* 75.2 Notes: S: 90A4; M: 92A5;
Parameters
Sub. 1
Sub. 2
Sub. 3
Mean
SD
Height (cm) 171.0 170.0 174.0 171.67 Weight (kg) 67.0 62.0 67.0 65.33 Chest circumference (cm) 93.2 89.0 90.0 90.73 Abdominal circumference (cm) 83.5 76.0 80.0 79.83 Buttock circumference (cm) 96.0 94.0 96.0 95.33 Neckbase circumference (cm) 40.0 37.5 39.5 39.00 Armscye circumference (cm) 40.0 38.0 42.5 40.17 Upper arm circumference (cm) 29.0 17.0 28.0 24.67 Waist back length (cm) 50.0 42.0 47.0 46.33 Sleeve length (cm) 58.5 55.0 58.0 57.17 Yukia (cm) 74.0 67.0 68.0 69.67 Bideltoid breadth (cm) 16.5 15.0 15.0 15.50 Biacromial arc (cm) 44.0 41.0 44.0 43.00 Interscye (cm) 41.0 36.0 37.0 38.00 Anterior chest arc (cm) 41.0 37.0 38.0 38.67 Chest breadth (cm) 34.0 30.5 31.1 31.87 Chest depth (cm) 21.0 20.5 22.6 21.37 Waist breadth (cm) 24.9 27.1 26.6 26.20 Waist depth (cm) 21.4 18.5 20.7 20.20 Shoulder slope (right) 21.0 20.0 24.0 21.67 Shoulder slope (left) 21.0 21.0 24.0 22.00 Rohrer indexb 134.0 126.2 127.2 129.12 Notes: aLength from cervicale to stylion ulnare, and bweight (kg) 107/height (cm)3
2.08 2.89 2.19 3.75 1.15 1.32 2.25 6.66 4.04 1.89 3.79 0.87 1.73 2.65 2.08 1.87 1.10 1.15 1.51 2.08 1.73 4.25
upper-limb parts of the dummy D2, D3, and D6 were made up of the same materials, and the body of the dummy D2 and D3 were developed by the difference compression hardness materials from the others. The compression hardness for trial experiment is based on the results of the compression characteristics of upper limb studied by Noriko et al. (1995), and the compression hardness was measured by the compression measuring equipment which was developed by Noriko et al. (1985). Compressed the meter to 9.8 kPa (100 gf/cm2) with the compressing speed of 10 mm/min, and then reduced the pressure to the same speed. To investigate how clothing pressure may serve as an index in evaluating the clothing comfort, the size of the dummy D2 and D3 were designed a little big. The sizes of dummy are shown in Table IV, the composition materials are shown in Table V, and the compression characteristics are shown in Figure 1. 3.4 Wearing measurements The wearing experiments were done on three chosen subjects who have the good Rohrer index from 125 to 135 and six kinds of body dummies with 15 kinds of suit jackets at random. Clothing pressures were measured on shoulder,
A study on clothing pressure
325
Table III. Characteristics of the subjects (N ¼ 3)
IJCST 15,5
326
Table IV. Characteristics of the dummies
Parameters
D1
D2
D3
D4
D5
D6
Chest circumference (cm) Abdominal circumference (cm) Neckbase circumference (cm) Armscye circumference (cm) Upper arm circumference (cm) Waist back length (cm) Bideltoid breadth (cm) Biacromial arc (cm) Interscye (cm) Anterior chest arc (cm) Chest breadth (cm) Chest depth (cm) Waist breadth (cm) Waist depth (cm) Shoulder slope (right,8)
93.5 84.0 40.0 43.0 28.2 38.5 15.0 44.0 40.0 37.0 32.2 21.7 28.3 21.5 21.0
99.0 84.0 42.0 43.0 30.5 40.0 15.0 44.5 41.0 40.0 34.0 23.5 28.5 22.0 19.0
99.0 84.0 42.0 43.0 30.5 40.0 15.0 44.5 41.0 40.0 34.0 23.3 28.5 22.0 19.0
92.0 80.0 41.0 43.0 28.2 41.5 15.0 45.0 41.0 38.5 34.5 22.3 27.0 20.4 22.0
92.0 80.0 40.0 43.0 28.2 41.5 15.0 46.0 42.0 39.0 35.0 21.7 27.0 21.0 22.0
92.0 80.0 40.0 43.0 25.2 41.5 15.0 46.0 42.0 39.0 35.0 21.7 27.0 21.0 22.0
D1
Table V. Composition materials of the dummies
D2
D3
D4
D5
D6
Materials Wood, of body Cloth, Others
Thick sponge, Polyurethane, Styrene foam, Wood, Elasticity cloth, Others
Thick sponge, Styrene foam, Wood, Elasticity cloth, Others
Improvement by China marketing, Thin sponge, Styrene foam, Wood, Cloth, Others
Improvement by China marketing, Thin sponge, Styrene foam, Wood, Cloth, Others
Improvement by China marketing, Thin sponge, Styrene foam, Wood, Cloth, Others
Materials Cotton, of arm Polyester, Cloth, Steel rod, Others Mobile
Sponge, Polyurethane, Wood, Cloth, Others Mobile
Sponge, Wood, Cloth, Others Mobile
Cotton, Polyester, Cloth, Steel rod, Others Mobile
Cotton, Polyester, Cloth, Steel rod, Others Mobile
Sponge, Polyurethane, Wood, Cloth, Others Mobile
front armscye, scapula, back armscye, and upper arm front five parts with normal standing posture and the posture flexing upper limbs 908. The wearing measurements of these five parts were based on the results of the subjective evaluation by the subjects in preliminary experiments (Mika et al., 2001), and the situations of the wearing measurements with the dummies are shown in Figure 2. The clothing pressure was measured with an air-pack type contact surface pressure sensor (AMI Techno. LTD) with which even about 0.03 kPa low pressure can also be obtained by amplifying the output voltage. To gather more accurate pressure data of each part, the meter was read when the pin
A study on clothing pressure
327
Figure 1. Compressive properties of the dummy and human body. Dummy: D1-D6; human: subject S
Figure 2. Wear measurement by the dummy
IJCST 15,5
328
became stable each time, and the measuring operation was repeated thrice, and then the average was taken. The diameter of the air bag was set at 3 cm, about 0.8 ml air was enclosed, and the thickness of the air bag was kept at 1 mm. In addition, the experiments were carried out at room temperature 18 ^ 18C and relative humidity 55 ^ 10 per cent. 4. Results and discussion In this study, for the sake of the number limitation of the subject and dummy in the wearing experiments, the data of the clothing pressure were limited to a range because of the small number of experimental data. The men’s suit, subject, and body dummy in the wearing experiment can be considered as one of the grey system, and it is thought that the relevance of the clothing pressure between the subjects and the dummies measured with normal standing posture and the posture flexing upper limbs 908 can be achieved effectively with grey incidence analysis. 4.1 Consideration of the relevance with normal standing posture First, the mean value of the clothing pressure on the shoulder parts with normal standing posture gathered from the three subjects was denoted as the characteristic behaviours Y1, and denoted the clothing pressure value on the shoulder parts gathered from the dummy D1-D6 with normal standing postures as the relevant factors X i ðX i [ X; i ¼ 1; 2; . . .; 6Þ: From the formula (3), the computation of the grey incidence coefficient of the kth to the relevant factors Xi over the characteristic behaviours Y1 can be accomplished. Furthermore, consider the distinguishing coefficient h ¼ 0:5; from formula (4), the degree of grey incidence to the comparison factor Xi of the clothing pressure value on the shoulder of the subjects with normal standing posture is {g1i } ¼ {g11 ; g12 ; . . .; g1i ; . . .; g1m } ¼ {0:6550; 0:5700; 0:5557; 0:5540; 0:5323; 0:5293}: The clothing pressure value ranking of the grey incidence degree on the shoulder parts of the dummy D1-D6 and the shoulder parts of the subjects with normal standing posture is D6 . D5 . D4 . D1 . D3 . D2: According to this result, it turns out that the degree of grey incidence to the relevant factor of the clothing pressure value on the shoulder parts of the subjects and D6 is the highest, and the degree of grey incidence to the relevant factor and D2 is the lowest. In other words, it is considered that the dummy D6 is best fit for measuring the clothing pressure on shoulder, and the dummy D2 is seldom with normal standing posture in this study. Similarly, denoted the mean clothing pressure value of four parts: front armscye, scapula, back armscye and upper arm front of the three subjects with normal standing posture as the characteristic behaviours Y2, Y3, Y4 and Y5, and denoted the mean clothing pressure value of the four parts: front armscye,
scapula, back armscye and upper arm front of the dummy D1-D6 with normal A study on standing posture as the relevant factors X i ðX i [ X; i ¼ 1; 2; . . .; 6Þ: From clothing pressure formula (3), the computation of the grey incidence coefficient of the kth to the relevant factors Xi over the characteristic behaviours Y2, Y3, Y4, Y5 can be accomplished. Furthermore, consider the distinguishing coefficient h ¼ 0:5; from formula (4), the degree of grey incidence to the comparison factor Xi of the 329 clothing pressure value on the four parts: front armscye, scapula, back armscye and upper arm front of the subjects with normal standing posture are {g2i } ¼ {g21 ; g22 ; . . .; g2i ; . . .; g2m } ¼ {0:9039; 0:8123; 0:4306; 0:9063; 0:8874; 0:9035} {g3i } ¼ {g31 ; g32 ; . . .; g3i ; . . .; g3m } ¼ {0:7619; 0:6752; 0:6648; 0:6070; 0:7187; 0:7260} {g4i } ¼ {g41 ; g42 ; . . .; g4i ; . . .; g4m } ¼ {0:6460; 0:5253; 0:6025; 0:6432; 0:6070; 0:5823} {g5i } ¼ {g51 ; g52 ; . . .; g5i ; . . .; g5m } ¼ {0:9020; 0:8970; 0:9324; 0:9219; 0:9091; 0:8934}: The clothing pressure ranking of the grey incidence degree of the front armscye parts of the dummy D1-D6 and the shoulder parts of the subjects with normal standing posture is D4 . D1 . D6 . D5 . D2 . D3: This result shows that the degree of grey incidence to the relevant factor of the clothing pressure value on the front armscye part of the subjects and D4 is the highest, D1 is fairly high, and D3 is the lowest. In other words, it is considered that the dummy D4, D1 and D6 are suitable for measuring the clothing pressure at the front armscye parts, and the dummy D3 is seldom with normal standing posture in this research. Furthermore, with the degree of grey incidence g3i, g4i, g5i of the clothing pressure values on the scapula, back armscye and upper arm front parts of the subjects and the dummy D1-D6, it is considered that the dummy D1 is suitable for detecting the clothing pressure on scapula and the dummy D4 is seldom regarded as suitable for measuring the clothing pressure with normal standing posture, the dummy D1 is suitable for detecting the clothing pressure on the back armscye parts and the dummy D2 is seldom suitable with normal standing posture, the dummy D3, D4 and D1 are suitable for detecting the clothing pressure on upper arm front and the dummy D6 is seldom with normal standing posture in this research.
IJCST 15,5
330
In short, the results of the degree of grey incidence of clothing pressure from the dummy D1-D6 and the subjects with normal standing posture turn out that the dummy D1 is suitable for detecting the clothing pressure on front armscye, scapula, back armscye and upper arm front parts with normal standing posture, the dummy D3 is suitable for detecting the clothing pressure on the upper arm front parts, and the dummy D4 is suitable for detecting the clothing pressure on the shoulder and the front armscye and the scapula parts. Moreover, from formula (6), it follows that X ½gj1 ¼ 3:7679 X ½gj2 ¼ 3:4391 X ½gj3 ¼ 3:1626 X ½gj4 ¼ 3:6340 X ½gj5 ¼ 3:6921 X ½gj6 ¼ 3:7601 So, we have that X X ½gj1 ¼ . ½gji
ð j ¼ 1; 2; . . .; 5; i ¼ 1; 2; . . .; 6; i – 1Þ
X X ½gj3 ¼ , ½gji
ð j ¼ 1; 2; . . .; 5; i ¼ 1; 2; . . .; 6; i – 3Þ
It turns out that the dummy D1 is most suitable, D6 is relatively suitable, D3 and D2 are seldom suitable for measuring the clothing pressure synthetically on shoulder, front armscye, scapula, back armscye and upper arm front five parts. It is considered that D3 and D2 are seldom suitable for measuring the clothing pressure with suit in which the sizes such as chest are too bigger than the size of standard type 92A5. 4.2 Consideration of the relevance with the posture flexing upper limbs 908 Denoted the mean value of the clothing pressure on the shoulder parts of the three subjects with the posture flexing upper limbs 908 as the characteristic behaviours
ð0Þ ð0Þ Y ð0Þ Y [ Y ; 1 1
and denoted the clothing pressure value on the shoulder parts of the dummy A study on D1-D6 with the posture flexing upper limbs 908 as the relevant factors clothing pressure
ð0Þ X ð0Þ X ð0Þ i i [ X ; i ¼ 1; 2; . . .; 6 : From formula (3), the computation of the grey incidence coefficient to the kth relevant factors X ð0Þ over the characteristic behaviours Y ð0Þ can be 1 i accomplished. Furthermore, consider the distinguishing coefficient h ¼ 0:5; from formula (4), the grey incidence degree of the comparison factor X ð0Þ i to the clothing pressure value on the shoulder of the subjects with normal standing posture is {gð0Þ 1i } ¼ {0:5850; 0:6518; 0:6958; 0:5845; 0:6487; 0:6348} the clothing pressure value ranking of the grey incidence degree on the shoulder parts of the dummy D1-D6 and the subjects with the posture flexing upper limbs 908 is D3 . D2 . D5 . D6 . D4 . D1: According to this result, it shows that the degree of grey incidence between D3 and the relevant factor of the clothing pressure value on the shoulder parts of the subjects is highest, next to it is D2, and the degree of grey incidence to the relevant factor and D1 is lowest. In other words, it is considered that the dummy D3 and D2 are most suitable for detecting the clothing pressure on the shoulder parts, and the dummy D1 is seldom with the posture flexing upper limbs, 908, in this study. Similarly, denoted the mean clothing pressure value of the four parts: the front armscye, scapula, back armscye and upper arm front of the three subjects with the posture flexing upper limbs, 908, as the characteristic ð0Þ ð0Þ ð0Þ behaviours Y ð0Þ 2 , Y 3 , Y 4 and Y 5 and denoted the mean clothing pressure value of the four parts: the front armscye, scapula, back armscye and upper arm front of the dummy D1-D6 with normal standing posture as the relevant factors
ð0Þ ð0Þ X ð0Þ X [ X ; i ¼ 1; 2; . . .; 6 : i i From formula (3), computation of the grey incidence coefficient of the kth ð0Þ ð0Þ ð0Þ ð0Þ relevant factors X ð0Þ i over the characteristic behaviours Y 2 , Y 3 , Y 4 , Y 5 can be accomplished. Furthermore, consider the distinguishing coefficient h ¼ 0:5; from formula (4), the grey incidence degree of the comparison factor X ð0Þ i to the clothing pressure value of the four parts: the front armscye, scapula, back armscye and upper arm front parts of the subjects with the posture flexing upper limbs, 908, is {gð0Þ 2i } ¼ {0:6241; 0:6471; 0:7072; 0:6322; 0:5982; 0:6423}
331
IJCST 15,5
{gð0Þ 3i } ¼ {0:7851; 0:4622; 0:4607; 0:7241; 0:6028; 0:6440} {gð0Þ 4i } ¼ {0:7841; 0:5902; 0:6057; 0:7849; 0:7287; 0:7399}
332
{gð0Þ 5i } ¼ {0:7679; 0:4189; 0:6958; 0:5637; 0:5595; 0:5413} The clothing pressure value ranking of the grey incidence degree of the front armscye part of the dummy D1-D6 and the shoulder part of the subjects with the posture flexing upper limbs 908 is D3 . D2 . D6 . D4 . D1 . D5: According to this result, it shows that the degree of grey incidence to the relevant factor of the clothing pressure value of the front armscye part with the subjects and D3 is the highest, the degree of grey incidence to the relevant factor and D2 is fairly high, and the degree of grey incidence to the relevant factor and D5 is the lowest. In other words, it is considered that the dummy D4, D1 and D6 are suitable for detecting the clothing pressure on the front armscye part, and the dummy D5 is seldom with the posture flexing upper limbs, 908, in this study. Furthermore, it is considered that the dummy D1 is suitable for detecting the clothing pressure at the scapular position and the dummy D3 is seldom suitable for normal standing posture, the dummy D4 is suitable for detecting the clothing pressure of the back armscye and the dummy D2 is seldom with the posture flexing upper limbs 908, the dummy D1 is suitable for detecting the clothing pressure of the upper arm front and the dummy D2 is seldom suitable for detecting the clothing pressure at the scapular position with the posture ð0Þ ð0Þ flexing upper limbs 908 with the degree of grey incidence gð0Þ 3i , g4i , g5i to the clothing pressure value of the scapula, back armscye and upper arm front with the dummy D1-D6 and the scapula, back armscye and upper arm front parts with the subjects in this study. In short, according to the clothing pressure results of the degree of grey incidence of the dummy D1-D6 and the subjects with the posture flexing upper limbs 908, the dummy D1 is suitable for detecting the clothing pressure on the scapula, back armscye and upper arm front parts. The dummy D3 is suitable for detecting the clothing pressure on the shoulder and front armscye parts, and the dummy D4 is suitable for detecting the clothing pressure on the back armscye. Moreover, from formula (6), it shows that Xh
i gð0Þ ¼ 3:5413 j1
Xh
i gð0Þ ¼ 2:7702 j2
Xh
gð0Þ j3
i
¼ 3:1651
Xh
i ¼ 3:2894 gð0Þ j4 Xh ð0Þ i gj5 ¼ 3:1379 Xh ð0Þ i gj6 ¼ 3:2023 So, we have that Xh ð0Þ i Xh ð0Þ i gj1 ¼. gji Xh ð0Þ i Xh ð0Þ i gj2 ¼, gji
ð j ¼ 1; 2; . . .; 5; i ¼ 1; 2; . . .; 6; i – 1Þ ð j ¼ 1; 2; . . .; 5; i ¼ 1; 2; . . .; 6; i – 2Þ
It turns out that the dummy D1 is most suitable, the dummy D2 is seldom suitable for measuring the clothing pressure synthetically on the shoulder, front armscye, scapula, back armscye and upper arm front with the posture flexing upper limbs 908. It is considered that D2 are seldom suitable for sizes such as the size of the chest are much bigger than the size of the standard type 92A5. 5. Conclusions In order to obtain the basic data for the clothing pressure evaluation of men’s suit, with the wearing experiment on the subjects and the body dummies, the clothing pressure value are measured with normal standing posture and the posture flexing upper limbs 908. To relate the pressure measured with the body dummies and the subjects’, grey incidence analysis theory was used to obtain the base data for clothing pressure evaluation of men’s suit. The correlation between the dummies and the subjects’ are given as follows. (1) It turns out that the dummy D1 and D6 are relatively suitable, D3 and D2 are seldom suitable for measuring the clothing pressure synthetically with the normal standing posture. (2) It turns out that the dummy D1 is suitable, D4 and D6 are relatively suitable, D2 is seldom suitable for measuring the clothing pressure synthetically with the posture flexing upper limbs 908. (3) It turns out that it is difficult to identify that dummy is much more suitable for measuring the clothing pressure with the normal posture. For further study on the presumption of clothing pressure of men’s suit with dummy which may replace human body with normal standing posture and the posture flexing upper limbs 908, the compression hardness of material, dummy’s form and size, and postures need to be carefully considered.
A study on clothing pressure
333
IJCST 15,5
334
References Deng, J. (1982), System and Control Letters, Vol. 5, p. 228. Deng, J. (1987), Fundamental Method of Grey System, Huazhong University of Science and Technology Press, Wuhan of China, pp. 17-31. Deng, J. (1993), The Journal of Grey System, Vol. 1, p. 1. Dongsheng, C. and Zhao, S. (1996), Text. Mach. Soc. of Japan, p. 154, The collection of the 49th exhibition summaries. Dongsheng, C., Kimi, S. and Nobuyuki K. (1997), Jpn. Res. Assn. Text. End-Uses, Vol. 51, p. 519. Haruko, M. and Sachiko, I. (1983), J. Home. Econ. Jpn., Vol. 34, p. 39. Hasebe, Y., Haruko, M., Takako, H. and Sachiko, I. (1976), Res. J. Liv. Sci., Vol. 27, p. 117. Masako, N. and Kimiko, I. (1985), J. Home. Econ. Jpn., Vol. 36, p. 779. Masako, N., Kimiko, I. and Yoko, Y. (1985), J. Home. Econ. Jpn., Vol. 36, p. 184. Masako, N., Yoko, Y. and Kimiko, I. (1984), J. Home. Econ. Jpn., Vol. 35, p. 854. Mika, K., Dongsheng, C., Tomoko, Y. and Noriko, I. (2001), Japan Society of Home Economics, The collection of the 53th exhibition summaries, p. 193. Momoko, S. and Mitsuo, M. (2000), Jpn. Res. Assn. Text. End-Uses., Vol. 41, p. 700. Nobuko, O. and Yayoi, K. (1993), J. Home. Econ. Jpn., Vol. 44, p. 573. Noriko, I., Michiko, Y., Takayuki, N. and Tsuneo, H. (1985), Jpn. Res. Assn. Text. End-Uses., Vol. 26, p. 204. Noriko, I., Yuriko, Y., Akemi, Y. and Tomoko, Y. (1995), Jpn. J. Clo. Res., Vol. 39, p. 57. Ruquan, Z. and Mitsuo, M. (1998), Jpn. Res. Assn. Text. End-Uses, Vol. 51, p. T87. Sachiko, I. and Haruko, M. (1983), J. Home. Econ. Jpn., Vol. 34, p. 30. Sifeng, L. and Yi, L. (1998), An Introduction to Grey Systems: Foundations, Methodology and Applications, IIGSS Academic Pub., USA, pp. 61-117. Ume, K. and Hasebe, Y. (1980), J. Home. Econ. Jpn., Vol. 31, p. 34. Ume, K. and Hasebe, Y. (1982), J. Home. Econ. Jpn., Vol. 33, p. 34. Wang, X. (1993), Concise Textbook of Grey System, Chengdu University of Science and Technology Press, Chengdu of China, pp. 8-20. Yuko, M. and Haruko, K. (1998), Jpn. Res. Assn. Text. End-Uses, Vol. 39, p. 452. Yuko, M. and Haruko, M. (1998a), Jpn. Res. Assn. Text. End-Uses, Vol. 39, p. 382. Yuko, M. and Haruko, M. (1998b), Jpn. Res. Assn. Text. End-Uses, Vol. 39, p. 636.
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Apparel sourcing: assessing the true operational cost Robert H. Lowson Strategic Operations Management Centre, School of Management, University of East Anglia, Norwich, Norfolk, UK
Assessing the true operational cost 335 Received May 2002 Accepted February 2003
Keywords Information research, Clothing, Sourcing Abstract In a earlier debate, it was suggested that for many reasons, the decision by a retailer to source low-cost clothing offshore from low-wage suppliers may be ill-advised. We were able to show that using lower priced textiles and apparel manufactured by foreign sources could be sub-optimal operations strategy. In numerous cases, those relying upon this form of procurement failed to consider all the relevant information. Despite the obvious attraction of low cost, there were serious trade-offs and disadvantages. We classified the latter as the hidden costs of importing (for example, delays, use of airfreight, administrative and quality costs, etc.) and the inflexibility costs. When properly attributed and quantified, these disadvantages often outweighed the benefits of low cost foreign supply. It was at this point that we proposed the need for an objective, axiomatic framework (widely accepted across the textile industry) to demonstrate the full implications of domestic versus offshore purchasing – a total acquisition cost model. Here, we expand this thinking, and begin to explore how such a model can be developed using the data obtained from a sample of international textile and clothing retailers and their suppliers.
The costs of operational inflexibility The hidden costs of apparel importing are relatively easy to identify and compute. Less apparent are the costs of inflexibility. We know that these involve issues such as longer lead-times and a general lack of flexibility and or response to demand changes (both before and during a sale season). Our aim was to better understand these factors as a first step to their quantification [1]. LISP interactions We wished to discover how factors such as lead-time, inventory, supplier performance and customer service level might be used to assess certain elements of an operations strategy such as sourcing. To appreciate the underlying impact of these elements, Figure 1 displays the relationships involved in what we termed as the LISP interactions. There are four fundamental quantities in any supply system. (1) Lead-time for supply; (2) Inventory at a particular supply pipeline stage; (3) Customer Service level; and (4) Supplier Performance, this involves two factors: supplier service level and supplier process time (in this retail example, to convert raw materials or components into finished goods).
International Journal of Clothing Science and Technology Vol. 15 No. 5, 2003 pp. 335-345 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310492606
IJCST 15,5
336
Figure 1. LISP interactions in an apparel supply network
These components interact and the implied value for one is derived from the other three. The result of this interaction is an important determinant of the inflexibility costs involved in an operations strategy. Figure 1 also sketches these connections between supply system tiers. Fisher (1997) points out that flexibility and speed of the supply system can help to overcome the impact of uncertainty, such as a marked change in demand profile, but typically at a cost. For any operation there are a number of relationships that can be used to assess these costs, whether they are in offshore or domestic supply systems. For the apparel retailer’s in this research, uncertainty came from many sources. In particular they cited: . inability to forecast customer demand (87 per cent); . primary and secondary tier manufacturer reliability/quality (especially offshore) (85 per cent); . logistical system reliability (77 per cent); . complexity in the manufacturing supply process (68 per cent); and . inaccuracies in point of sale data or bar codes (57 per cent).
In the next three figures we attempt to quantify these LISP relationships for the retailer’s[2] in this study as a way to compare the costs of uncertainty in sourcing (offshore compared with onshore). First, Figure 2 examines how customer service level is related to retail store inventory for short, medium and long order and re-order lead times. We can adopt a quantitative measure for lead-time performance whether long (1-50 per cent), medium (51-80 per cent) or short (81-100 per cent). The LISP relationships can then be plotted for each of the lead-time performance profiles. For example, at 25 per cent along the lead time contour, a retailer’s apparel inventory levels are likely to be relatively high (the aim being to decrease them to a holding level matching real-time customer demand), and customer service levels low (line a-a). In the medium range at performance 75 per cent, the reduction in lead times has had a proportionate effect upon customer service levels and inventory (line b-b). With much shorter lead-time (say, performance 90 per cent) we can see that again customer service level and clothing inventory levels for the retailer would improve (line c-c). However, in real-life things are not always this simple. Our research demonstrates that, in fact, these relationships will not always be linear, that is, continual improvement in lead-time will not continue to provide similar-sized improvements in customer service and inventory statistics (Hunter et al., 2002). As shown in Figure 2, an optimum point for improvement will often be reached. After that, further reductions in lead-time for the retailer, say, between the performance points 90 and 100 per cent, although bringing commensurate increases in customer service, will be at the expense of slightly increasing finished goods inventory needed to support an improvement in the other variables. Thus, for any organization, operational improvement is to a degree a matter of moving between curves rather than along them.
Assessing the true operational cost 337
Figure 2. The relationship between the customer service level and inventory for the lead-time performance
IJCST 15,5
338
Figure 3. The relationship between the customer service level and inventory for the supplier performance (service level)
We can now undertake a similar analysis for the relationship between customer service levels and clothing inventory for high, medium and low supplier service performance. The interactions in Figure 3 are the first element (service level) in an examination of supplier performance. These demonstrate the effect of increasing the supplier performance level from low, medium to high. There are subsequent improvements upon both finished goods apparel inventory at store level (a reduction) and customer service level. However, there will again be a critical point after which further improvement in supplier service level will be at the expense of increasing the inventory levels over and above what is needed to satisfy customer demand (between, for example, 90 and 100 per cent). The final LISP example is shown in Figure 4. The relationship between the customer service level and inventory of textile and clothing for low, medium and high supplier process time. This is the second element of supplier performance ( process time). In this scenario, as the process times improve we see a related reduction in inventory levels and an increase in customer service until the watershed point (say, around 90 per cent). Thereafter, a continued reduction in process time will eventually start to drive-up inventory over and above the actual customer demand requirements. We then have a situation whereby the supply system uses buffer inventory, as the process time is so short that the retailer is over-stocked and the manufacturer struggles to comply with schedules. The above LISP analysis could be performed equally well for a supplier rather than a retailer, and could be focused up- or down-stream. However, further research is required to better understand and quantify this break-even or watershed point for all the LISP interactions.
Assessing the true operational cost 339
Figure 4. The relationship between the customer service level and inventory for the supplier performance (process time)
Three-dimensional application of LISP interactions The next stage was to apply this thinking to the firms taking part in the research. A European-based textile and apparel retailer (Retailer “A”) agreed to take part. This enterprise supplies high quality, fashionable clothing, with both men’s and women’s ranges, aimed at the 20-30 year-old customer. It has outlets in most European and North American cities. The various garments it supplies can be classed as follows. . “Basic” (generally sold all round the year with infrequent changes); . “Seasonal” (shelf life of, on average, 12-25 weeks); and . “Short-season” (shelf life of 6-10 weeks or less). For the purposes of this research we chose an example from each category. A short-season product (women’s styled summer shirts), a seasonal garment (women’s denim jeans) and a basic item (women’s underwear). It is our contention that the various LISP elements will vary in importance between the different product groups. In a normal season, Retailer “A” will combine two sourcing strategies for the supply of many of these garments. (1) Offshore, low cost (OLC ). The bulk of its product requirements are agreed with a number of Far East suppliers (mainly Hong Kong, Taiwan and Malaysia; although some of the firms re-import to Chinese manufacturers). Manufacturing commitments are often negotiated between 8 and 12 months in advance. During the sale season there is little flexibility for re-estimation and re-order, and most of the garments are delivered in advance of the particular open-for-sale date.
IJCST 15,5
340
(2) Domestic, responsive and flexible (DRF). Retailer “A” also uses domestic suppliers (mainly European and North American SMEs) that offer a fast and flexible reaction to changes in demand (whether volume, size, colour or style preference). Using operations strategies such as quick response, time-based competition and agile production, these firms provide fast-turn manufacturing and shipment of products that are in demand. Some pre-capacity agreements are made in advance, but mainly the suppliers have to be prepared to deal with last minute orders from the retailer with little advance warning (a situation that is far from satisfactory). It is important to remember that we are looking at a method to quantify the performance of each of the following variables as part of the LISP interaction. (1) Lead time; (2) Supplier service level (supplier performance); and, (3) Supplier process time (supplier performance). Each being correlated with: (1) customer service level; and (2) inventory for retail finished goods levels. We are not merely dealing with individual and isolated measures of performance (for lead-time, for example). With the LISP model, we asked a number of senior managers involved in making procurement decisions to rank and score the LISP variables for each of the three product groups using both sourcing strategies (OLC and DRF). This activity was by no means easy, as the final scores reflect a number of three-dimensional interactions that can be applied to whatever sourcing strategy is being used (offshore or onshore). These relations were considered in the following format for the chosen product groups (see Figures 2, 3, and 4): . Case 1. An interaction between lead-time (dependent variable) and customer service level and inventory (independent variables). . Case 2. An interaction between supplier service level (dependent variable) and customer service level and inventory (independent variables). . Case 3. An interaction between supplier process time (dependent variable) and customer service level and inventory (independent variables). Table I demonstrates the results of this performance scoring. A moments reflection is in order. In the following we compare the performance of each of the product groups examined. . Women’s shirts (short-season). When lead-time performance (Case 1) was ranked against the levels of customer service and inventory, Retailer “A”
40
47
73
83
81
Case 2. Supplier service level performance (dependent variable); customer service level and inventory (independent variables)
Case 3. Supplier process time performance (dependent variable); customer service level and inventory (independent variables)
Total performance (per cent)
45
47
87
Case 1. Lead-time performance (dependent variable); customer service level and inventory (independent variables)
LISP interactions
Women’s styled summer shirts (short season) Domestic, responsive, Offshore, low flexible cost
73
80
80
60
42
63
30
33
Women’s denim jeans (seasonal) Domestic, responsive, Offshore, low flexible cost
41
47
30
47
Domestic, responsive, flexible
34
11
40
52
Offshore, low cost
Women’s underwear (basic)
Main sourcing strategy performance (per cent)
Assessing the true operational cost 341
Table I. LISP performance scores
IJCST 15,5
342
.
.
scored the performance of its domestic, responsive and flexible operations strategy as being nearly twice as efficient as the offshore, low cost option. The same was true of supplier service level (Case 2) and supplier process time (Case 3) – in each, the performance was in excess of 44 per cent higher. Women’s denim jeans (seasonal). Here again, for all three dependent variables, the domestic option scored far higher (in the magnitude of a 57 per cent improvement). Women’s underwear (basic). This is an interesting situation. The offshore, low cost operations strategy outperforms its domestic counterpart in two of the three categories (Cases 1 and 2). Only in process time performance (Case 3) do we see the domestic strategy being considerably more advantageous. Does the volatility of the product group make a difference?
We can now see that when the costs of inflexibility begin to be measured, the low cost procurement operations strategy is not always preferable. Admittedly, this research case has used a somewhat subjective approach. However, as part of a full total acquisition cost model, the system is designed to be applied at a strategic level by management from a number of firm’s in a supply network. Not only do the results from Table I help us ascertain the costs of inflexibility, they can also help to shape the type of operations strategy used and its components. In the work of Lowson (2002), research findings demonstrated that a number of possible operations strategies existed. Further, each of these had common “building blocks” that involved strategic decisions about: . core competencies, capabilities and processes; . technologies; . resources; and . key tactical activities vital to support a particular strategy or positioning. However, despite having some common components, it also became clear that these strategies were unique in the way they were blended into a particular architecture matching the individuality of the situation. An emphasis was evidently placed upon some higher order strategic themes and their interconnections, over and above other support activities. Returning to the research conducted with Retailer “A”, we can now use the various flexibility and inflexibility performance costs to assess and to shape a firm’s operations strategy (Table II). Table II uses just one product as an example (women’s short season summer shirts) and compares the two sourcing strategies, DRF versus OLC. The first task for the executives from Retailer “A” was to agree on a factor weighting for each of the LISP three-dimensional performance interactions. Using the graphics in Figures 2-4, the management team reached a consensus as to
33 42 12 23 62 55 63
55 45 33 60 78 62 81 46
44 57 70
67 72 86
66
61
OLC 55 (per cent)
87
DRF 92 (per cent)
59
24 55 76 75
34 15
66
80 60 87
78
49
30 49 60 60
49 15
44
48 59 66
65
Case 2. Supplier service level (DV), customer service level and inventory (IV) Factor weighting DRF 88 OLC 64 (per cent) (per cent)
50
34 43 55 78
23 0
58
67 52 66
74
DRF 72 (per cent)
44
17 37 41 65
41 1
32
46 65 72
63
OLC 38 (per cent)
Case 3. Supplier process time (DV), customer service level and inventory (IV)
23 49 52 63 OSIR 47
OSIR 58
44 9
36
46 60 69
63
OLC (per cent)
39 59 64 78
34 16
60
71 61 80
80
DRF (per cent)
Component weighting bands (per cent)
Notes: SS ¼ Short season product group; DV ¼ Dependent variable; IV ¼ Independent variable; DRF ¼ Domestic, responsive, flexible; OLC ¼ Offshore low cost; OSIR ¼ Operations strategy importance ranking
Total (per cent)
Fast re-estimation and re-order systems Continual replenishment Electronic reorder Compressing open-to-buy dates Shared and open inventory management systems Container shipping codes Consumer demographic information systems Internet connectivity Store ready deliveries Universal product codes PoS data sharing between customers and suppliers
Operations strategy components
Case 1. Lead-time (DV), customer service level and inventory (IV)
Product group women’s styled summer shirts (SS) LISP three-dimensional performance interactions for particular sourcing strategy used (offshore or onshore)
Assessing the true operational cost 343
Table II. Retailer “A” operations strategy composition summary
IJCST 15,5
344
(1) the importance of the dependent variable for the particular sourcing strategy; and then (2) the influence of the independent variables upon it. This factor weighting was expressed as a score out of 100 and marked as the optimal point on one of the curves. Next, taking a selection of operations strategy components (we have not included the full list of possibilities), a score out of 100 was given for each component of the operations strategy. Here, the aim was: (1) to decide the importance for each element for the particular operations strategy, and (2) the influence upon the dependent variable in each LISP three-dimensional interaction. The two factor scores were then turned into a percentage. The analysis in Table II gives Retailer “A” the following information. (1) A percentage score indicating the importance of a particular component for each operations strategy. The score takes into account not only the type of sourcing strategy, but also the importance of the various 3D LISP performance interactions in the supply system. (2) In the final column, the component weighting bands, we can see the overall importance of each component for the particular strategy type. This information is vital in allowing the organisation to customise its operations strategy by placing a particular emphasis upon those building blocks that score particularly highly. (3) In the total row at the bottom, the percentage score indicates the importance of this combination of components for each of the LISP performance interactions and this type of strategy. (4) Finally, the operations strategy importance ranking (OSIR) allows a comparison of average strategy/component scores for this product group with others. It is then also possible to customise the resultant strategy by product group (or even customer), and the components that will suit those particular circumstances. Conclusion and further research This paper reported the initial work aimed at producing a total acquisition cost model (TACM) that could be used to quantify the true cost of importing foreign apparel from low wage economies. The costs involved in this type of sourcing strategy can be classified as hidden and inflexibility costs. The latter was our primary concern. We demonstrated, using empirical research, the main three-dimensional interactions in a supply system (LISP variables). These were used to compare the inflexibility costs of offshore procurement versus domestic, responsive and flexible supply. These concepts were tested by a leading retailer in order to compare the performance of the two different
procurement strategies – in this way the costs of inflexibility could be measured and then contrasted with its responsive and flexible counterpart. The greater understanding of inflexibility cost also brought a secondary benefit. The information gained allowed the retailer involved to view its existing operations strategy as an architecture containing a fusion of interconnected “building blocks”. The metrics obtained from the earlier work could be used to assess the contribution of the operations strategy and its core components. In this way, the strategy could be customised by emphasising certain key elements; thus becoming “tailored” to the needs of each product and demand situation. The development of the TACM continues. In particular, further quantification of the costs of inflexibility and the hidden costs of importing are required in order to provide a more detailed assessment. In addition, a second stage is necessary that will analyse and quantify the advantages and disadvantages of foreign as opposed to domestic supply of textiles and clothing in terms of gross margins, sales revenue, etc. Work is also needed to refine the three-dimensional LISP interactions and to make them more user-friendly. Futuristically, we also seek to establish whether the volatility of apparel categories (basic, seasonal and short-season) has any bearing upon the various costs involved. Needless to say, we welcome any constructive suggestions from industry and academia. Notes 1. This publication draws upon data from a current 4-year empirical research study of the clothing goods sector. An international initiative, covering retailers and their suppliers in both North America and Europe, the overall objective of the project is a better understanding of operations strategies in an increasingly volatile and dynamic commercial setting. In this part of the study, senior managers from 167 different apparel and textile retailers were identified. The sample was intended to be representative of the sectors that provide such clothing normally available in a shopping centre, high street or department store. Of the 167 mailed surveys, 78 were completed (an adjusted return of 48 per cent). Each of the respondents was subsequently interviewed (many more than once) for clarification and expansion of some issues. 2. This type of analysis applies equally well at any clothing supply pipeline stage. References Fisher, M.L. (1997), “What is the right supply chain for your product?”, Harvard Business Review, pp. 105-16. Hunter, N.A., King, R.E. and Lowson, R.H. (2002), The Textile/Clothing Pipeline and Quick Response Management, The Textile Institute, Manchester, UK. Lowson, R.H. (2002), Strategic Operations Management: The New Competitive Advantage, Routledge, London. Further reading Lowson, R.H. (2001), “Analysing the effectiveness of European retail sourcing strategies”, European Management Journal, Vol. 19 No. 5, pp. 543-51.
Assessing the true operational cost 345
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
IJCST 15,5
346
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Three-dimensional garment modelling using attribute mapping Yves Chiricota
Received November 2002 Accepted April 2003
De´partement d’informatique et mathe´matiques, Universite´ du Que´bec a` Chicoutimi, Chicoutimi, Que´bec, Canada Keywords Modelling, Clothing Abstract We propose a three-dimensional (3D) geometrical modelling algorithm based on the mapping of 2D objects on a 3D model. Our methodology can be applied to the automatic modelling of many “secondary” garment parts like collars, waist bands, pockets, etc. The results obtained are accurate in relation to the original flat patterns. Our approach is oriented towards the automation of the process of 3D garment modelling from flat patterns. An underlying constraint behind our approach consists in minimizing user intervention in the modelling process. Our method leads to an intuitive interface for novice users.
International Journal of Clothing Science and Technology Vol. 15 No. 5, 2003 pp. 346-358 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310492615
Introduction Three-dimensional (3D) modelling of clothing is a topic of premium importance for the garment industry, as reduction of production costs is a motivation for the research in this field. Many researchers have introduced software systems aimed at solving this problem, an aspect of which consists in creating realistic models representing garments from flat patterns. A system has been developed by Okabe et al. (1992) which maps a 2D pattern on a 3D mannequin and then applies a physical simulation which leads to the garment model. Their system can also achieve the inverse operation, that is, generate 2D patterns from 3D models. Another system has been introduced by Stylios et al. (1996), for the presentation of a virtual fashion show. For their part, Kang and kim (2000 a, b) introduced a system based on a scripting language that allows the calculation of 3D models of garments. Moreover, Moccozet et al. (2001) developed a system allowing garment modelling and animation. An interesting application of their work is the development of a virtual fitting room (Protopsaltou et al., 2001). Production of 2D patterns from 3D shapes has been specifically addressed by some researchers. A system was introduced in Hinds and McCartey (1990) and Hinds et al. (1992, 1999) which allows to design a garment in a 3D environment as a collection of panels placed around a mannequin by the user. Another system, by Bonte et al. (2000), allows to design a garment directly in a 3D environment. Using a similar methodology, Cugini and Rizzi (2002) introduced a system allowing to generate 2D patterns for men’s jackets from the 3D shapes.
In this paper, we introduce a 3D geometrical modelling algorithm based on the mapping of 2D objects on a 3D model. Our approach is oriented towards the automation of the process of 3D garment modelling from flat patterns. An underlying constraint behind our approach consists in minimizing user intervention in the modelling process. This point of view is motivated from the observation that pattern makers often have difficulties when using 3D tools to manipulate garments in CAD systems. Most of them are not expert users, which is not the case in other fields, for example, in computer animation. It is substantially easier for most pattern makers to work in a 2D environment instead of using 3D tools. We propose a methodology that can be applied to the automatic modelling of many “secondary” garment parts like collars, waist bands, pockets, etc. The results obtained are accurate in relation to the original flat patterns. Our method leads to an intuitive interface for novice users. In fact, all the information necessary to the modelling process is fed by the user in a 2D window presenting flat patterns. The remaining of the modelling process is achieved by the program. The reader may refer to Chiricota et al. (2001) for the modelling of the main part of the garment. One may notice that the approach introduced here is general enough to be applied to other modelling methods. In the next section, we will introduce some notation and briefly recall the modelling algorithm used for the main part of a bodice. Then, we will present point mapping and curve mapping, which are the basic blocks in the construction of more sophisticated mappings. Next, applications are presented, in relation to the modelling of collars and pockets. Moreover, we introduce a device called repulsers, preventing surface interpenetration during a physical simulation. Mapping is used to define these objects.
Garment modelling
347
Attribute mapping Let us begin this section by the introduction of some notations. Notations We represent flat patterns as collections of polygonal curves (O’Rourke, 1994) lying in a bi-dimensional Euclidian space (denoted by E 2). A simple closed polygonal curve defining the contour of the piece, called contour curve, corresponds to each garment piece. Other curves are commonly included inside the contour, which may represents darts, notches, etc. Figure 1 shows a typical piece corresponding to a sleeve.
Figure 1. An example of a garment piece
IJCST 15,5
348
Let us introduce some notations related to planar triangulations and 3D meshes. This will serve in the description of our algorithms. For a complete exposition, one may refer to de Berg et al. (1997). Given a triangulation D, triangles t will be denoted by t ¼ ðP 0 ; P 1 ; P 2 Þ, where Pi [ E2 with Pi– Pj if i–j. We will denote edges e with e ¼ ðP 0 ; P 1 Þ, where P0–P1. Given a triangulation D, the set of points belonging to D will be designated by p(D). We will denote the edges of D by Y(D) and the triangles by d(D). The same notation will be used for meshes M. Let c be a simple polygonal curve and A, B [ c. We will denote by Gþ (A, B; c) (resp. G2 (A, B; c)) the sub-curve of c going from A to B clockwise (resp. counter-clockwise). In the following, we will consider parametrized curves (Farin, 1996). This is an application c : ½0; 1 ! E2 (or c : ½0; 1 ! E3 ). One can define a parametrized curve from a polygonal curve in the following manner. Let us first define the parametrized segment corresponding to an edge e ¼ ðP 0 ; P 1 Þ as sðP 0 ; P 1 ; uÞ ¼ sðuÞ ¼ uP 0 þ ð1 2 uÞP 1 . We will also write e(u) instead of s (u). Given two parametrized curves c0 and c1, we define the concatenation cðuÞ ¼ ðc0 þ c1 ÞðuÞ of these two curves. Let a ¼ ‘ðc0 Þ=‘ðc0 Þ þ ‘ðc1 Þ, the concatenation c is given by 8 u if u [ ½0; aÞ > < c0 a cðuÞ ¼ a > : c1 u2 if u [ ½a; 1 12a Using these definitions, the parametrized curve corresponding to the list of edges (e0, e1,. . .,ek ) is simply the concatenation e0 þ e1 þ · · · þ ek of the parametrized segments corresponding to these edges. The length of the curve c denoted ‘(c) is defined by ‘ðcÞ ¼
X
‘ðei Þ;
i
where ei ¼ ðP i ; P iþ1 Þ and ‘ðei Þ ¼ jP iþ1 2 P i j is the length of the edge ei. Garment modelling The mapping method presented in this paper may be used conjointly to any modelling process which transforms flat pattern pieces to 3D meshes representing the garments. In the present work, we apply it to the modelling process introduced in Chiricota et al. (2001). This process serves to calculate the main part of a bodice and consists mainly of two steps, the first of which leads to a 3D geometrical model calculated from the information added to the pattern by the user. Then, the model is converted to a masses-springs system, and a physical simulation is performed. The following is a brief description of the process.
The information added by the user in the flat pattern is of two kinds: assemblies and reference points. Assemblies consist of sewing information between pieces. Reference points are special points aimed at identifying specific location on garments. For each type of garment, these points are predefined. For example, in the case of a bodice, there are points identifying the center of the garment, the under-arm, and so on (Figure 2). Two important reference points defined by the user are the center points. The line going from the center up to the center down points is called center line of the fragment. Once the required information has been added by the user, the remaining of the process is automatic. This part of the algorithm is designed to be applied to standardized pieces called fragments. Fragments are predefined for each type of garment. For a bodice, there are four fragments denoted by f0, f1, f2, and f3, corresponding to front/back and left/right combinations (Figure 3). Fragments are calculated using information from reference points and assemblies by joining initial pieces together [1]. Once the fragments have been calculated, each of the corresponding contour curves is triangulated. The triangulation associated to a fragment f will be denoted by D( f ) and the contour curve will be denoted by g ( f ). From the measures taken on the fragments, a 3D geometrical model composed of polygonal meshes is calculated. The meshes are calculated so that the triangles in the meshes are related to triangles belonging to fragment triangulations in such a way that there is a one-to-one correspondence t 7 ! t between the two sets of triangles and another correspondence e 7 ! e for the edges. One may note that we use the same notation for these correspondences. The correspondences preserve the adjacency relation between the triangles. That is, if two triangles t0 and t1 are linked by an
Garment modelling
349
Figure 2. Reference points
Figure 3. Bodice fragments
IJCST 15,5
350
edge e in a fragment, then the corresponding triangles t 0 and t 1 will be linked by the edge ¯e. In fact, one can consider the modelling algorithm as a deformation of the fragments in the 3D space leading to the meshes. The correspondence t 7 ! t between triangulations and meshes is fundamental in our algorithm. We use it to map objects from E 2 to objects in the 3D space using two methods. First, the correspondence allows to locate any point inside the fragments in the 3D space using barycentric coordinates, and second, it allows to map curves. Let us look at each of these two kinds of mapping in detail. Mapping This section presents the concept of mapping used in the remaining of this paper. Point mapping. The mapping method presented here relies on the correspondance t 7 ! t between triangulations and meshes representing a garment. We consider two different ways for mapping points between flat triangulation and 3D models. The first is simply a direct one-to-one correspondance between the points in p (D) and p (M) induced by the correspondance t 7 ! t. We call this as direct point mapping. The second type of mapping relies on barycentric coordinates (Figure 4). Given a point P in E2, the method to find the corresponding point m(P) in E3 consists of the following steps. . Find a fragment f containing P and let D( f ) be the associated triangulation. . Find t [ D( f) the triangle containing P (with t ¼ ðP 0 ; P 1 ; P 2 ÞÞ. Let t ¼ ðP 0 ; P 1 ; P 2 Þ be the triangle belonging to the mesh that corresponds to t. . Calculate the barycentric coordinate of P in relation to the triangle t, that is, calculate a0,a1,a2 [ R such that ai $ 0, a0 þ a1 þ a2 ¼ 1 and P ¼ a0 P 0 þ a1 P 1 þ a2 P 2 . The vector (a0, a1, a2) contains the barycentric coordinates of P in relation to t. . Finally, let mðPÞ ¼ a0 P 0 þ a1 P 1 þ a2 P 2 .
Figure 4. Barycentric point mapping
The point P ¼ mðPÞ in the model corresponding to the point P in the pattern is located at the same position in ¯t as P in relation to t. We call the mapping based on this technique as barycentric point mapping. Curve mapping. In our software, polygonal curves are implemented from lists of edges from triangulations (in E2) or meshes (in the 3D space). These lists have the form c ¼ ðe0 ; e1 ; . . .; ek Þ where ei [
Garment modelling
351
Applications The first application concerns the calculation of collars in 3D from the flat pattern, using our algorithm which allows to automatically calculate a 3D model from a flat pattern. Next, we present the repulsers application. Repulsers
Figure 5. Global curve mapping
IJCST 15,5
352
Figure 6. Sub-fragment of collars
are used to prevent interpenetration of layers of fabric during a masses-springs simulation. These 3D objects are calculated from the flat patterns using direct point mapping. Finally, pocket modelling is described. Collar We will show the application of our method to the modelling of shirt collars. The calculation of the 3D model for a shirt collar is a two fold process. First, a flat pattern is measured. Then, the measures are used to calculate the 3D model. Curve mapping is applied in the second step to calculate the location points in 3D. Now let us describe the measurement process and the construction of the collar model in 3D. The following exposition relies on the supposition that the main part of the bodice model is constructed. The curves used in the implementation of the model are B-spline curves. We use ruled surfaces and Coons patches to calculate meshes. The reader will find definitions and properties of these objects in Farin (1996). A typical pattern for shirt collars is divided into two sub-polygons corresponding to the base and a lapel. This is shown in Figure 6(a). Three curves are calculated from the original contour curve c. Using the notation defined earlier, these curves are the base curve b ¼ G2 B; C; c, the fold curve f, and the lapel curve l ¼ Gþ A; D; c. The curve f is interpolated from the other two. In fact, this is the quadratic curve passing through the points b(0), 2/3b(1/2) + 1/3l(1/2) and b(1). These curves serve to measure the base and lapel sub-polygons, for which the measurement consists in calculating couples of points (b(uk), f(uk)) calculated along the curves b and f, where uk ¼ k=n, k ¼ 0; 1. . .; n (Figure 6(b)). For each k, we obtain a measurement vk ¼ jbðuk Þ 2 f ðuk Þj, that is the distance between b(uk) and f(uk). The measurement of the lapel sub-polygon is similar to the previous one, but the points used are f(uk) and l(uk) 0 leading to the measurement vk ¼ j f ðuk Þ 2 lðuk Þj. Moreover, in this case, an angle is also measured. This is the angle uk between the vector f ðuk Þ 2 lðuk Þ and the tangent vector to the curve f at the point f(uk). Once the measurements have been made, we proceed with the construction of the 3D model. This is based on the mapping of the collar base curve b belonging to the flat pattern to the collar curve c in the 3D bodice model (Figure 7). Using the earlier notation, this mapping is written as fðbÞ ¼ c. The collar is calculated in two steps. First, the base is constructed and then the lapel. The first step in the construction of the base
consists in calculating ˆf, the 3D curve corresponding to the fold curve f. This is achieved by calculating the points Qk from the measurements using the following formula: Qk ¼ Bk þ vk M k , where Bk ¼ fðbÞðuk Þ ¼ cðuk Þ and Mk is the unit vector normal to the surface corresponding to the bodice, at point Bk. Next, the curve ˆf is obtained from an interpolation through these points. Finally, these two curves lead to a ruled surface defined by the formula ^ þ ð1 2 vÞcðuÞ ðu; v [ ½0; 1Þ. This surface is triangulated, s ðu; vÞ ¼ vfðuÞ leading to a mesh corresponding to the 3D model for the collar base. The calculation of the collar lapel is based on the mapping fð f Þ ¼ f,^ that is, the mapping of the folding curve from the flat collar pattern to the corresponding curve in the 3D collar model. This calculation is shown in Figure 8. First, the algorithm carries out the calculation of the 3D lapel curve ˆl. The curve, ˆl, is obtained 0from interpolation through the points Zk calculated from the measurement vk as 0follows (Figure 8). First, for each k, a point Z 0k is ^ k Þ and Nk is the unit vector defined by Z 0k ¼ Qk þ vk N k , where Qk ¼ fðu ^ k ÞÞ. The point Z 0 is then normal to the surface s at point sðuk ; 1Þ ¼ ðfðu k rotated about the axis T k £ N k (where T is the vector tangent to the curve ˆf k 00 ˆ at point f(uk), leading to a point Z k . The rotation is done in such a way so as to have an angle uk between vectors Z k00 2 Qk and Tk. This part of the calculation is shown in Figure 8(a).
Garment modelling
353
Figure 7. Modelling of collar base in three dimensions
Figure 8. Modelling of collar lapel in three dimensions
IJCST 15,5
354
00
Next, the point Zk is obtained from the rotation of the point Z k about the vector Tk. This rotation is such that the distance between the resulting point and the surface corresponding to the bodice is a relatively small00 value d 0 . 0. Figure 8(b) shows this part of the calculation. In detail, let RðZ k ; cÞ be the point 00 resulting from the (right-handed) rotation of Z k in relation to the vector Tk with angle c, and let 6 be the surface corresponding to the bodice. The point Zk is 00 obtained by solving the equation dðRðZ k ; cÞ; 6Þ 2 d0 ¼ 0 for c, where d(Q,6) represents the distance between a point Q and the surface 6. We use the Newton-Raphson method to solve this equation. The curve, ˆl, is calculated from interpolation through points Zk. Finally, a ruled surface s 0 is defined from the curves ˆf and ˆl, which leads to the mesh representing the collar lapel. In applying the above calculation, the position of points relative to the flat pattern is preserved in the 3D model. Hence, the shape of the collar is preserved. It should be noted that the shape of the 3D model obtained by our method is close to the position that one will obtain in a physical simulation. So, it is natural to use this model as initial state in order to speed-up such a simulation. Figure 9 shows the calculation of a collar with our algorithm. In our implementation, we have used global curve mapping. The use of local curve mapping is also possible, provided the number of edges on the curves involved in the calculation is the same. Repulsers One feature that is of importance to pattern makers is the correctness in the disposition of a layer of fabric during a physical simulation. It is necessary to avoid the interpenetration of layers in order to produce realistic 3D models. In a CAD system based on the direct manipulation of meshes in 3D by the user, it is possible for her or him to arrange fabric layers manually. However, this approach does not correspond to our point of view which is focused on automatic calculation of the models. We address the problem of layer interpenetration in a different way here. In fact, we use direct point mapping
Figure 9. Collar modelling
together with the information fed by the user in the flat pattern, to calculate the special device introduced during the simulation. These objects, called repulsers, are based on the masses-springs methodology and serve to prevent interpenetration of layers of fabric during a simulation. A repulser corresponds to a collection of springs conveniently placed between the layers of fabric. They are automatically calculated by our algorithm and are introduced during the simulation. Let us explain how the repulsers are calculated. We apply them to the center front part of a bodice. The first step in the calculation of repulsers consists in finding the overlapping region around the front center line. The calculation is done with relation to fragments f0 and f1. The polygonal curves g ( f0) and g ( f1) are translated and rotated in such a way that the center lines of each fragment coincide. Then, the regions r0 and r1 corresponding to the intersection of the two contour curves is calculated. Figure 10 shows the result. Next, pairs of points are defined from the regions ri in the following manner. For each point p in r0, we define the set
Garment modelling
355
C 1 ð pÞ ¼ {ð p; qÞ : q [ r1 ; distð p; qÞ , 2d m }; where dm is the mean edges length. We call the set C1( p) the cone at p (Figure 11).
Figure 10. Overlap calculation
Figure 11. Cone
IJCST 15,5
356
The same calculation is done for point q [ r1 , that is, we calculate C 0 ðqÞ ¼ {ðq; pÞ : p [ r0 ; distðp; qÞ , 2d m }: Let S ¼ ð
Figure 12. A garment calculated with and without repulsers
Conclusion The algorithms introduced here have been applied to calculate the models of secondary parts of garments, for example, shirt collars. Our method can be applied to any modelling process based on the construction of a 3D model from a flat pattern, provided triangulation and meshes are used. It is easily feasible to extend the method to other garment parts, like waist bands. Our method is well suited for 3D garment modelling where user intervention must be kept at minimum. Moreover, the calculation of models is almost immediate. Future work will be devoted to the extension of these algorithms to the automatic modelling of other parts of garments. In particular, we plan to apply it to vest collar. Moreover, we expect to use our methodology in the context of garment animation in the industrial context. The information fed by the user into the flat pattern will be used to set some parameters for animation. For example, using this information it could be feasible to restrict the movement of a 3D collar using devices calculated in the same way as repulsers, from the flat pattern.
Garment modelling
357
Figure 13. Pocket mesh
Figure 14. Pocket modelling
IJCST 15,5
Note 1. This part of the process may introduce small inaccuracies in relation to the area of the pieces, due to the difference in shape of pieces sewed together. However, for a large number of patterns we have encountered, these inaccuracies are almost unnoticeable in the resulting garment model.
358
References Bonte, T., Galimberti, A. and Rizzi, C. (2000), “A 3d graphic environment to design garment”, in Cugini, U. (Ed.), Proceeding of the 7th GEO Workshop, pp. 189-206. Chiricota, Y., Cochaux, O. and Provost, A. (2001), “Geometrical modelling of garment”, International Journal of Clothing Science and Technology, Vol. 13 No. 1, pp. 38-52. Cugini, U. and Rizzi, C. (2002), “3d design and simulation of men garments”, Proceeding Workshop WSCG 200, pp. 4-8. de Berg, M., van Kreveld, M., Overmars, M. and Schwarzkopf, O. (1997), Computational Geometry, Springer, Berlin. Farin, G. (1996), Curves and Surfaces: A Practical Guide, Academic Press, New york. Hinds, B.K. and McCartey, J. (1990), “Interactive garment design”, The Visual Computer, Vol. 6 No. 2, pp. 53-61. Hinds, B.K., McCartey, J., Hadden, C. and Diamond, J. (1992), “3d cad for garment design”, International Journal of Clothing Science and Technology, Vol. 4 No. 4, pp. 6-14. Hinds, B.K., McCartey, J., Hadden, C. and Diamond, J. (1999), “The flattening of triangulated surfaces incorporating darts and gussets”, Computer-Aided Design, Vol. 31 No. 4, pp. 249-60. Kang, T.J. and Kim, S.M. (2000a), “Development of three-dimensional apparel cad system. Part 1: flat garment pattern drafting system”, International Journal of Clothing Science and Technology, Vol. 12 No. 1, pp. 26-38. Kang, T.J. and Kim, S.M. (2000b), “Development of three-dimensional apparel cad system. Part 2: prediction of garment drape shape”, International Journal of Clothing Science and Technology, Vol. 12 No. 1, pp. 39-44. Moccozet, L., Thalmann, N.M. and Volino, P. (2001), “Designing and simulating clothes”, International Journal of Image and Graphics, Vol. 1 No. 1, pp. 1-17. Okabe, H., Imaoka, H. and Niwaya, H. (1992), “Three-dimensional apparel cad system”, Computer Graphics, Vol. 26 No. 2, pp. 105-10. O’Rourke, J. (1994), Computational Geometry in C, Cambridge University Press, New york. Protopsaltou, D., Luible, C., Arevalo, M. and Thalmann, N.M. (2001), “A body and garment creation method for an internet based virtual fitting room”, University of Bradford, CGI2002. Stylios, G., Wan, G. and Powell, N.J. (1996), “Modeling the dynamics drape of garments on synthetics human in a virtual fashion show”, International Journal of Clothing Science and Technology, Vol. 8 No. 3, pp. 95-112.
The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister
The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0955-6222.htm
Designing knitted apparel by engineering the attributes of shape memory alloy
Designing knitted apparel
359
Rachael C.C. Winchester and George K. Stylios Riflex, School of Textiles and Design, Heriot-Watt University, Netherdale, Galashiels, Selkirkshire, United Kingdom
Received March 2003 Accepted May 2003
Keywords Textiles, Clothing, Design Abstract This paper discusses how shape memory alloy can be engineered to provide new design concepts that enhance the aesthetic appeal of knitted textiles for apparel. It focuses on the range of techniques and processes used to accommodate the specific characteristics and requirements of the shape memory alloy and it describes a new approach to the overall design construct.
1. Introduction The dynamics of textile design has entered a new and challenging era. Previously, designers were limited in their use of traditional fibres and production methods with the primary rational for output being either decorative or functional. This has changed fundamentally with the emergence of new and innovative technologies, which allow the designer to produce fabrics and clothing that combine function with aesthetics. Fibres and textiles whose applications were traditionally engineering, geo-textiles and space technology, have, with the benefit of new processes, been engineered to meet the specific requirements of the apparel textile industry. These new processes provide the opportunity to improve upon and to generate new functional attributes whilst at the same time, creating the potential for new design aesthetics. These innovations, together with the change in demographics and lifestyle of the consumer have generated an upsurge in the development and production of and demand for multifunctional and adaptive textiles and clothing. This diversification has led to new methods and approaches to design. Multidisciplinary design teams, which use the skills of designers, technologists and engineers, are becoming the established norm. New areas of clothing and textiles are being created with many established fashion and textile companies seeking to exploit the potential of this new technology. This convergence of craft and technology, design and science will see a significant increase in high-tech materials being used in the fashion and apparel industry. Shape memory alloy (SMA) is one such material whose potential application in the development of multifunctional textiles for apparel has yet to be fully explored.
International Journal of Clothing Science and Technology Vol. 15 No. 5, 2003 pp. 359-366 q MCB UP Limited 0955-6222 DOI 10.1108/09556220310492624
IJCST 15,5
360
Figure 1. SMA: MT
2. The principles of SMA SMA is an adaptive, temperature sensitive alloy that has the ability to return to a pre-programmed shape when stimulated by heat or an electrical current; this is known as the shape memory effect (SME). The crystal lattice structure of the alloy is called the martensitic phase transformation (MT) and allows the SME to take place. The MT is a two-phase diffusionless state and relies on the co-operative movement of its atoms, generating elongation, contraction and shear along the planes of the alloy. The two-phase transformation comprises an austinite or parent phase and a martensite phase as shown in Figure 1. Due to the self-accommodating movement of the atoms, a uniform stacking order is achieved with the corresponding crystal structures remaining adjacent to each other. This occurs during the programming and subsequent transition from the martinsite, a lower temperature phase to austinite, the high temperature phase (Otsuka and Wyman, 1998). To create the SME, the alloy is formed and held mechanically in the required shape. Through a process of heating and rapid quenching, the relationship between the different crystal structures is set. The SMA retains its programmed shape at ambient temperature but when deformed it will always return to its original shape when the appropriate stimulus is applied. Historically, SMA has been used for biomedical and engineering applications (Duerig et al., 1997). With regard to textiles, the focus of this material has been the functional aspect, primarily filtration for technical textiles and vascular stents. Its use in apparel textiles is scarce and any inclusion of SMA has been primarily for the purpose of the functional attributes of the material (Wilson, 2002) and not for enhancing its aesthetics. For the purpose of this research 50 per cent equiatomic nickel-titanium (Ni-Ti) wire was selected as the ideal, due to its superiority in terms of transformation, recovery and bio-compatibility. There are, however, particular mechanical properties inherent in Ni-Ti, which can cause malfunction and or degradation of the SME, which need to be considered during the conditioning
and programming of the SMA. Titanium is highly susceptible to oxidation at high temperatures (Otsuka and Wyman, 1998). Therefore, as the alloy is programmed above 4008C, an inert atmosphere is required.
Designing knitted apparel
3. Implementation of aesthetics in knitted structures In order to bring out the aesthetic attributes from the point-of-view of the conceptual design methodology, a number of areas needed to be overcome; stabilising the SMA, determining its optimum diameter size, spinnability and suitability of the knitting process.
361
3.1 Yarn development Although the design methodology has been led by the mechanical attributes of the SMA, traditional design inspiration from a visual motif was a major contributory factor in the development and integration of colour, texture, form and pattern facilitating the creation of a new aesthetic design. It was considered that the aesthetic attributes of the SMA lay solely in its mechanical properties, not in its external appearance, therefore it was used as the core for all yarn developments. Important considerations of handle and drape of the finished textile were also addressed. Challenges have been met in the requirement that the conventional yarns used in conjunction with the SMA need to withstand the extreme temperatures required to successfully programme the SMA. In order that the SME could be produced, it was necessary to programme the SMA wire at temperatures exceeding 6008C for a period of 4 h. Conventional yarns used in the spinning of a yarn composite suitable for apparel fabrics are not able to withstand this treatment. Processing such yarns at high temperatures would cause them to burn, scorch or melt. Consequently, the SMA was programmed prior to its inclusion into the core of the yarns. In preliminary spinning investigations, nickel wire was used as a substitute to SMA because of its similar properties. These trials served a two-fold purpose; first, to gain a technical awareness of how successfully wire could be incorporated into a yarn composite; second, to develop the aesthetic dimension. Through investigation, it was determined that the most efficient way of incorporating wire into the core of the yarn was to feed it horizontally from an X-frame situated behind the spinning machine. The creel frame normally used to hold yarns during the spinning process proved ineffective. Placing the wires package vertically on the creel allowed the wire to spring from the package causing it to kink and twist. Following initial trials with wires of different diameter values, it was deemed that a diameter of between 0.100 and 0.200 mm would be the most appropriate. All yarns were produced using the Gemmel and Dunsmore Fancy Wrap Spinner. The first developments produced yarns that were very twist lively and lacked dimensional stability. The stability of the yarn was also compromised when the SME was induced. As the configuration of the SMA altered to its programmed shape, the core protruded from the centre of the yarn. Solutions to
IJCST 15,5
362
improve the stability of the yarns were explored, with further tests adjusting the levels of twist and the incorporation of different yarns until an even and balanced yarn composite was produced. A variety of yarn types were constructed to accommodate different knitting techniques. These consisted of a composite of synthetic multifilament and staple yarns. The range of conventional yarns included Tencelq, polyester, viscose and polyamide. A fine tex count was necessary in order to cover the wire and prevent its protrusion from the core of the yarns. Two factors needed to be taken into consideration when determining the level of twist to be applied. The first was the coverage of the wire and the second was that the yarn composite would be suitable for apparel end use. Shape change is the fundamental attribute of the SMA and consequently, the yarns were developed to capitalise on this effect. Optical effects have been created by the use of contrasting yarns of matt and shine and light and dark to generate reflective light properties. Figures 2 and 3 show examples of yarns with these properties. The design concept of these yarns is that as the SME takes place, it creates a fabric that has alternating light and colour perspectives. Careful selection of yarn has allowed these effects to be both subtle and dramatic. Overfeeding and the use of textured yarns, as shown in Figure 4, have also been used to enhance the knitted structure. These yarns would be concealed within a knit structure. The SME would cause the fabric to open, revealing the textured yarn beneath. 3.2 Knitted development of structures An in-depth understanding of the technical aspects and requirements of the knitting process needed to take precedence in order not to compromise the
Figure 2. Reflective light
Figure 3. Reflective light
design. The predominant aspects of producing a wire textile is the ability to control flyer pay-off and its associate problem of the “slip/stick/slip” phenomenon. Flyer pay-off occurs when the yarn is drawn off the supply package during the knitting process. The relative cause of the “slip/stick/slip” phenomenon is the lack of extensibility in wire yarn, thus causing it to intermittently stick as it is delivered from the package through the tension feeds. The force applied to the yarn as the cam carriage is taken across the needle bed, produces a sudden release of tension. This results in the yarn springing from the package in long spiral lengths, producing a fluctuation in the tension. The yarn twists on itself, causing kinks to develop along its length. Solutions to these problems included the introduction of a “cats whisker” device. Variation of structure and technique and the philosophy of craft and technology working in tandem were fundamental to the evolution of the aesthetic design. Consequently, hand knitting and the use of two different knitting machines were employed. Each system contributes its own unique stitch and patterning capabilities and allows a greater variety of yarn types to be used. The first textile developments were small sample pieces to establish the handle and drape of textiles constructed from individual yarns. In most samples, the top and bottom of each piece would roll to the middle on the face side of the fabric, forming a cylindrical shape. Unlike traditional fibrous yarns, wire only exhibits elasticity under considerable mechanical stress. Due to the lack of extensibility and the propensity of the yarn to easily deform when knitted, complex structures were not possible. From these results, it was considered that the most appropriate method of introducing SMA to the knitted structure would be in the selected areas of a fabric. The principal aim of the research is to redirect the aesthetic aspects of the fabric from something static to something active, to produce “living” fabrics that evolve; to present the industry as sensual and tactile. The shape memory effect creates a dual, decorative fabric that can exhibit different aesthetic characteristics within the same cloth. Three-dimensional (3D) and sculptural fabrics have been developed that enhance these concepts. Figure 4 shows a fabric that metamorphoses into a 3D surface pattern when the SME takes place. In Figure 5, a 3D pattern already exists in the form of half-moon structures. In this case, these structures evolve into circular forms in response to the SME.
Designing knitted apparel
363
Figure 4. Overfeed
IJCST 15,5
364
Figure 5. Three-dimensional effect
Figure 6. Three-dimensional effect
Figure 7. Mobility
Figure 6 shows a fabric that has been designed with consideration to the placement of an SMA that responds at different rates, thus giving a textile that has the concept of movement around the body (Figure 7). Figures 8 and 9 show how these textile developments could be included in whole garment concepts. They illustrate the subtle way in which the SME would take effect creating 3D textiles. Garments could also be produced that
Designing knitted apparel
365
Figure 8. Garment concept
Figure 9. Garment concept
IJCST 15,5
366
not only change their surface characteristics, but also garment dimensions and form, accommodating the needs of the wearer. 4. Conclusion The inclusion of shape memory materials (SMMs) into the knitted structure introduces a new garment form. Application of SMMs is not limited to fashion apparel only. Its inclusion in athletics sports wear as an aid to physiological performance and comfort and bio-medical textiles i.e compression bandages, could be the potential areas of investigation. Similarly, in the performing arts of theatre and film, this technology could be used to create garments with unique aesthetic attributes that enhance the performance appeal. Knowledge of SMA technology enables the designer to fully exploit these materials. In this research, the contribution of SMA to the textile is aesthetic rather than to the performance and functionality only. Future research will continue the development of yarns and the development and utilisation of more complex structures to engineer new and exciting textile designs. References Duerig, T.W., Pelton, A.R. and Stockel, D. (1997, 2000), khttp://www.devicelink.com/mpb/archive/ 97/03/003.htnl. Otsuka, K. and Wyman, C.M. (1998), Shape Memory Materials, 1st ed., Cambridge University Press, Cambridge. Wilson, A. (2002), “The transfer list”, Future Materials, 2-4 July/August.
International Journal of Clothing Science and Technology
ISSN 0955-6222 Volume 15 Number 6 2003
International textile and clothing research register Editor-in-Chief George K. Stylios
Access this journal online __________________________
2
Editorial __________________________________________
3
Research register __________________________________
5
Research index by institution_______________________ 119 Research index by country _________________________ 120 Research index by subject__________________________ 121 Research index by principal investigator ____________ 123
Access this journal electronically The current and past volumes of this journal are available at
www.emeraldinsight.com/ft You can access over 100 additional Emerald journals, each with a comprehensive searchable archive of articles (many dating back to 1989), a detailed classification system and links to referenced material. See page following contents for full details of what your access includes.
CONTENTS
www.emeraldinsight.com/ijcst.htm As a subscriber to this journal, you can benefit from instant, electronic access to the content of this title. Your access includes a variety of features that increase the value of your journal subscription.
E-mail Services Emerald’s range of free e-mail alerting services is designed to deliver personal notification of news and features in a number of different interest areas.
Automatic permission to make up to 25 copies of individual articles
Research Register Located at www.emeraldinsight.com/researchregister, the Emerald Research Register is an Internet research forum where you and your readers can identify information on research activity world-wide. This feature is invaluable if you are seeking information or wish to disseminate information about your own research.
This facility can be used for teaching purposes, training courses, meetings or seminars. This only applies to articles of which Emerald owns copyright. For further details visit www.emeraldinsight.com/copyright
How to access this journal electronically To benefit from electronic access to this journal you first need to register on the Internet. Registration is simple and full instructions are available online at www.emeraldinsight.com/register Once completed, your institution will have instant access to the journal content from www.emeraldinsight.com or via the journal homepage www.emeraldinsight.com/ijcst.htm Our liberal institution-wide licence allows everyone within your institution to access your journal electronically, thus making your subscription more cost-effective. Our Web site has been designed to provide you with a comprehensive, simple system that needs only minimum administration. Access is available via IP authentication or username and password.
Key features of Emerald electronic journals Online Publishing and Archiving You can gain access to past volumes as well as new material from this journal on the Internet via Emerald Fulltext. You can browse or search the database for relevant articles. Reference Linking Direct links are provided from the journal article references to abstracts of the most influential articles cited (where possible this is to the full text of the article). E-mail an Article This facility allows you to e-mail relevant and interesting articles in PDF format to another PC for later use, reference or printing purposes. Non-Article Content The non-article content of our journals (product and materials information, industry trends, company news, etc.) is available online so that it can be accessed by users. It is possible to browse or search non-article content by keyword.
Additional Complementary Services Available Your access includes a variety of features that add to the functionality and value of your journal subscription:
Support Resources Comprehensive librarian and user toolkits have been created to help you get the most from your journal subscription. For further information about what is available visit www.emeraldinsight.com/usertoolkit
Choice of Access Electronic access to this journal is available via a number of channels, enabling end users and libraries to reach the content through their preferred delivery system. The Emerald Fulltext Web site – www.emeraldinsight.com/ft – is the recommended means of electronic access, as it provides fully searchable and high value-added access to the complete content of the journal. Subscribers can also access and search the article content of this journal through the following journal delivery services: EBSCOhost Electronic Journals Service (EJS) ejournals.ebsco.com Huber E-Journals e-journals.hanshuber.com/english/index.htm Minerva www.minerva.at OCLC Firstsearch Electronic Collections Online www.oclc.org/firstsearch/ SilverLinker www.silverplatter.com Swets Blackwell’s SwetsWise www.swetswise.com
Emerald Customer Support Services For customer service and technical help, contact: E-mail
[email protected] Tel +44 (0) 1274 785278 Fax +44 (0) 1274 785204
Editorial The International Textile and Clothing Research Register Championing the Research Efforts of the Community The International Textile and Clothing Research Register (ITCRR) is in its ninth year of publishing the research efforts of our community. It provides a breadth of activity in the field of textile and clothing research and it encourages participation and dissemination to those working in this discipline and further afield. Indeed as you will see in this edition textile and clothing research is increasing in volume, in quality and in diversity, all good news for all of us involved in it. Research, development and innovation can, without doubt, give us more wisdom, enable our industries to become more competitive, and contribute to our quality of life. I believe that registering research projects will provide the due credit to originators of the research and contribute much more to the future development of this field. Groups of expertise can be identified in this manner, repetition and re-invention can be avoided, leading to best utilisation of time and funding for faster and better directed research in the international arena, since globalisation is on everybody’s agenda. The ITCRR has been set up with all these things in mind. Textiles and clothing originate from the physiological need to protect ourselves from the environment. This has made necessary the art of hand knitting and weaving, and cut and sew processes, which have been evolving for many centuries. Although the original need for clothing has somewhat changed, mechanisation of this process started after the industrial revolution and has continued this century, with automation developments on a massive scale. Considering the upstream part of the whole textile and clothing production chain, yarn-making is the most highly automated area, followed by fabric, with high speed knitting and weaving machines. The downstream part of garment making, however, still remains probably the less developed connection in this chain; one that no doubt many of us have our eyes on as the candidate for development into the new century. With massive computerisation over the last 20 years, logistics and sales have also changed dramatically from pen and paper to electronic data interchange and electronic point of sale. Consistent and extensive research and development in textile and clothing science and technology underpin all these developments by the international research community, whether in educational establishments, in research trade organisations, or in companies. IJCST has been set up 14 years back, as a platform for the promotion of scientific and technical research at an international level. With the statement that the manufacture of clothing in particular needs to change to more q Professor George Stylios
Editorial
3
International Journal of Clothing Science and Technology Vol. 15 No. 6, 2003 pp. 3–4 MCB UP Limited 0955-6222
IJCST 15,6
4
technologically advanced forms of manufacturing and retailing, IJCST continues to support the community in these and other efforts. The journal continues with its authoritative style to accredit the original technical research. The refereeing process will continue, but we will be implementing a pilot scheme, which will be designed to reduce waiting time during the refereeing process. IJCST will be instrumental in supporting conferences and meetings from around the world in its effort to promote the science and technology of clothing. I praise the enthusiasm of our research community and those authors who have made IJCST an invaluable resource to all involved with textiles and clothing. I thank our editorial board for their continuous support and our colleagues who have acted in a refereeing capacity, with commitment to progress our research efforts. I take the liberty to list some of those names below (apologies in advance if anyone has accidentally been omitted from this list): Dr Norman Powell, University of Bradford Dr Taoruan Wan, University of Bradford Professor David Lloyd, University of Bradford Dr Jim Betts, University of Bradford Dr Steve Heycock, University of Bradford Dr G.A.V. Leaf, University of Leeds Dr David Brook, University of Leeds Dr C. Iype, University of Leeds Dr Jaffer Amirbayat, UMIST Dr David Tyler, Manchester Metropolitan University Dr Jintu Fan, Hong Kong Polytechnic University Dr Lubas Hes, University of Minho Dr Jelka Gersak, University of Maribor Professor Paul Taylor, University of Newcastle Professor Haruki Imaoka, Nara Women’s University Professor Mario De Araujo, University of Minho Professor H.J. Barndt, Philadelphia College of Textiles and Science Professor Masako Niwa, Nara Women’s University Professor Jachym Novak, Vysoka Skola Sronjni a Tectilffi Professor Isaac Porat, UMIST Professor Roy R Leitch, Heriot-Watt University Professor Ron Postle, The University of New South Wales Professor Gordon Wray, Loughborough University of Technology Professor Witold Zurek, Lodz Technical University Thank you all subscribers, authors, editorial board members, referees, publishing team, colleagues and students for your support and note that my address for correspondence is: School of Textiles, Heriot-Watt University, Netherdale, Galashiels, Selkirkshire, TD1 3HF, Scotland, E-mail:
[email protected] George K. Stylios Editor-in-Chief
Research register Alcoy-Alicante, Spain Aitex, Asociacion de la Investigacion de la Industria Textil, P1, Emilio Sala, 103801 Alcoy-Alicante Tel: +0034 965 5442200; Fax: +0034 9655442200; E-mail:
[email protected] Chemical Laboratory Reyes Botı´
Near infrared analyzing method for properties of feathers and down: Feathers and Down Craf-1999-71488 Other partners: De Vries Holland Feathers and Down B.V. (Coordinator) Interplume, Ducky Dons Netherlan B.V. A. Molina and C. Spa Finnish Feather Naturtex TNO Academic Industrial None None Project started: 1 September 2002 Project ends: 31 August 2004 Finance/support: 331.150e Source of support: European Commission Keywords: NIRA, Feathers and down A fast and objective measurement technique has been identified that can help the European down and feather industry to reduce costs by decreasing the testing time and increase productivity. The implementation of this new testing method (NIRA) will represent technology transfer from the NIR industry to a more traditional testing environment. Thereby, its use will help to improve the working conditions in test laboratories in the down and feather industry and will lead to a marked reduction in chemical use. The net prospective cost savings to the European down and feather industry will be about 45 million euros. European down and feather products can be put in a competitive position and therefore imports from Far East countries can be
Research register
5
IJCST 15,6
6
blocked. The new testing method is expected to be commercially implemented 12 months after the completion of the project. Project aims and objectives Industrial objectives and targets (1) To provide a reliable, fast and easy to operate method for determining the most important quality aspects of feather and down at the same time: . content of feather and down (percentage), species of the material. . fat content (percentage). (2) To be able to determine the quality aspects before production starts and during the production, with a method robust enough and suitable to apply in the typical conditions in feather and down plants. (3) To better control our own production quality. (4) To develop an analysing method that can be commercially available for other feather and down companies, and will be paid back in 1 or 2 years at most. (5) To implement the developed technique into the existing labelling standards (e.g. EN 12934). Technical objectives (1) A quick availability of the results less than a few minutes. (2) Determination of the feather and down ratio with a deviation of less than 5 per cent. (3) Undoubtedly distinguish the species of the feather and down material. (4) Determination of the fat content with a deviation of less than 0.1 per cent. (5) A procedure for carrying out the measurements, including a working description. (6) The libraries of NIR-spectra and data of samples with known content of feather and down, species and fat content. These libraries are the essential part of the technique to be able to measure the desired parameters. (7) The feather and down samples can be used as reference material for future work in improving or adapting the technique or for other purposes. This can save time and money involved with the preparation of the samples. (8) A draft proposal for incorporating the technique into the existing standard methods will be set up and that may be evaluated by concerning parties like the companies and the European Down and Feather Association (EDFA). Economic objectives (1) Reduction in production losses because of poor quality. (2) Reduction of the time and cost needed to carry out testing.
(3) New European suppliers of high quality feather and down can be established, because the import from Asian countries will be reduced. This will generate an extra turnover of e 10 million. (4) Increase of the competitiveness of the European feather and down industries as a whole. Research deliverables (academic and industrial) . Report on latest developments; . Samples of feather and down material to build up the library; . Measuring techniques (feather and down content, species, fat content); . Spectra library; . Measuring procedure and pilot demonstration; . Technical evaluation and cost-benefit analysis; . Progress report about draft proposal; . Business plan. Publications None
Alcoy-Alicante, Spain Aitex, Asociacion de la Investigacion de la Industria Textil, Pl, Emilio Sala, 103801 Alcoy-Alicante Tel: +0034 965 5442200; Fax: +0034 9655442200; E-mail:
[email protected] Chemical Laboratory Ana Carbonell
Optimization of cotton fabric processing for a flame retardant finish: G1ST-CT-2002-50270 Other partners: THOMOGLOU (Coordinator), ASTIR, LOUFAKIS, CLOTEFI, HABO, IFP, MOLTO, PYMAG, AITEX, LICOLOR, INOTEX, TECHNA Academic Industrial None None Project started: 1 March 2003 Project ends: 28 February 2005 Finance/support: 661.460e Source of support: European Commission Keywords: Flame retardant, Added value textiles, Safety
Research register
7
IJCST 15,6
8
The flammability of textile fibers is a setback to many end-use applications of textile materials. Textile fibers when subjected to sources of ignition can ignite and burn, creating risk for the consumer. Most textile fibers are flammable in air unless they have been modified chemically during fiber production or processing to render them flame retardant (FR). The end-uses for fibers span an enormous range of products and activities and in many of these flame retardancy is an important aspect of the product performance: flame retardancy is required to protect life and property. The need for FR textiles is clearly demonstrated by the statistical analysis of fire deaths and fatalities, related costs and loss of property. In the UK, each year, while 20 per cent of dwelling fires are caused by textiles – being the first material to be ignited – the percentage of deaths associated with these fires is increased up to 50 per cent. This disproportionate fatality emphasizes the need to develop successful FR systems for textile materials. The current range of available chemicals, which are used to impart FR properties to textiles, is mainly based on scientific developments during the 1950s and 1960s, when the emphasis laid mainly on providing durable ignition resistance at an affordable cost. More recently and alongside with the need for improved safety, issues such as environmental sustainability and toxicological properties of FR materials and handle and comfort properties of FR fabrics, have to be compromised. Project aims and objectives (1) Textile industry goals: . innovate and produce new added value products environmental friendliness and non-toxicity of FR chemicals, advanced fabric performance and quality and reengineered FR systems towards end use requirements; . enlarge current markets; and . anticipate niche markets. (2) Social goals: . improvement of health and safety; . protection of the environment; and . improvement of quality of life. Research deliverables (academic and industrial) . Technological, economic and environmental assessment of the existing FR products and processes. . Data on the evaluation of the quality performance of typical FR fabrics. . Guidelines for the development of new FR formulations. . Optimal operational finger prints on FR processes.
.
. . . . . . . . . .
. .
Detailed description of the new sequence of mechanical treatments to follow FR process. Data on FR behaviour of treated fabrics. Data on mechanical and comfort performance of FR fabrics. FR formulations. Environmental and economic assessment of FR formulations proposed. Guidelines for the full scale production. Technical training on the new systems proposed. Implementation reports. FR fabrics produced covering various end uses. Performance criteria according to various end uses. Specification profiles of FR fabrics according to various end uses and operational guidelines for their production. Financial statements. Exploitation policy.
Publications None
Bolton, UK Bolton Institute, Deane Road, Bolton, BL3 5AB, UK Tel: 01204 903108; Fax: 01204 399074; E-mail:
[email protected] Leah Higgins, Advanced Materials Research Center Research staff: Prof. S. C. Anand (Director of Studies), Dr D. A. Holmes (Supervisor), Dr M. E. Hall (Supervisor)
Effect of laundering on dimensional stability, distortion and other properties of cotton fabrics (Ph.D. Project) Other partners: Academic Industrial None Whirlpool Corporation, USA Project started: March 2000 Project ends: February 2003 Finance/support: $16,000 per annum Source of support: Whirlpool corporation Keywords: Cotton, Laundering, Dimensional stability, Wrinkling
Research register
9
IJCST 15,6
10
New cotton garments tend to shrink and distort when first laundered, particularly if they are tumble dried. Although this behaviour is often an inevitable consequence of the manufacturing and finishing process consumers tend to blame their laundering equipment. This project is the second Ph.D. project in an ongoing collaboration between Bolton Institute and Whirlpool Corporation. The underlying aim of this research is to gain a better understanding of how different laundering factors effect the levels of shrinkage, distortion and also wrinkling observed in new cotton fabrics on initial laundering. This knowledge will allow the project sponsors to develop laundering equipment that is less damaging to the appearance and dimensional stability of cotton fabrics. Publications Higgins, Anand, S. C., Hall, M. E. and Holmes, D. A. (2002a), ‘‘Factors during tumble drying that influence dimensional stability and distortion of cotton knitted fabrics’’, International Journal of Clothing Science and Technology (in Preparation). Higgins, L., Anand, S. C., Hall, M.E., Holmes, D.A. and Brown, K. (2001), ‘‘Effect of repeated laundering of dimensional stability and distoration of weft knitted cotton fabrics’’, IAT Conference Proceedings March 2001. Higgins, L., Anand, S.C., Holmes, D.A., Hall, M. E. and Underly, K. (2002b), ‘‘Rinse cycle softener and drying method and the effect of tumble sheet softener and tumble drying time’’, Textile Research Journal (in Preparation).
Bucharest, Romania The Research – Development National Institute for Textile and Leather, 16th Lucretiu Patrascanu Street, Sector 3, Bucharest, Romania Tel: 004 021 340 4921; Fax: 004 021 340 5515; E-mail:
[email protected] Testing, experienting of technologie, equipment, products Research staff: Nicula Gheorghe (Eng.)
Filtering products meant for reducing the impact of the industrial processes over the environment Other partners: Academic Industrial None None Project started: 1 June 2000 Project ended: 15 November 2002 Finance/support: 5,500 EUR Source of support: Ministry of Education, Research and Youth Keywords: Environment protection, Polluting phenomena, Man protection
The filtering products considered in this paper aim at solving the problems connected to the negative effect of the industrial processes over the environment. The industrial fields for which the filtering products have been mainly accomplished are: food, chemical, metallurgical, mining, as well as the fields connected to these, where there take place physico-chemical processes accompanied by state transformations or structural modifications, which generate polluting phenomena. There have been accomplished and experimented, for the first time in Romania, the following: . polyphenylsulphuric fibres, . PP filament yarns, and . spun and filament polyester yarns. As a consequence of the experiments, the following products have been tested and introduced into production: . Filter for waste waters – accomplished of PA6 yarns, 940 dtex count/140 f/70Z, meant for the mining industry and the concrete mixing plants. . Galvanic bath filter – accomplished of PP yarns, 640 dtex count, 124 f/120 Z, meant for the electrotechnical, metallurgical, chemical industry. . Filters meant for cooling emulsions for the ball bearing industry, accomplished of polyester yarns of the count 76/32 f 4 dtex in the warp and Nm 70/3 or Nm 40/4 in the weft. . Hot gas filters for the ferro – alloy industry, accomplished of polyester yarns of the count 76/32 f 4 dtex. Project aims and objectives . Designing, accomplishing, experimenting, and certifying of the filtering materials with special structures obtained by spinning, weaving, and special treatments by using fibres having special characteristics. . Ensuring the technical and qualitative level of the products in accordance with the requirements imposed by the European norms. . Employing the experience belonging to the specialists working with the organizations which are partners in the field of accomplishing and IPE testing new fabrics by means of performant technologies. . Developing the collaboration relationships with the potential end-users. Research deliverables (academic and industrial) None Publication The periodical Industria Textila (2001), No. 3, pp. 164-6.
Research register
11
IJCST 15,6
12
Bucharest, Romania The Research – Development National Institute for Textile and Leather, 16th Lucretiu Patrascanu Street, Sector 3, Bucharest, Romania Tel: 004 021 340 4921; Fax: 004 021 340 5515; E-mail:
[email protected] Textile mechanical processing Research staff: Doina Toma (Eng.), Professor Dr Eftalea Carpus (Eng.), Ileana Iorga (Eng.), Claudia Niculescu (Eng.)
Cold protection equipment Other partners: Academic Industrial None None Project started: 1 June 2000 Project ended: 15 October 2002 Finance/support: 17,500 EUR Source of support: Ministry of Education and Research Keywords: Heat insulation, Risk factor, Performance level, Protection individual equipment Working in low temperature environments and the contact with cold objects implies the beginning of energy transfer from the body towards the environment. During longer time period, transfer is affected by cooling the body and by overstraining of the physiological functions, possibly by reaching severe illness states: getting cold, freezing of the body extremities, and coma. The prevention of the above-mentioned symptoms makes the interdepositing of an insulating layer between the body and the environment necessary by using IPE that is used against cold. Based on the protection requirements and the minimum specified needed performance parameters, the following IPE types used against cold have been accomplished. For working indoors (1) Quilted costume made up of a blouse and trousers, containing three layers of material: . Outer layer. Woven fabric of 80 per cent cotton/20 per cent PA. . Intermediate layer. Non-woven of 80 per cent PES/20 per cent thermoadhesive fibres. . Interior layer. 100 per cent cotton woven fabric.
(2) IPE used against cold, made up of two layers: . Exterior layer. Costume made up of a blouse and trousers, accomplished of 99 per cent Rhovyl/1 per cent Bekinox; . Interior layer. Heatinsulating undergarment, containing a quilted blouse and quilted trousers, manufactured of 100 per cent cotton woven fabric, face – back and non-woven 100 per cent Rhovyl. For working outdoors (1) Quilted costume made up of a blouse and breastplated trousers, manufactured of three layers of material: . Exterior layer. Woven fabric of 100 per cent cotton/20 per cent PA. . Intermediate layer. Non-woven of 80 per cent PES/20 per cent thermoadhesive fibres. . Interior layer. 100 per cent woven fabric. The accomplished products, which were tested and certified according to the European Normatives, satisfy the essential requirements of safety and health corresponding to the envisaged utilization fields. Project aims and objectives . Designing, accomplishing, experimenting and certifying the IPE made up of functional and intelligent textile materials, having special structures obtained by spinning, weaving, needlepunching, manufacturing, by using fibres having special characteristics. . Ensuring the technical and qualitative level of the products according to the requirements imposed by the European Normatives. . Utilization of the experience of the specialists belonging to the partner organizations in the field of accomplishing and testing IPE, as well as the material basis, with a view to accomplishing new products by means of performant technologies. . Developing the collaboration relationships established with the potential beneficiaries. Research deliverables (academic and industrial) None Publication Toma, D., Carpus, E., Iorga, I. and Niculescu, C. (2003), ‘‘ Individual protection equipment used against cold ’’, Industria Textila, No. 2, p. 75, ISSN 1222-5347.
Research register
13
IJCST 15,6
14
Bucharest, Romania The Research – Development National Institute for Textile and Leather, 16th Lucretiu Patrascanu Street, Sector 3, Bucharest, Romania Tel: 004 021 340 4921; Fax: 004 021 340 5515; E-mail:
[email protected] The Department of Textile Product Testing, Control and Notification The Romanian enterprises manufacturing textile soil coverings, wishing to import and export, as well as producers from abroad having markets in Romania Research staff: Ramona Buriceanu (Eng.)
The developing of the methodologies for testing the textile soil coverings according to the European Norms EN 1307 and EN 1470 Other partners: Academic None
Industrial Industrial partners manufacturing fibres, yarns, woven fabrics, knits, textile soil coverings, technical articles, etc. Project ended: 15 December 2001
Project started: 21 January 2000 Finance/support: 20312,5 EURO Source of support: 96.08 per cent Budget 3.92 per cent cofinancing Keywords: Textile soil coverings, European norms, EN 1307, EN 1470, Quality
As part of the project, there has been effected the complex studying and analysing of all the quality parameters that characterize the textile soil coverings based on the standards EN 1307 and EN 1470, which can be applied to the textile soil coverings which are needle punched in flat form, knitted, conventional, woven and raised. The project has achieved the classifying of the textile soil covering utilizations by comfort, the aspect preserving and wear resistance, by way of establishing the significant technical parameters that arise in classifying the textile soil coverings. Fifty six normative references (national and international standards) have been evaluated and studied, the textile soil covering characteristics have been studied, and the results have been obtained based on the effected tests also have been implemented at the industrial partner. The activity carried out as part of the project allowed the tackling and the clarifying of the following technico-scientific aspects:
the description of the categories of textile soil coverings (depending on mass and pile thickness); . establishing the identifying requirements and integrating tolerances of the characteristics; . establishing the basic and supplimentary requirements with a view to the classification of the textile soil covering by wear (integrating into 1, 2, 3, 4 classes); . establishing the classification requirements of the textile soil coverings by aspect; and . establishing the classification requirements and classification categories LC1, LC2, LC3, LC4, LC5 of the textile soil coverings by comfort. The results of the tests have been presented and analysed with the specialists of the partner for the purpose of finding out the technological solutions of integrating into the imposed quality classes, of improving the technical performances of the products and of increasing the enterprise competitiveness at the European level. .
Project aims and objectives . Creating a coherent system of qualitative evaluation of the textile soil coverings, based on the European Standards EN 1307 and EN 1470. . Classifying the textile soil coverings utilization by comfort, aspect and depending on wear resistance. . Creating the technico-scientific basis that is needed for certifying the products belonging to the range of textile soil coverings. . Facilitating the textile soil covering exportation on the EU market by way of creating the conditions for mutual acknowledging of the INCDTP laboratory tests with the similar EU organisms. Research deliverables (academic and industrial) . Strengthening the position on the internal and international market of the enterprises manufacturing textile soil coverings will lead to the increase in the competitional capacity of the textile products, having in view the forming of the unique European market. . Increasing the product quality and competitiveness. . Aligning to the normative European requirements regarding the classification of the textile soil coverings. . Increasing the qualification degree by acquiring new knowledge concerning the covering classification. Publications None
Research register
15
IJCST 15,6
16
Bucharest, Romania The Research – Development National Institute for Textile and Leather, 16th, Lucretiu Patrascanu Street, Sector 3, Bucharest, Romania Tel: +40-021-340.49.28; Fax: +40-021-340.55.15; E-mail:
[email protected] The Department of Laboratory Equipment and Apparatus Meant for the Textile Industry Romanian Research and Education Ministry Research staff: Daniela Isar (Senior Researcher), Virgil Motocu (Senior Researcher), Cristian Jipa (Senior Researcher)
Ecological textile technical articles as asbestos substitutes for the industry; equipment for accomplishing these Other partners: Academic Industrial None Fermit SA – Ramnicu Sarat Enterprise Project started: 20 December 1999 Project ended: 29 January 2003 Finance/support: 33,000 EUR Source of support: Governmental budget, The Research-Development, National Institute for Textile and Leather Keywords: Asbestos – free yarns, Asbestos substitute yarns, Technical textiles, Composite materials, Eco-technologies, Environmental health, Quality of life The asbestos extensive use is the cause of severe damages both for the human health and the environment. The cancerous character of this mineral, under whose generic title there are included many silicates, is a worldwide acknowledged fact. During the past 15 years, laws have been enacted to prohibit the use of asbestos. In recent years, sustained efforts have been made to find substitutes for asbestos. The present project aimed at accomplishing of the technology and a specialized machine meant for obtaining a composite cord as an asbestos substitute, meant for producing the automotive clutch disks. This technical article is made up of versions of three components: textile yarns, glass yarns and metal yarns. The machine does the twisting – cabling of the three yarn components with variations of the winding parameters, so that it would allow obtaining of the characteristics needed for the respective technical norms imposed to the clutch disks.
The stages applied as part of the project were the following: . establishing the working conditions specific for the automotive friction elements; . textile and non-textile raw material selection based on which asbestos substitute composite structures can be produced; . technology for accomplishing versions of asbestos substitute composite structures; . design and accomplishing the necessary machines meant to obtain these composite structures; and . instrumental techniques for investigating and characterizing the obtained products. Project aims and objectives The project refers to the obtaining of the technology and the machine necessary to achieve the asbestos substitute composite yarn, meant to accomplish the friction elements in the automotive industry. Research deliverables (academic and industrial) . Technology of accomplishing the asbestos substitute composite yarn meant for producing automotive clutch disks. . Project and equipment meant for accomplishing the asbestos substitute composite yarn. . The production of textile technical articles: asbestos substitute composite yarn meant for producing automotive clutch disks. Publications The Romanian periodicals: Industria Textila (2000), No. 2, p. 83. Industria Textila (2003), No. 2, p. 145.
Bucharest, Romania The Research – Development National Institute for Textile and Leather, 16th Lucretiu Patrascanu Street, 3rd Sector, Bucharest, Romania Tel: +40(021)340.49.28; Fax: +40(021)340.55.15; E-mail:
[email protected] Special Products Romanian Research Ministry Research staff: Claudia Niculescu (Senior Researcher), Florica Ionescu (Senior Researcher)
Research register
17
IJCST 15,6
18
Procedure to obtain Neoprene microcellular backed plates. The design and the achieving of the technology of making-up the professional suit for independent divers Other partners: Academic None
Industrial ARECA S.A. Enterprise – Bucharest ARGOS S.A. Enterprise – Cluj-Napoca
Project started: 1992 Finance/support: 13,000 e Source of support: Governmental Budget Keywords: Composite materials, Safety life, Procedure microcellular plates, No conventional technology The project consisted in achieving of a procedure to obtain Neoprene backed microcellular plates on textile support destined for making the professional suits for independent divers and designing the suit. The stages of project were: . achieving of a procedure to obtain Neoprene microcellular plates covering both sides with Lycra knit; . designing the suit according to the anthropometric sizes of the male population of Romania; . testing the performances: thickness and mass, comfort at wear, handle at dressing and undressing, thermal isolation degree at diving under water of depth of 3-51 m and temperatures between 4 and 25 C; . homologation of the prototype. The obtained results shall be applied for obtaining the other new products such as: life saving suits, mountings for cars, splits for emergency hospitals and rescue centers, thermal isolations for pipes, phonon – absorbent upholstery, suits for surfing, professional suits for ski. Project aims and objectives The projects aim was to design and develop a professional wet suit for divers with high characteristics of thermal isolation, resistance and elasticity. The project objectives were as follows: . designing and achieving of a procedure to obtain Neoprene microcellular plates on textile support destined for making the professional suits for independent divers; . designing and manufacturing the prototype of suit according to the anthropometric sizes of the male population in Romania; and . homologation of the suit.
Research deliverables (academic and industrial) . Technical documentation and blue prints. . Technology of obtaining Neoprene microcellular plates coating on both sides with Lycra knit. . Establishing the making-up technology of suit. . Prototype of suit. Publications None
Bucharest, Romania The Research – Development National Institute for Textile and Leather, 16th Lucretiu Patrascanu Street, 3rd Sector, Bucharest, Romania Tel: +40(021)340.49.28; Fax: +40(021)340.55.15; E-mail:
[email protected] Special Products Romanian Research and Education Ministry Research staff: Claudia Niculescu (Senior Researcher)
Parachutists’ Life Preserver Other partners: Academic None
Industrial ARECA S.A. Enterprise – Bucharest STINGO – SOMET S.A. Enterprise – Buzau
Project started: 1997 Finance/support: 16.300 e Source of support: Governmental Budget Keywords: Composite materials, Textile materials, Life safety, No conventional technology, Inflatable equipment The projects intend to design and develop a technology for manufacturing a parachutists’ life preserver to be worn when the parachutist is undertaking a descent over or near the sea. The life preserver, in one size, consisting in a waistcoat with inflatable stole, inflation equipment and survival equipment. The life preserver was conceited to be worn over the parachutists’ normal clothing and equipping before the parachute harness, the reserve parachute and any suspended load.
Research register
19
IJCST 15,6
20
The life preserver was manufactured of a cotton gabardine and a polyester webbing harness. The stole is inflatable from manually operated CO2 cylinder or an oral inflation tube and valve. The survival equipment consists of whistle and a life line. The stages of project were: . research, selection and achievement of the materials, . selection of the survival equipment, . design of the waistcoat, . design of the inflatable stole, . design of the inflatable equipment, . achievement of the life preserver, and . testing the performances of the life preserver. Project aims and objectives The projects aim was to design and develop a technological process for a Parachtists’ Life Preserver. The project objectives are as follows: . designing the parachutists’ life preserver, . designing the CO2 cylinder and the operating head, . achievement of the parachutists’ life preserver, and . salvage the life of parachutists. Research deliverables (academic and industrial) Technical documentation and blue prints for waistcoat, CO2 cylinder, operating head and inflatable stole. Manufacturing technological process. Prototype of parachutists’ life preserver. Publications None
Bucharest, Romania The Research – Development National Institute for Textile and Leather, 16th Lucretiu Patrascanu Street, 3rd Sector, Bucharest, Romania Tel: +40(021)340.49.28; Fax: +40(021)340.55.15; E-mail:
[email protected]
Special Products Romanian Research and Education Ministry Research staff: Claudia Niculescu (Senior Researcher)
Research register
The flying and survival suit on the sea Other partners: Academic None
Industrial ARECA S.A. Enterprise STINGO-SOMET S.A. Buzau Enterprise
Project started: 1997 Finance/support: 38.700 e Source of support: Governmental budget Keywords: Sea rescue equipments, Composite materials, No conventional technology, Hypothermia protection, Inflatable equipment The projects intend to design and develop a technology for manufacturing an individual equipment for helicopter aircrew that fly over or near the sea. The equipment consists of a waterproof suit, a thermal insulated suit and a survival west. The stages of project were: . research, selection and manufacturing the materials; . designing the waterproof suit, thermal insulated suit, survival west, inflatable stole, CO2 cylinder and the operating head; . manufacturing the suit; . testing the performances of the suit: waterproof at a pressure of 2,000 mm water column, physiological comfort, resistant at high temperature, thermal isolation, buoyancy for 80 daN, time of filling of pneumatic cushion pillow with liquefied CO2, possibility to localize the shipwrecked person with light and acoustic signals, possibility to recover the shipwrecked person with the force system. Project aims and objectives The projects aim was to design and develop a technological process for an individual equipment for helicopter aircrew. The project objectives are as fallows: . designing of a suit with a view to protect the life of helicopter aircrew in situation of a flay accident over the sea; . development and manufacturing the suit; and . homologation of the suit.
21
IJCST 15,6
Research deliverables (academic and industrial) . Technical documentation and blue prints. . Manufacturing technological process. . The flying and survival suit for helicopter aircrew
22
Publication Hight Performance Textile (2001), ‘‘Flaying and survival suit for marine use’’, p. 151-7.
Bucharest, Romania The Research – Development National Institute for Textile and Leather, 16th Lucretiu Patrascanu Street, 3rd Sector, Bucharest, Romania Tel: +40(021)340.49.28; Fax: +40(021)340.55.15; E-mail:
[email protected] Special Products Romanian Research Ministry Research staff: Doina Grecu (Senior Researcher), Claudia Niculescu (Senior Researcher), Radu Radulescu (Senior Researcher)
Personnel parachute system for paratroopers with backward attachment Other partners: Academic Industrial None AEROSTAR S.A. Bacau Enterprise Project started: 1992 Finance/support: 50.000 e Source of support: Governmental budget Keywords: Aeronautic accessories, Aerodynamic studies, Parachute systems, Paratroopers deployment The projects consisted in designing and achieving of the parachute system (main and the reserve parachute) has dorsal attachment. It is conceived for paratroopers launching at low altitude and with forced opening for the main parachute. The system was conceited for low altitude launching, with forced opening for the main parachute and hand opening for the reserve parachute. The stages of project were:
. .
. . .
.
aerodynamic calculation of the main parachute and the reserve parachute, designing of the parachutes, riser system, parachutes’ pack and harness assembly, manufacturing of the parachute system, manufacturing technological process, testing and evaluation of the performances: deployment and inflation, stability, rates of descent and minimum deployment altitude for main and reserve parachute, maximum horizontal rate and maximum horizontal rate for main parachute and homologation of the prototype.
Project aims and objectives The aim of the projects was to design and manufacture of a back attached parachute system assembly for preparing, training and delivering into battle the parachute troopers and their specific weaponry. The objective was: . designing of a parachute system for equipping the paratroopers, . manufacturing the parachute system, and . homologation of the system. Research deliverables (academic and industrial) Parachute documentation and blue prints. . Manufacturing technological process. . Manufacturing of a parachute system (main parachute, reserve parachute, riser system, parachute pack and harness assembly). Publications None
Bucharest, Romania The Research-Development National Institute for Textile and Leather, 16th Lucretiu Patrascanu Street, Sector 3, Bucharest, Romania Tel: 004 021 340 4921; Fax: 004 021 340 5515; E-mail:
[email protected] Departament of Medical Article Research Alexandra ENE (Eng.) Research staff: Carmen Mihai, (Eng.) Adriana Petrescu, Maria Bulearca (Techn.)
Research register
23
IJCST 15,6
24
Technologies for accomplishing a new generation of textile structures meant for implantable medical articles having utilization in cardiovascular surgery Other partners: Academic Industrial The Research-Development None National Institute for Chemicals and Pharmaceuticals, Ministry of Health and Family Project started: 30 September 2000 Project ended: 30 September 2002 Finance/support: 23, 000 EUR Source of support: 70 per cent governamental, 30 per cent own Keywords: Implantable products, Weaving technologies, Cardiovascular surgery The identification of the biological requirements of the most complicated mechanism, that is the human body, correlated with the possibilities of the current technique, has enabled the successful achievement, in Romania, for the first time, by a team of specialists from The Research-Development National Institute for Textile and Leather, of a new generation of surgical implants meant for the cardiovascular surgery. The accomplishing of new vascular prosthesis types has imposed a high precision degree in the making of a tube with diameter included, so as to coincide with that of the blood vessel with which it is to be coupled. In this respect, for the accomplishing of these products, the following aspects were aimed at: . choosing the weaving model, . adequate choice of the yarn length density, . exact establishing of the textile backing density, . establishing of the yarn number in weft, and . establishing of the internal diameter. To design the fabrics destined for the vascular prostheses, the minimal requirements have been considered for the biofunctional characteristics imposed by the clinical usage field as a consequence of which there resulted the following: . the product geometry imposes the use of the tubular structure and the maximal work-width, and . the impermeability imposes main parameters of fabric designing (achieving of a structure for which untwisted yarns are used, the densities of the two systems and the product mass, etc.)
Project aims and objectives . Accomplishing of a new generation of products based on textile materials having special structures, achieved through weaving technologies, meant for cardiovascular surgery as vascular prostheses and textile patches. . Assurance of the technical and qualitative level, having in view the physico-chemical, physico-mechanical, biological and microbiological characteristics in accordance with the European norms in force. . Implementation of the GMP procedures in the manufacturing of the products. Research deliverables (academic and industrial) The research results are meant for the following fields: . as sterile products in sanitary network (hospitals and clinics), . as didactic material (teaching aids) for universities having chemical, pharmaceutical, biological and medical profiles, and . as technical documentation for final producers. Publications New generation of vascular prostheses obtained through weaving technologies (2003), Medtex’ 03 International Conference, Bolton, UK. The diagnosis of the implantable product sector meant for cardiovascular surgery (2002), Revista Industria Textila, No. 3, pp. 14-8.
Bucharest, Romania The Research-Development National Institute for Textile and Leather, 16th Lucretiu Patrascanu Street, Sector 3, Bucharest, Romania Tel: 004 021 340 4921; Fax: 004 021 340 5515; E-mail:
[email protected] Information – Automation Research staff: Dr Emilia Visileanu (Eng.), Carmen Ghituleasa (Eng.), Mihai Stan (Math.)
Informatic system of simulating the textile industry processes (spinning and weaving) Other partners: Dr Tudor Sireteanu (Math.), Virgil Mitre (Eng.), Gheorghe Ghita (Eng.) Academic Industrial S.C. The Institute for the Mecha-Nics of Solids POSTAVARIA Bucharest ROMANA S.A.
Research register
25
IJCST 15,6
Project started: January 2000 Project ended: December 2002 Finance/support: 25.000 EURO Source of support: MEC program Relansin Keywords: Informatic system, Simulation, Spinning, Weaving
26
The informatic product can be applied to the promotion of the new products and it eliminates the preliminary experimental stages of establishing the technological parameters and the characteristics of these. The achieved informatic product (one for the spinning process, and the other for the weaving process, respectively) is made up of a main menu having five options: data bases, price simulation, help, exit. The data base contains the elements associated with a previous experience that included data regarding the raw material, equipment and the characteristics of the already accomplished products. The ‘‘simulation’’ component. In the case of spinning, there is a calculated spinnability index that allows the selection of a spinning plan based on the technological parameters necessary to accomplish the new product. As part of this option, too, there is the calculated break resistance of the yarn that is going to be produced (the Ning Pan algorithm). Within the weaving process, the activation of the ‘‘simulation’’ option allows the establishment of the creation elements, colour range, the warp and weft yarn number, when previously, by a selection procedure, the technological parameters corresponding to the new woven fabric, etc. have been established. The price option achieves the calculations of an economic nature concerning the accomplishing of the new product (yarn or woven fabric). The help option helps the utilizer in using the informatic product. Project aims and objectives . Accomplishing an informatic product having a technological applicability. . Modernizing of the designing systems in the spinning and weaving mills. . Reducing of the costs of promoting the new products (time and labour force). Research deliverables (academic and industrial) Informatic product that can be used for establishing the spinning plans from the wool and wool type spinning mills. Informatic programme of establishing the characteristics of designing the woven fabrics (structure and colour). Publication The Periodical Magazine (2002), Industria Textila, No. 3, pp. 153-6.
Bucharest, Romania The Research-Development National Institute for Textile and Leather, 16th Lucretiu Patrascanu Street, 3rd Sector, Bucharest, Romania Tel: +40(021)340.49.28; Fax: +40(021)340.55.15 ; E-mail:
[email protected] Special Products Romanian Research and Education Ministry Research staff: Claudia Niculescu (Senior Researcher), Doina Grecu (Senior Researcher), Alexandra Ene (Senior Researcher)
Developing of manufacturing technologies for new types of textile and plastic medical articles for emergency purposes Other partners: Academic None
Industrial IZOLATORU S.A. Enterprise ARECA S.A. Enterprise Project ended: 2002
Project started: 2000 Finance/support: 26.000 e Source of support: Governmental budget Keywords: Medical articles, Emergency service, Human health, Manufacturing technology
The projects intend to develop a technology for manufacturing the emergency medical articles such as: splints for arm and leg, arm and forearm holder, extrication spine splint, vacuum mattress and transfer sheet. These products are intended for the use of emergency hospitals, mountain rescue services, civil protection, extrication services and ambulance services. These products were manufactured with textile materials (polyamide, polyesters), plastic materials (PVC) and rubber (neoprene). Each of these medical articles is conceived with a precise destination such as: arms and legs immobilization, spine immobilization, arms sustain, injured people transport etc. The stages of project were: . research, selection and achievement of the materials, . design of the emergency articles, . elaboration of the manufacturing technology for the emergency articles, . manufacturing of the prototypes,
Research register
27
IJCST 15,6
.
.
28
testing the performances: physical-chemical, biological, functional and clinical, and homologation of the products.
Project aims and objectives The projects aim is to develop a technology for manufacturing the emergency medical articles. The project objectives are as follows: . development of non-conventional assembly technologies for the subassembly of the medical articles with textile structures, . development of technology to manufacture the pellicle fabric for the joints through HF welding, . development of manufacturing technologies of the medical articles for external use depending on the quality and technical requirements, . design of medical articles for external use, . manufacture of medical articles for external use (splints for arm and leg, arm and forearm holder, extrication spine splint, vacuum mattress, transfer sheet), . developing technical specification papers, and . physical-chemical, biological, functional and clinical testing. Research deliverables (academic and industrial) (1) Technical documentation and blue prints. (2) Manufacturing technological process. (3) Prototypes: . splints for arm and leg, . arm and forearm holder, . extrication spine splint, . vacuumed mattress, and . transfer sheet. Publication Textile Industry Magazine (2002), ‘‘ Medical emergency articles for external use ’’, No. 2, pp. 88-93.
Budapest, Hungary Budapest University of Technology and Economics, H-1521 Budapest, Hungary Tel: +36-1-463-1376; Fax: +36-1-463-1376; E-mail:
[email protected]
Department of Plastics and Rubber Technology Prof. Judit Borsa
Research register
Development of special properties on cotton cellulose Other partners: Academic Industrial Budapest University of Technology None and Economics – Department of Chemical Engineering, Department of Chemical Technology, Department of General and Analytical Chemistry, Department of Organic Chemical Technology, and Department of Physical Chemistry. Hungarian Academy of Sciences, Chemical Research Center – Institute of Chemistry, and Institute of Isotope and Surface Chemistry. Bay Zoltan Institute for Materials Science and Technology, Hungary Johan Bela National Center of Epidemology, Hungary Johannes Kepler University, Linz, Austria Cornell University, Ithaca, NY, USA Project started: 1 January 2001 Project ends: 31 December 2004 Source of support: OTKA (Hungarian National Science Fund), NKFP (Hungarian National Research and Development Fund) Keywords: Cotton, Cellulose, Swelling, Quaternary ammonium hydroxide, Mercerization, Chemical modification, Carboxymethylation, Soil release, Protective clothing, Hospital infection, Medical textile, Antimicrobial textile New properties of cotton fiber are developed by physical and/or chemical modification. Structure and morphology of modified fiber are studied (1 and 2), furthermore, research on functional textiles is on-going (3). (1) Swelling of cotton cellulose. Quaternary ammonium hydroxides are intracrystalline swelling agents of cellulose, moreover, the large molecules can even dissolve it. Effect of tetramethylammonium hydroxide (TMAH) on the properties of cotton cellulose (degradation, supermolecular structure, morphology, degree of mercerization) and on
29
IJCST 15,6
30
purification of cellulose (retting of hemp, scouring of cotton) is studied in comparison with sodium hydroxide. (2) Chemical modification of cotton cellulose. 2-3 per cent of hydroxyl groups of cotton cellulose are substituted by bulky carboxymethyl groups. Modified fiber retains its fibrous nature while many of its properties differ from those of the original fiber. Some properties of fiber (in this project mainly accessibility, hydrophyl character, swellability) as a function of reaction parameters are studied. (3) Functional textiles. . Soil release: in a former project partially carboxymethylated cotton fabric was effectively used to limit the dermal exposure of workers handling pesticide. The modified fabric has dual effect: it entraps large amount of pesticide during contamination and releases its large ratio during washing. In this project partial carboxymethylation as a durable soil release finishing is studied. . Protective clothes/Antimicrobial textiles: cationic antimicrobial agent is bounded on the anionic groups of cellulose fiber. The survival time of some microorganisms (Candida albicans, Staphylococcus aureus) on this fabric is studied. Project aims and objectives Cotton is the most commonly used fiber in clothing, moreover, it is an excellent model to study cellulose itself. Cellulose, as a renewable raw material has a special significance in the sustainable development, hence any information on its possible improvement can be interesting for areas outside textiles as well. The aim of the project is to modify cellulose by physical and/or chemical methods to change its properties. It can be important both from the scientific and practical point of view. The project has three main objectives: (1) physical modification by swelling with quaternary ammonium hydroxide, (2) physical and chemical modification by carboxymethylation, and (3) functional textiles. . Swelling. TMAH seems to be a special activating agent of cellulose: due to its slightly apolar character it can penetrate into the apolar parts of the cellulose structure, too. Scientific literature on its swelling effect is very limited. The aim of the project is to obtain the basic information on the effect of TMAH on the structure and morphology of cellulose and on purification of cellulose sources (hemp and cotton) in comparison with sodium hydroxide. . Chemical modification. Substitution of hydroxide groups by bulky carboxymethyl groups is a good tool to loosen the ordered structure
.
of cellulose. Carboxymethyl cellulose of high degree of substitution (DS , 0.6) is soluble in water. Introducing some carboxymethyl groups (DS , 0.08. . .0.1) into cotton cellulose can significantly change the properties of fiber. The aim of the project is to characterize the cotton cellulose modified by various technologies, and to study possible applications of modified fiber. Functional textiles. New, valuable properties of cotton fabric can be developed by proper modification. The project focuses on two topics: durable soil release finishing and antimicrobial textiles/protective clothes (hospital textiles protect people from infection, but according to international statistical data, they are also carriers of micro-organisms; the aim of the project is to reduce the survival time of the micro-organisms on the fabric as a part of the strategy against nosocomical infection).
Research deliverables (academic and industrial) Academic deliverables . Swelling with TMAH: important information can be obtained about the activation of cellulose by a slightly apolar swelling agent. It can be a starting point for further research on special modifications of cellulose. . Chemical modification: especially, the structure of the amorphous phase might be interesting. . Functional textiles: study of the soil release from the carboxymethylated cotton fabric can help in the further understanding of fiber swelling and highly anionic fiber surface on the release mechanism. Industrial deliverables Results might be applied in textile finishing (durable soil release, antimicrobial textile), and possibly in purification (delignification) of cellulose. Publications Borsa, J., Racz, I. and Toth, T. (2002), ‘‘Chemical modification of cotton cellulose for medical textiles’’, 2nd AUTEX Conference, Brugges, Belgium. Borsa, J., Toth, T. and Takacs, E. (2003), ‘‘Radiation modification of swollen and chemically modified cellulose’’, Rad. Phys. Chem., Vol. 67, pp. 509-12. Borsa, J., Toth, T., Racz, I. and Balkan, D. S. (2002a), ‘‘The role of anionic group and improved accessibility of chemically modified cellulose in adsorption and release of cationic molecules’’, 1st International Cellulose Conf., ICC 2002, Kyoto, Japan. Borsa, J., Tanczos, I., Pokol, Gy., Toth, T. and Schmidt, H. (2002b), ‘‘The effect of tetramethylammonium hydroxide in comparison with the effect of sodium hydroxide on the slow pyrolysis of cellulose’’, Pyrolysis 2002, Leoben, Austria. Borsa, J., Toth, T., Takacs, E., Sajo, I. and Tanczos, I. (2002c), ‘‘Activation of cotton cellulose by tetramethylammonium hydroxide’’, 1st Int. Cellulose Conf., ICC 2002, Kyoto, Japan. Borsa, J., Zala, J., Kiss, K., Lazar, K., Toth, T. and Horvath, E. (2003), ‘‘Antimicrobial cotton fabric for hospital use’’, 2nd European Conf. on Protective Clothes, Montreux, Switzerland.
Research register
31
IJCST 15,6
32
Obendorf, S. K. and Borsa, J. (2001), ‘‘Lipid soil removal from cotton fabric after mercerization and carboxymethylation finishing’’, J. Surfactants and Detergents, Vol. 4, pp. 247-56. Tanczos, I., Borsa, J., Sajo, I., Laszlo, K., Juhasz, Z.A. and Toth, T. (2000), ‘‘Effect of tetramethylammonium hydroxide on cotton cellulose in comparison with sodium hydroxide,’’, Macromolecular Chem. Phys., Vol. 201 No. 17, pp. 2550-6. Tanczos, I., Pokol, Gy., Borsa, J., Toth, T. and Schmidt, H. (2003), ‘‘The effect of tetramethylammonium hydroxide in comparison with the effect of sodium hydroxide on the slow pyrolysis of cellulose’’, J. Analytical and Applied Pyrolysis (in press). Toth, T., Borsa, J. and Tanczos, I. (2002), ‘‘Equilibrium swelling of cotton cellulose in tetramethylammonium hydroxide’’, 10th Oesterreichische Chemietage, Linz, Austria. Toth, T., Borsa, J., Takacs, E. and Sajo, I. (2003), ‘‘Effect of preswelling on radiation of cotton cellulose’’, Rad. Phys. Chem., Vol. 67, pp. 513-15. Toth, T., Borsa, J., Reicher, J., Sallay, P., Sajo, I. and Tanczos, I. (2003), ‘‘Mercerization of cotton cellulose with tetramethylammonium hydroxide’’, Textile Res. J., Vol. 73 No. 3, pp. 273-8.
Budapest, Hungary Budapest University of Technology and Economics (BUTE), Budapest, M} uegyetem rkp. 3., H-1111 Hungary Budapest, H-1521, Hungary Tel: +36-1-463-1376; Fax: +36-1-463-1376; E-mail:
[email protected] Prof. Judit Borsa Department of Plastics and Rubber Technology
Effect of quaternary ammonium compounds on structure and reactivity of cellulose Other partners: Academic Industrial Department of Analytical Chemistry, None BUTE Department of Physical Chemistry, BUTE Department of Chemical Technology, BUTE Chem. Research Center of the Hungarian Academy of Sciences Johannes Kepler University, Linz, Austria Dr Habil Ildiko Tanczos Project started: 1 January 1999 Project ended: 31 December 2002 1 January 2002 (the project will be continued in 2003) Finance/support: Euro 15,000, Euro 5,000
Source of support: Hungarian National Research Fund and Austrian-Hungarian Scientific Exchange Program Keywords: Cellulose, Cotton, Tetraalkylammonium compounds, Tetramethylammonium hydroxide, Sodium hydroxide, Swelling, Mercerization
Research register
Quaternary ammonium compounds are intracrystalline swelling agents of cellulose, moreover, they are also its good solvents in the case of sufficiently large size of molecule. Scientific literature on the effect of tetraalkylammonium hydroxides on cellulose is very limited, partly due to the relatively high price of these chemicals. Tetramethylammonium hydroxide has recently been applied in the electronic industry for surface cleaning, hence its price has significantly decreased. Some more information about the interaction of cellulose with these swelling agents, compared to sodium hydroxide, could be interesting both from a scientific and technological point of view.
33
Project aims and objectives The aim of the work is to study the interaction of tetramethylammonium hydroxide (TMAH), the smallest member of the tetraalkylammonium hydroxide family, with cotton cellulose. Crystallinity, sorption capacity, water retention, dye uptake, effect of high energy irradiation etc. were investigated. Purification of various cellulose sources (wood, hemp, cotton) was also studied. Research deliverables (academic and industrial) It was found that TMAH is a more effective swelling agent of cellulose than sodium hydroxide. It was explained by its large size, partly apolar character, and extremely high activity. This property of TMAH might be used in various areas including textile industry. Publications Borsa, J., Ta´nczos, I., Sajo´, I., Juha´sz, Z.A. and To´th, T. M. (1999), ‘‘Activation of cellulose with tetramethylammonium hydroxide’’, Advances in Wood Chemistry, International Symposium (Proceedings), Wien, Austria. Taka´cs, E., Wojna´rovits, L., Fo¨ldva´ry, Cs., Borsa, J. and Sajo´, I. (2001), ‘‘Radiation activation of cotton cellulose prior to alkali treatment’’, Res. Chem. Intermediates, Vol. 27, pp. 837-45. Ta´nczos, I., Putz, R. and Borsa, J. (1999), ‘‘Comparative study on the effects and mechanism of the new quatam pulping’’, 10th International Symposium on Wood and Pulping Chemistry, Main Symposium (Proceedings), Yokohama, Japan, Vol. II, pp. 288-91. Ta´nczos, I., Borsa, J., Sajo´, I., La´szlo´, K. and Juha´sz, Z.A. (1998), ‘‘Comparison of the effect of sodium hydroxide and tetramethylammonium hydroxide on cotton cellulose’’, International Symposium in Wakayama on Dyeing and Finishing of Textiles (Proceedings), Wakayama, Japan, pp. 276-7. Tanczos, I., Borsa, J., Sajo, I., Laszlo, K., Juhasz, Z.A. and Toth, T.M. (2000), ‘‘Effect of tetramethylammonium hydroxide on cotton cellulose in comparison with sodium hydroxide’’, Macromolecular Chemistry and Physics, Vol. 201 No. 17, pp. 2550-6. ´ Toth, I., Borsa, J., Reicher, J., Sallay, P., Sajo´, I. and Tanczos, I., ‘‘Mercerization of cotton with tetramethylammonium hydroxide’’, Textile Research Journal (in preparation).
IJCST 15,6
34
Budapest, Hungary Budapest University of Technology and Economics (BUTE), Budapest, Mu¨egyetem rkp. 3., H-1111, Hungary, Budapest, H-1521, Hungary Tel: +36-1-463-1376; Fax: +36-1-463-1376; E-mail:
[email protected] Prof. Judit Borsa, Department of Plastics and Rubber Technology
Advanced textiles Other partners: Academic Industrial Chem. Research Center of the None Hungarian Academy of Sciences Bay Zolta´n Institute for Materials Science and Technology Ilona Ra´cz Ph.D. Cornell University, Ithaca, New York, USA Prof. S. Kay Obendorf, Johan Be´la National Center of Epidemiology Project started: 1 January 2001, Project ends: 31 December 2004, 1 July 2001 31 July 2004 Finance/support: Euro 25,000, Euro 30,000 Source of support: Hungarian National Research Fund, Hungarian National Research and Development Fund Keywords: Cellulose, Cotton, Chemical modification, Carboxymethylation, Pesticide protective clothes, Soil release, Medical textile, Antimicrobial textile Supermolecular structure and morphology of cellulose can significantly be modified by chemical modification. Slight carboxymethylation of cotton cellulose improves the accessibility of the fiber, which can be used for various purposes. Project aims and objectives The aim of the work is to find useful applications for a fiber with very high accessibility (sorption capacity). Slight carboxymethylation as durable finishing for various aims (pesticide protection, lipid soil release, antimicrobial properties) has been studied. Research deliverables (academic and industrial) Highly accessible cotton fiber was used for pesticide protective clothes. Durable carboxymethylation finish has been used on cotton fabrics to trap the pesticide
on the fabric decreasing the transfer to the skin and also enhancing the removal of the pesticide by laundering. This finish improved also the lipid soil removal from cotton fabric. Studies on antimicrobial fabric are going on. Publications Borsa, J., Ra´cz, I., Obendorf, S.K. and Bodor, G. (1999), ‘‘Slight carboxymethylation of cellulose’’, Lenzinger Berichte, Special Symposium Issue, pp. 19-25. Borsa, J., Racz, I., Obendorf, S.K. and Bodor, G. (1999), ‘‘Slight carboxymethylation of cellulose’’, Advances in Wood Chemistry, International Symposium, Wien, Austria. Csisza´r, E., Borsa, J., Ra´cz, I. and Obendorf, S.K. (1998), ‘‘The reduction in human exposure to pesticide through selection of clothing parameters: fabric weight, chemical finishing, and fabric layering’’, Archives of Environmental Contamination and Toxicology, Vol. 35, pp. 129-34. Obendorf, S.K. and Borsa, J. (1999), ‘‘Carboxymethylierung von Baumwollflaeche zur Verbesserung der Trageeigenschaften’’, International Textile Bulletin, Vol. 45, pp. 40-2. Obendorf, S.K. and Borsa, J. (2001), ‘‘Soil removal from chemically modified cotton’’, Detergent and Surfactant, Vol. 4 No. 3, pp. 247-56. Ra´cz, I., Obendorf, S.K. and Borsa, J. (1998), ‘‘Carboxymethylated cotton fabric for pesticide protective work clothes’’, Textile Research Journal, Vol. 68, pp. 69-74.
Enschede, The Netherlands Textile Technology, Department of Chemical Engineering, University of Twente, P.O. Box 217, 7500 AE, Enschede, The Netherlands Tel: +31-(0)53-489 3596; Fax: +31-(0)53-489 3849; E-mail:
[email protected] or
[email protected] Dr V.S. Moholkar, Textile Technology Research staff: Prof. Dr ir. M.M.C.G. Warmoeskerken
Ultrasound enhanced mass transfer in wet textile processes Other partners: Academic None Project started: 1 July 1998
Industrial Stork Brabant, The Netherlands Project ended: 30 June 2002 (the project will be continued in 2003) Source of support: Stork Brabant, The Netherlands Keywords: Ultrasound, Enhanced mass transfer, Sono-process engineering, Cavitation, Process intensification
Research register
35
IJCST 15,6
36
One of the main problems in wet textile processes is the relatively slow transport processes in the porous structure of the textile substrate. Due to the complex geometry of textile materials these processes are mainly diffusion controlled. It is believed that ultrasonic waves can enhance these processes. The current project is aimed at understanding the mechanisms of ultrasound waves and their effect on the enhancement of the transport processes by inducing convective diffusion in the pores of textile materials. The mechanisms of ultrasound waves are being investigated in terms of acoustic cavitation phenomena and acoustic streaming. The theoretical analysis is supported by model experiments. Project aims and objectives The relatively slow transport processes in the porous structure of the textile substrate form one of the main problems in wet textile processes. Due to the complex geometry of textile materials these processes are mainly diffusion controlled. The aim of the project is to intensify the mass transfer process in the pores of textiles by acoustic cavitation. The focus of the project is on the mechanisms of ultrasound waves and their effect on the enhancement of the transport processes by inducing convective diffusion in the pores of textile materials. Publications Moholkar, V.S. (2002) ‘‘Intensification of textile treatments: sonoprocess engineering’’, PhD thesis, University of Twente, The Netherlands. Moholkar, V.S. and Warmoeskerken, M.M.C.G. (2000a), ‘‘Mechanistic studies in ultrasonic textile washing’’. AATCC Annual Book of Papers-2000 (CD-ROM version), Section 18, 1-8. Moholkar, V.S. and Warmoeskerken, M.M.C.G. (2000b), ‘‘Scale-up and optimization aspects of an ultrasonic processor’’, Proceedings of 21st Annual European AIChE Colloquium, AIChE NL-BE Section, pp. 59-66. Moholkar, V.S. and Warmoeskerken, M.M.C.G. (2001), ‘‘Intensification of mass transfer in textile materials’’, Proceedings of the 1st AUTEX Conference (Technitex), Povoa do Varzim, Portugal, June 26-29, 2001, pp. 204-13. Moholkar, V.S. and Warmoeskerken, M.M.C.G. (2002a), ‘‘The mechanism of ultrasonic mass transfer enhancement in textiles’’, Proceedings of the 2nd Autex Conference, Bruges, Belgium, 1-3 July, p. 561. Moholkar, V.S. and Warmoeskerken, M.M.C.G. (2002b), ‘‘Mechanistic aspects and optimization of ultrasonic washing’’. AATCC Review, Vol. 2 No. 2, pp. 34-7. Moholkar, V.S., Pandit, A.B., and Warmoeskerken, M.M.C.G. (1999), ‘‘Characterization and optimization aspects of a sonic reactor’’, Proceedings of the International Conference and Exhibition on Ultrasonics (ICEU-99), Ultrasonics Society of India, Vol. 1, pp. 17-22. Moholkar, V.S., Rekveld, S. and Warmoeskerken, M.M.C.G. (2000), ‘‘Modeling of the acoustic pressure fields and the distribution of the cavitation phenomena in a dual frequency sonic processor’’, Ultrasonics, Vol. 38, pp. 666-70. Moholkar, V.S., Huitema, M., Rekveld, S. and Warmoeskerken, M.M.C.G. (2002), ‘‘Characterization of an ultrasonic system using wavelet transforms’’, Chem. Eng. Sci., Vol. 57 No. 4, pp. 617-29.
Enschede, The Netherlands Textile Technology, Department of Chemical Engineering, University of Twente, P.O. Box 217, 7500 AE, Enschede, The Netherlands Tel: +31-(0)53-489 3596; Fax: +31-(0)53-489 3849; E-mail:
[email protected] or
[email protected] M. Lopez-Lorenzo M.Sc., Textile Technology Research staff: Prof. Dr ir. M.M.C.G. Warmoeskerken, Dr ir. V.A. Nierstrasz
Enzymatic upgrading of the properties of recycled paper fibers Other partners: Academic Technical Univ. Eindhoven, Wageningen Agricultural Univ.,
Industrial Kappa RP Europe, Sappi, So¨dra, Voith Sulzer, Buckman, DSM, Novozymes
ATO-DLO, KCPK, TNO Paper and Board Project started: 1 September 2000 Project ends: 31 August 2004 Finance/support: N/A Source of support: Ecology, Economy and Technology program (EET) from the Dutch Ministry of Economic Affairs Keywords: Cellulase, Enzyme technology, Cellulose, Surface modification It is expected that within 10 years the processes of textile production will be shifted substantially due to increasing governmental and environmental restrictions and the availability of fresh water. Enzyme technology is a promising technology to fulfill expected future requirements. In 1998 the Textile Technology Group has taken the initiative to start research on fundamental aspects on enzyme applications. This project focuses the surface modification of cellulose fibres. Paper is a non-woven material formed by cellulose fibres. The recycling of cellulose fibres is limited since the tensile strength decreases during this process. Generally, it is assumed that the deterioration of properties of recycled paper is mainly due to structural changes in the fiber cell wall caused by drying. The project aims to improve the tensile strength of recycled paper, for which different concepts for the surface modification of the fibres will be developed.
Research register
37
IJCST 15,6
Recycled fibers can be upgraded through enzymatic treatments of these fibers. Enzymatic hydrolysis of cellulose by cellulases can improve fibrillation and flexibility, enabling the formation of a fibre-network, which gives improved strength characteristics to the paper.
38
Project aims and objectives The overall aim of the project is ‘‘fibre technology for a durable production of paper and board’’. Our task in this project is to minimise the decrease in the tensile strength of paper during the recycling process using enzyme technology. We will establish the connection between bonding and strength properties and how fibres are affected due to the enzymatic modification. Publications Lenting, H.B.M. and Warmoeskerken, M.M.C.G. (2001a), ‘‘Mechanism of interaction between cellulase action and applied shear force, an hypothesis’’, Journal of Biotechnology, Vol. 89 No. 2-3, 217-26. Lenting, H.B.M. and Warmoeskerken, M.M.C.G. (2001b), ‘‘Guidelines to come to minimized tensile strength loss upon cellulase application’’, Journal of Biotechnology, Vol. 89 No. 2-3, pp. 227-32. Lopez-Lorenzo, M., Nierstrasz, V.A. and Warmoeskerken, M.M.C.G. (2002a), ‘‘Enzymatic modification of cellulosic fibers: from lyocell to recycled paper’’, Book of abstracts of the 2nd International Symposium on Biotechnology in Textile Industry (INTB conference), Athens, Georgia, USA, 3-6 April, pp. 25-6. Lopez-Lorenzo, M., Nierstrasz, V.A. and Warmoeskerken, M.M.C.G. (2000b), ‘‘Enzymatic modification of cellulosic fibers: from lyocell to recycled paper’’, Proceedings of the 2nd Autex Conference, Bruges, Belgium, 1-3 July, p. 563.
Enschede, The Netherlands Textile Technology, Department of Chemical Engineering, University of Twente, P.O. Box 217, 7500 AE, Enschede, The Netherlands Tel: +31-(0)53-489 3596; Fax: +31-(0)53-489 3849; E-mail:
[email protected] or
[email protected] P.B. Agrawal M.Sc., Textile Technology Research staff: Prof. Dr ir. M.M.C.G. Warmoeskerken, Dr ir. V.A. Nierstrasz
Bioscouring of cotton fabrics Other partners: Academic TNO Textiles, TU Graz, UMinho, UPC Terrassa
Industrial Textile Alberto de Sousa (Portugal), Tinfer (Spain)
Project started: 1 December 2000 Project ends: 30 November 2004 Finance/support: N/A Source of support: EU 5th framework Keywords: Bioscouring, Enzyme technology, Cotton, Pectinase The traditional alkaline scouring process can be replaced with an enzymatic scouring process, in which impurities such as protein, wax and ash are efficiently removed prior to further processing of the cotton fabric. Our aim in this project is the development of a new environmentally and industrially viable (enzyme based) continuous process for the scouring of cotton. In order to design a stable enzymatic pre-treatment process, it is necessary to understand the structure of a cotton fiber that will help to make a targeted attack on non-celluloses. Due to the high substrate specificity of most enzymes it is necessary to have sufficient detailed information about the substrate composition and structure to design and introduce a robust pre-treatment process. On the basis of the structure of the cotton fiber an alternative process was proposed. Project aims and objectives The overall aim of this project is the development of a new environmentally and industrially viable enzymatic continuous and batch processes for the scouring of cotton fabrics. Publications Agrawal, P.B., Nierstrasz, V.A. and Warmoeskerken, M.M.C.G. (2002a), ‘‘Bioscouring of cotton textiles: the structure of cotton in relation to enzymatic scouring processes’’, Book of abstracts of the 2nd International Symposium on Biotechnology in Textile Industry (INTB Conference), Athens, Georgia, USA, 3-6 April, pp. 21-2. Agrawal, P.B., Nierstrasz, V.A. and Warmoeskerken, M.M.C.G. (2002b), ‘‘Bioscouring of cotton textiles: the structure of cotton in relation to enzymatic scouring processes’’, Proceedings of the 2nd Autex Conference, Bruges, Belgium, 1-3 July, p. 562. Lenting, H.B.M., Zwier, E. and Nierstrasz, V.A. (2002), ‘‘Identifying important parameters for a continuous bioscouring process’’, Textile Research Journal, Vol. 72 No. 9, 825-31.
Enschede, The Netherlands Textile Technology, Department of Chemical Engineering, University of Twente, P.O. Box 217, 7500 AE, Enschede, The Netherlands Tel: +31-(0)53-489 3596; Fax: +31-(0)53-489 3849; E-mail:
[email protected] or
[email protected] T. Topalovic M.Sc., Textile Technology Research staff: Prof. Dr ir. M.M.C.G. Warmoeskerken, Dr ir. V.A. Nierstrasz
Research register
39
IJCST 15,6
40
Catalytic bleach processes Other partners: Academic TNO Textiles, TNO Nutrition
Industrial Procede Twente B.V., Boessenkool B.V., TDV B.V., Emiod B.V., Vlisco Helmond B.V. Project ends: 30 October 2006
Project started: 1 November 2002 Finance/support: N/A Source of support: Ecology, Economy and Technology Program (EET) from the Dutch Ministry of Economic Affairs Keywords: Oxidative catalysts, Process intensification, Bleach In the traditional bleaching process high temperatures and high concentrations of peroxide are needed. In this recently honored project innovative bleach processes will be developed for the pre-treatment of textile materials using oxidative catalysts. Processes on the basis of oxidative catalysts can be performed at much lower temperatures (40 C) and with a significant reduction of the concentration of chemicals compared to the traditional process. In this project more environmentally acceptable and more efficient pre-treatment processes for textile materials will be developed on the basis of such oxidative catalysts. Project aims and objectives The aim of this project is the development and introduction of more environmentally acceptable and more efficient pre-treatment processes for textile materials on the basis of oxidative catalysts.
Enschede, The Netherlands Textile Technology, Department of Chemical Engineering, University of Twente, P.O. Box 217, 7500 AE, Enschede, The Netherlands Tel: +31-(0)53-489 3596; Fax: +31-(0)53-489 3849; E-mail:
[email protected] or
[email protected] Textile Technology Research staff: Prof. Dr ir. M.M.C.G. Warmoeskerken, Dr ir. V.A. Nierstrasz
Dynamic and advanced wetting of textile materials Other partners: Academic Industrial None None Project started: 1 May 1999 Project ended: – Finance/support: W/A Source of support: The Dutch Foundation for Technology of Structured Materials Keywords: Wetting, Dynamic surface tension, Contact angle, Nanotechnology, Surface heterogeneity, Surface roughness Wetting is key in transport processes in wet textile processing, like washing and dyeing. The knowledge of liquid flow through and the wetting of complex materials is often limited and the process conditions are usually chosen on an empirical basis, rather than a fundamental one. Recent advances in the characterization of surface properties of materials make it in principle possible to relate the macro- and meso-scopic properties of the textile to its nanoscopic properties, even under conditions far from equilibrium. In this project the inter relations between wetting properties, surface roughness, surface heterogeneity and adhesion forces will be studied. The Textile Technology Group has advanced facilities available such as an auto-porosimeter, equipment to measure the dynamic surface tension and high resolution ADSA equipment to measure dynamic and equilibrium wetting characteristics. Project aims and objectives The aim of this project is to generate fundamental knowledge about wetting phenomena in textile materials and to develop tools to predict the wetting behavior in wet textile processes. Publication Nierstrasz, V.A. and Warmoeskerken, M.M.C.G. (2002), ‘‘Dynamic wetting of textile materials’’, Proceedings of the 2nd Autex Conference, Bruges, Belgium, 1-3 July, p. 564.
Ghent, Belgium Ghent University, Department of Textiles, Faculty of Applied Sciences, Technologiepark 907, 9052 zwijnaarde gent Tel: 09/2645411; Fax: 09/2645846; E-mail:
[email protected];
[email protected] Lieva Van Langenhove and Paul Kiekens Research staff: Els Van Nimmen, Kris Gellynck
Research register
41
IJCST 15,6
42
The use of spider silk for scaffolds in soft tissue engineering Other partners: Peter Verdonk, Fredrik Almqvist, Johan Mertens, Domir De Bakker, August Verbruggen Academic Industrial None None Project started: 1 January 2002 Project ends: 31 December 2005 Source of support: BOF – University Ghent Keywords: Spider, Silk, Cartilage, Tissue engineering, Scaffold Injured cartilage, not accreting by itself, often decreases quality of life. The chondrocytes need an implanted support to bridge and recover the wound with extra-cellular matrix products forming fresh cartilage. Advances in cell biology and biomaterial research have led to new possibilities in tissue engineering. Transplanted scaffolds, holding a 3D cell culture, should copy the cartilage characteristics. Strength and flexibility are important, but even more an adequate porosity, so the chondrocytes can migrate through the matrix, but are not able to float around. Looking for regeneration and not a repair, we want the scaffold material to disappear while real cartilage is healing the wound. In this way, the material and its hydrolysis products that are frequently toxic of synthetic polymers have to be biocompatible and harmless. Spider silk is a promising fibre for many applications. Completely made out of protein, a suspected biocompatibility is already proven. The harmless amino acid hydrolysis products make the silk a good candidate for creating a bioresorbable textile scaffold. The chondrocytes cells adhere quite well on the spider cocoon silk threads. Cocoons that could be obtained each autumn in large numbers from the Araneus diadematus garden spider. The mechanical properties of the silk is more appropriate than polymeric gels, like hyaluronic acid, collagen, alginate, which proved to be successful in 3D immobilisation and maintaining the differentiated phenotype of chondrocytes. The phenotypical products collagen II and aggrecan were also detected around the cells growing on the spider cocoon silk. A silk 3D textile could possibly be applied in combination with a polymer gel, probably alginate in order to achieve some biomechanical stability. While biodegradation is occurring, the silk textile is overgrown with real cartilage and eventually the wound will be recovered without any definitive synthetic implants. Project aims and objectives The project-purpose is to compare the properties of spider silk fibres and the textile fabrics that can be made out of these fibres towards the conditions of a good scaffold for soft tissue engineering. A scaffold has to support the
migrating and growing cells and replace the soft tissue while biodegrading during the regeneration. A textile should be engineered with an adequate porosity; porous enough so cells have space to multiply, not too porous so the cells stay attached to the scaffold. The cells should attach on the scaffold so problems due to cytotoxicity are to be excluded and the adhesion of the cells tested. Just as soft tissue the textile scaffold should be strong and flexible. An important condition in tissue regeneration is the biocompatibility of the textile fabric, but also of the hydrolysis-products after biodegradation. Research deliverables (academic and industrial) . Some textile fabric prototypes; . Proof of biocompatibility: in vitro and in vivo tests; . Single fibre properties; . Proof of good adhesion of different cells on the scaffold; . Proof of regrowth of soft tissue in the scaffold. Publications De Bakker, D., Gellynck, K., Van Nimmen, E., Mertens, J. and Kiekens, P. (2002), ‘‘Structural analysis and differences in cocoon structure revealed by means of scanning electron microscopy between several spider species’’, 20th European Colloquium of Arachnology, 22-26 July 2002, Szombathely, Hungary. De Bakker, D., Van Nimmen, E., Baetens, K., Gellynck, K., Mertens, J. and Kiekens, P. (n.d.) ‘‘Structure and use of different silk threads produced by the water spider Argyroneta aquatica’’, Belgian Journal of Zoology (in press). Gellynck, K., De Bakker, D., Van Nimmen, E., Mertens, J., Kiekens, P. and Van Langenhove, L. (2002), ‘‘Research and development of a spider silk textile for cartilage regeneration’’, Poster PhD-Symposium, 11 December 2002. Gellynck, K., Verdonk, P., Almqvist, F., Van Nimmen, E., De Bakker, D., Mertens, J., Kiekens, P., Van Langenhove, L. and Verbruggen, A. (2003), ‘‘A spider silk supportive matrix used for cartilage regeneration’’, 2nd Annual Meeting of the European Tissue Engineering Society, 3-6 September 2003, Genova, Italy. Gellynck, K., Verdonk, P., Almqvist, F., Van Nimmen, E., Van Langenhove, L., De Bakker, D., Mertens, J., Verbruggen, A. and Kiekens, P. (2003), ‘‘A spider silk supportive matrix used for cartilage regeneration’’, Healthcare and Medical Textiles ’03, 8-9 July 2003, Bolton, UK. Gellynck, K., Verdonk, P., Almqvist, F., Van Nimmen, E., De Bakker, D., Mertens, J., Kiekens, P., Van Langenhove, L. and Verbruggen, A. (2003), ‘‘Spider silk-based scaffolds for cartilage repair’’, 18th European Conference on Biomaterials, 1-4 October 2003, Stuttgart, Germany. Van Nimmen, E., Kiekens, P. and Mertens, J. (2002), ‘‘Some material characteristics of spider silk’’, International Journal of Materials and Product Technology (in press). Van Nimmen, E., Gellynck, K., De Bakker, D., Gheysens, T., Mertens, J., Kiekens, P. and Van Langenhove, L. (2002), ‘‘Research and development of spider silk for biomedical applications’’, Proceedings SEM Annual Conference on Experimental and Applied Mechanics, 10-12, June 2002, Biological Inspired and Multi-Functional Materials and Systems, Milwaukee, Wisconsin, USA.
Research register
43
IJCST 15,6
44
Ghent, Belgium Ghent University, Technologiepark 907, B-9052 Zwijnaarde-Gent Tel: +32-9-264 57 35; Fax: +32-9-264 58 46 Department of Textiles Paul Kiekens and Kathleen Van de Velde
Biomedical textiles from dibutyrylchitin and chitin Other partners: Academic Industrial UGent (Belgium), TUL (Poland), UNIAN (Italy) Several European partners Project started: 1 January 2003 Project ends: 31 December 2005 Source of support: EC Keywords: Biomaterials, Chitin, Wound dressings On the European market there is a lack of innovative biomaterials that aid in regeneration of wound tissue. The project aims the design and development of optimal textile forms for medical applications made from dibutyrylchitin (DBC) and chitin, from fishery byproducts. Recently, a method for the synthesis of DBC is developed. DBC is easily soluble in common recyclable solvents and has film/fibre forming properties. Chitin can be regenerated (RC). The products are assumed to have wound healing properties. This opens the way for the development of novel functional biomaterials made from DBC and RC. Technologies for the production of fibres, yarns, non-wovens and knittings will be developed. The textiles will be designed to requirements and characterised on mechanical, physico-chemical and biochemical-medical (in vitro, in vivo) properties. Initiations for approval of medical products will be made. Project aims and objectives Chemical and textile companies will make joint effort with universities to develop the production of DBC, spinning technology of DBC and low dose DBC/cellulose fibres and yarns, production of DBC textiles, regeneration of chitin in manufactured DBC products. An industrial company active in the medical field will co-operate with the laboratories/hospital, where the bioactivity and biocompatibility of these novel biomaterials can be assessed. The project will generate novel biomaterials and medical items, that accelerate the wound healing with no scar formation and undesirable effects, are easy to handle and could be prepared as self-adhering dressings. This kind of dressing would reduce the pain and suffering of patients. The project is oriented and targeted to improve the competitiveness of European industry and enhancing the quality of
life of the EU citizen through the sustainable production and rational utilisation of natural resources with the special emphasis on new technologies.
Research register
Research deliverables (academic and industrial) See project description. Publications None
I˙zmir, Turkey Engineering Faculty, Department of Textile Engineering, Ege University, E.U Mu¨h. Fak. Tekstil Mu¨h. Bo¨lu¨mu¨, 35100 Bornova, I˙zmir, Turkey Tel: 90-232-3887859; Fax: 90-232-3887859; E-mail:
[email protected] Prof. Dr Is¸ık Tarakc¸iogˇlu, Department of Textile Finishing Research staff: Assistant Prof. E. Perrin Akcakoca (Ph.D.), Assistant Prof A. Taner O¨zgu¨ney (Ph.D.), Research Assistant Arzu Ozerdem (M.Sc.)
An investigation of the application of RF dryer combined with steam as a reactor in textile pretreatment, dyeing and printing Other partners: Academic Industrial None None Project started: June 2002 Project ended: June 2003 Finance/support: 35,000 US$ Source of support: Textile Research Center of The Scientific and Technical Research Council of Turkey Keywords: Microwave energy, RF dryer, Reactive dyeing, Pretreatment of cotton, Reactive printing, Ager, Reactor High frequency HF energy has been used in industrial dyeing processes for a long time. Besides drying process it is also possible to make use of HF dryers as a reactor in pretreatment and dyeing processes. It was observed that, in the investigations carried out in our University, the usage of microwave combined with steam has given satisfying results. It was figured out that application of two-step microwave steam process (3 min+3 min ¼ 6 min dwell time) in pretreatments of cotton substrates has given results (desizing, absorbency, seed removal, degree of whiteness), which
45
IJCST 15,6
46
are equivalent to that of conventional two-step pad steam processes (10 min+10 min ¼ 20 min dwell time). In addition, it was observed that, in many cases, one-step microwave steam combination with 3-4 min dwell time has achieved adequate results. The appropriate results in reactive dyeing and printing was also obtained. These experiments were carried out in a kitchen type modified microwave oven (2,450 MHz). In this study the usability of a modified (direct steam feedable) continuous laboratory type RF (27.12 MHz) dryer as a reactor in the pretreatment of cotton fabrics and their dyeing and printing with reactive dyes will be investigated. Project aims and objectives In this study, it is aimed to reduce energy consumption and time saving by using the steam fed RF dryer as a reactor (ager) and consequently to reduce the manufacturing costs without decreasing quality. Research deliverables (academic and industrial) Experiments in progress Publication Tarakcioglu, I. and Anis, P. (1996), ‘‘Microwave processes for the combined desizing, Scouring and Bleaching of Grey Cotton Fabrics’’, Journal of Textile Institute, Vol. 87, pp. 602-8.
I˙zmir, Turkey ¨ BI˙TAK TAM The Scientific and Technical Research Council TU ¨ niversitesi of Turkey Textile Research Center, Ege U ¨ BI˙TAK Tekstil Aras¸tirma Merkezi, Bornova, I˙zmir TU TURKEY Tel: 90 232 3887859; Fax: 90 232 3887859; E-mail:
[email protected] Prof. Dr Kerim Duran, Faculty of Engineering, Textile Department, Ege University Research staff: Assistant Prof. Dr Ays¸egu¨l Ekmekci, Emrah Bilgin, Cem Karabogˇa
Usage of ultrasound combined with hydrogen peroxide and UV light in textile wet processing Other partners: Academic Ege University, Faculty of Engineering, Textile Department Finance/support: $30,124
Industrial None
¨ BI˙TAK TAM Source of support: TU Project started: 1 August 2002 Project ended: 1 August 2003 Keywords: Cotton, Ultrasound, Washing, Hydrogen peroxide, UV light, Textile pretreatment
Research register
It is very important to remove foreign matters from cotton because of quality of textile materials and success in following processes. Effective washing and pretreatment processes depend on three factors: . chemicals and auxiliaries, . mechanical effects, . parameters (temperature, time, materials, etc.). The use of ultrasound in textile finishing offers many potential advantages: . energy savings, . reduced processing times, . reducing environmental problems. In this project, ultrasound will be combined with hydrogen peroxide and UV light. New ultrasound equipment will be used in textile finishing. The project is the first part of great project. All the experiments are preparatory in this year. According to results of the experiments, main project will be expanded.
47
Project aims and objectives . new textile finishing technology, . new ultrasound equipment for textile finishing, . energy savings in textile finishing, . reducing environmental problems, . reducing processing times. Research deliverables (academic and industrial) Publications None
Kaunas, Lithuania Faculty of Design and Technologies, Kaunas University of Technology, Studentu str. 56, Kaunas LT-3031, Lithuania Tel: +370 37 300205; Fax: +370 37 353989; E-mail:
[email protected]
IJCST 15,6
Dr Eugenija Strazdiene Department of Clothing and Polymer Products Technology Research staff: Doctoral students J. Domskiene, V. Dobilaite, V. Sidabraite, K. Dapkuniene
48
The effect of physical-mechanical properties upon the tailorability and appearance of textiles Other partners: Academic Industrial None None Project started: 15 January 2002 Project ends: 15 January 2006 Finance/support: N/A Source of support: Kaunas University of Technology Keywords: Technical textiles, Mechanical properties, Laminates, Coating, Shearing, Buckling, Surface Roughness, Image analysis The research project is orientated towards modern textiles, which is characterized by new original properties extending their functionality and the range of their end use. Laminates and coatings bring textile technology into a new dimension. High performance fabrics are used for leisure, sports, industrial and military garments. In this sense woven structure is very attractive as the reinforcement for composites because it is lightweight, flexural and strong. This makes textile composites suitable for the parts of complicated or curved shape due to their formability properties. Though technical textiles are designed for a specific performance, aesthetics for such garments is very important, too. Thus the main goal of this research is to study the effect of shearing and buckling properties for the behavior of technical textiles, especially for fitting such materials on three-dimensional surfaces without wrinkling and to analyze the conditions when these fabrics start to loose their stable shape, i.e. their surface becomes waved. Project aims and objectives An experimental method based on image capturing and image analysis able to characterize deformational properties of coated and laminated composites in uniaxial and spatial tensile deformations will be developed. Comparative analysis of buckling wave’s propagation, i.e. alterations of its shape (changes of image intensity in certain zone of the specimen) and dimensions (length, width and number of waves) in uniaxial tension of bias (45 ) orientated samples will be performed. Furthermore, the ability of coated and laminated textile composites to be deformed into three-dimensional curvature, i.e. to obtain spherical shape of different diameter will be analyzed from the standpoint of such properties as
tensile extension, shear stiffness, shear hysteresis, bending rigidity and bending hysteresis.
Research register
Research deliverables (academic and industrial) Theoretical results of this research will deepen the knowledge of technical textile mechanical behavior not only under in-plane and perpendicular loading conditions but also in spatial shape formation and will be used in textile and polymer garments design process.
49
Publications Domskiene, J., Strazdiene, E. (2002a), ‘‘Shearing behavior of technical textiles’’, Material Science (Medziagotyra), Vol. 8 (in press). Domskiene, J., Strazdiene, E. (2002b), ‘‘Deformational properties of coated and laminated textile composites’’, Proceedings of the 2nd AUTEX World Textile Conference, Bruges, Belgium, p. 552. Domskiene, J., Strazdiene, E., Dapkuniene˙, K. (2002), ‘‘The evaluation of technical textiles shape stability by image analysis’’, Material Science (Medziagotyra), Vol. 8 (in press).
Kaunas, Lithuania Kaunas University of Technology, Faculty of Design and Technologies, Studentu str. 56, Kaunas-3031, Lithuania Tel: 370-7-767066; Fax: 370-7-353989; E-mail:
[email protected] Department of Clothing and Polymer Products Technology Hab. Dr Professor Matas Gutauskas Research staff: Dr E. Strazdiene, Dr L. Papreckiene, Dr V. Daukantiene and master student G. Martisiute
New method of textile hand evaluation Other partners: Academic Industrial None None Project started: June 2001 Project ends: December 2004 Keywords: Textiles, Membrane, Pulling through, Biaxial deformation, Geometry, Wave, Jamming Although a large part of textiles is concerned with imparting desirable physical properties, theoretical understanding of the effect of these properties on material response is limited. The research project will provide new method for the evaluation of planar anisotropic material’s behavior and original experimental
IJCST 15,6
50
information will be obtained, contributing to the existing knowledge in the field of mechanics of heterogenic, e.g. textile structures. New pulling through a hole method is similar to a well-known punch test used to control strength parameters of knitted material. The difference is that rounded specimen is not firmly fixed by its external contour and the diameter of the hole and the punch are relatively small compared to that of the specimen. Latest investigations have shown that this method is sufficiently informative and able to characterise such hand properties as softness, slippery, roughness, etc. Besides, it provides useful information for the evaluation of textile’s drape and anisotropy. Project aims and objectives The aim of the research is to set the relationship between the geometry and resistance parameters of textile membrane due to its type and testing conditions. Tests are performed by the original device mountable on the standard tensile testing machine. It consists of two perpendicular plates, replaceable stand with the hole in the centre and supporting plate with the hole of the same radius. Spherical punch is used to pull rounded specimen through the hole of the stand. The investigations are realised by two pulling through cases: free pulling through the hole of the stand; restrained pulling simultaneously through the limited crack of the plate and through the same hole of the stand. Research deliverables (academic and industrial) New testing method for textile and its experimental base will extend the existing laboratory of material testing and will be used in the study process of the Kaunas University of Technology. Theoretical results of this research will deepen the knowledge of textile spatial shape formation and will be used in textile and polymer garments design process. Publications Martisiute, G. and Gutauskas, M. (in press), New Approach to the Evaluation of Fabric Handle, Materials Science, Kaunas. Martisiute, G. and Gutauskas, M. (in press), ‘‘Pulling through process of knitted membrane: analysis of geometry’’, Proc. of the Conf. Design and Technology of Consumer Goods - 2001, Kaunas (in Lithuanian). Strazdiene, E., Martisiute, G., Gutauskas, M. and Papreckiene, L. (in press), ‘‘New method for the objective evaluation of textile hand’’, Journal of the Textile Institute, UK.
Kettering, UK Satra Technology Centre, Satra House, Rockingham Road, Kettering, Northamptonshire NN16 9JH Andrea Wilford, CTC
Clothing comfort (intended) Other partners: Academic None Project started: – Finance/support: N/A Source of support: SATRA Keywords: Clothing, Design .
.
.
Industrial None Project ends: Ongoing
A study of the factors important for comfortable clothing and development of a method to measure and assess comfort. To provide the clothing industry with a practical technique to quantify the comfort of clothing, highlighting strengths and weaknesses in design and making recommendations on how to improve the comfort of the garment. To provide guidance in product design, product manufacture and purchasing.
Project aims and objectives The development of a Clothing Comfort Index. A study of the factors important for comfortable clothing and to develop a practical technique for the quantification of clothing comfort. Academic deliverables Production of a Comfort Index System. Industrial deliverables A design guide for manufacturers and retailers to enable them to source materials and provide goods which meet consumer needs. Publication SATRA, ‘‘Clothing-Closeup’’.
Kettering, UK Satra Technology Centre, Satra House, Rockingham Road, Kettering, Northamptonshire NN16 9JH. Austin Simmons, CTC Research staff: Mark Gamble
Swimwear degradations Other partners: Academic None
Industrial None
Research register
51
IJCST 15,6
Project started: June 1998 Finance/support: N/A Source of support: SATRA Keywords: Elastane, Swimwear
52
Limited wear trials were conducted on a variety of elastane-containing swimwear and examinations were carried out on a selection of failed swimwear garments. The common feature of each garment’s failure was the breakdown of the elastane component. In each case of failure it was noted that garments exhibited a particular pattern of wear. The project currently being undertaken aims to reproduce the wear patterns in a laboratory setting. It is intended to investigate the flow of swimming bath water through the fabric structure of different garments and the effect of flow restriction in preserving the life of a garment. It is also intended to develop a test rig for assessing the effects of combined chemical and mechanical action on swimwear garments.
Project ends: Ongoing
Project aims and objectives To establish an effective means of assessing the likely wear performance of elastane-containing swimwear. The means of assessment to incorporate a chemical and mechanical system for degrading swimwear materials. Academic deliverables An understanding of the mechanisms which contribute to elastane failure in swimwear garments. Industrial deliverables A test apparatus for predicting garment performance. Publication SATRA, ‘‘Clothing-Closeup’’.
Kettering, UK Satra Technology Centre, Satra House, Rockingham Road, Kettering, Northamptonshire NN16 9JH Andrea Wilford, CTC Research staff: David McKeown, Mark Gamble
Water-resistant permeable membranes Other partners: Academic None Project started: –
Industrial None Project ends: Ongoing
Research register
Finance/support: N/A Source of support: SATRA Keywords: Garments, Wear Limited wear trials have been undertaken on a series of commercially available garments which incorporate membrane structures. The results of this testing, which may be advanced to moisture and temperature recording of wear trialled products, using data-loggers, will be made available to SATRA members. Much of the work that is to be undertaken will complement SATRA’s current Comfort Index work. Project aims and objectives To understand the mechanisms at work in water permeable membranes which are used for clothing. We aim to draw on the expertise developed in the use of such materials in footwear. Academic deliverables To develop performance guidelines for current market products. Industrial deliverables To provide a service to industry for the development of membrane materials (and their testing). Publication SATRA, ‘‘Clothing-Closeup’’.
Kowloon, Hong Kong The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong Tel: (852) 2766 6470; Fax: (852) 2773 1432; E-mail:
[email protected] Institute of Textiles and Clothing Professor Xiaoming Tao Research staff: Dr Kwok-po Cheng, Yam-kuen Yip, Dr Ka-fai Choi, Sing-kee Wong, Dr Ka-kee Wong, Dr Bingang Xu, Chak-lam Leung, Charlotte Murrells, Tao Hua, Kun Yang
Novel ring yarns and production technology Other partners: Government Innovation and Technology Commission
Industrial Central Textiles (HK) Limited Chip Tak Weaving Factory Limited
53
IJCST 15,6
54
The Government of the Hong Kong Special Administrative Region
Fountain Set (Holdings) Limited Perfecta Dyeing Printing and Weaving Works Limited Project ends: 31 January 2005
Project started: 1 November 2002 Finance/support: HKD 6,000,000. Source of support: Innovation and Technology Commission; The Government of the Hong Kong Special Administrative Region; Central Textiles (HK) Limited; Chip Tak Weaving Factory Limited; Fountain Set (Holdings) Limited; Perfecta Dyeing Printing and Weaving Works Limited Keywords: Residual torque, Torque free, Knitting yarn, Weaving yarn Residual torque or twist liveliness of a twisted yarn is the most prominent and fundamental factor contributing to the spirality of single jersey knitted fabrics and distortion of woven fabrics. These occur when the residual torque, developed in the component fibres of a yarn by the twisting action during the spinning process, is released. Various techniques have been used in the past to reduce or eliminate these fabric imperfections. However, several drawbacks are associated with these techniques, such as unsatisfactory fabric performance and high production costs. Therefore, a new technique has been invented to produce torque free single ring yarns with a single step on ring spinning machines. Our laboratory results show that the resultant fabrics have clear and smooth surfaces, soft handle, lower pilling and good strength, in addition to very low spirality after washing-tumble-dry cycles. Furthermore, the newly developed technology can save production cost substantially in the yarn production and fabric finishing as well as material cost in the spreading and cutting stage of garment manufacturing. The project is intended to further develop these new technologies in the laboratory and transfer them to the Hong Kong textile and apparel companies for industrial application. Optimization of the machine design, product and processing parameters in terms of product performance and cost will be carried out with measurement methods of the yarn performance and structures. Project aims and objectives (1) Optimization of machine designs in terms of yarn performance and structures. (2) Optimization of processing parameters in terms of yarn performance and structures. (3) Optimization of product parameters in terms of yarn performance and structures. (4) Successful industrial production of torque free single spun yarns for both weaving and knitting sectors. (5) Establishment of quality assurance procedure in the production and application of torque free single spun yarns.
Research deliverables (academic and industrial) (1) Devices for ring yarn modification . Optimized devices for producing torque single yarns for weaving and knitting: Laboratory trials: 7, 10 16 and 20Ne. Production trials: 7Ne for weaving and 20Ne for knitting. (2) Test devices and methods for modified yarns . Yarn residual torque, yarn appearance. (3) Product and process parameters . Raw material, preparation procedure, yarn count/twist, splicing/ cleaning procedure, dye/finishing. (4) Completed industrial reports . Raw materials, yarn, fabric, finishing, garment, tests, wear trial and QA procedure. Publications None
Kowloon, Hong Kong The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong Tel: (852) 2766 6470; Fax: (852) 2773 1432; E-mail:
[email protected] Institute of Textiles and Clothing Professor Xiaoming Tao Research staff: Professor Chung Loong Choy; Professor Xiao Ming Tao; Dr John Xin; Dr Yau Shan Szeto; Dr Chun Wah Marcus Yuen; Professor Lai Wah Chan Wong; Professor C Surya; Dr Pei Li; Dr Mei Yi Leung; Dr Willy Chan; Dr C L Mak; Dr Pu Xue; Dr Chen Wei; Sun Xiao-hong; Dr Cheng Xiao-yin; Dr Ye Wei-jun; Tsang hing-yee, Kong Yee-yee; Dr Deng Jian-guo; Dr Li Chang-sheng
Nanotechnology center for functional and intelligent textiles and apparel Other partners: Government Innovation and Technology Commission
Industrial Bondex International (HK) Limited Cha Textiles Limited
Research register
55
IJCST 15,6
The Government of the Hong Kong Special Administrative Region
56
Project started: 1 June 2003 Finance/support: HKD 14,481,762 Source of support: Innovation and Technology Commission; The Government of the Hong Kong Special Administrative Region; Bondex International (HK) Limited; Cha Textiles Limited; Glorious Sun Holdings Limited; Sun Hing Elastic and Lace Fty. Ltd; Wah Tai Piece Goods Ltd; Artex Fashions (Asia) Ltd; Link Dyeing Works Limited Keywords: Nanotechnology, Functional and intelligent, Textiles and apparel
Glorious Sun Holdings Limited Sun Hing Elastic and Lace Fty. Ltd Wah Tai Piece Goods Ltd Artex Fashions (Asia) Ltd Link Dyeing Works Limited Project ends: 31 May 2006
Nanotechnology Center established specifically for the textile and apparel industry in Hong Kong, which is one of the five largest exporters of textiles and apparel products in the world. This Nanotechnology Center will further strengthen the competitiveness of the industry by achieving the fourfold objectives listed in the following. Nanotechnology has been regarded as an essential enabling technology for the next generation of fiber-based functional and intelligent textile materials and apparel. Our multi-disciplinary research team actively worked in the area and demonstrated several new technologies with very promising industrial application potentials in the past. The 3 year program will extend our past research activities and develop the fundamental research into technology for industry. The program of the center will focus on nano-finishing systems and nanotechnology for intelligent textiles and apparel products. The projects are devoted to investigation and development of environmentally friendly and effective nano-finishing processing systems for textile fabrics and garments, including surface polymerization system, systems for precise manufacture of nano-particles, nano-scaled polymer bulk treatment system and printing/chemical vapor deposition system. These newly-developed processing systems will be used for producing various functional or smart/intelligent products, such as sensing textiles and apparel as well as nano-structured photonic fibers and films. Project aims and objectives The main objectives for the Nanotechnology Center for Functional and Intelligent Textiles and Apparel are: (1) to provide research and develop infrastructure for textiles and apparel related nanotechnology,
(2) to develop new nano-materials, new processing technologies and products for high value added functional and intelligent textiles and apparel, (3) to facilitate technology transfer to and collaboration with the industry, and (4) to provide training to postgraduate students and company technical personnel. Research deliverables (academic and industrial) (1) Optimized surface polymerization systems for UV-blocking, stain-, oil-, water-repellent, anti-bacteria finishing of cotton, polyamide and polybenzimidazole fabrics, nano-pigment coloration system. (2) Customer tailored synthesis systems for precise size and sensitivity control of nano-structures for functional finishing and photonic fibers. (3) Optimized fabrication system for conductive textiles sensing devices for strain, temperature and relation humidity, and a prototype of electrical sensing apparel. (4) Prototypes of photonic fibers that can regulate light intensity and color and a prototype of two-colored display fabric made from such fibers. In addition, the center will train several postgraduate research students, conduct training courses for company personnel, and carry out other promotion activities, etc. Publications None
Kyeongsan, Korea School of Textiles, Yeungnam University, 214-1 Daedong Kyeongsan 712-749, Korea Tel: 82-53-8102771; Fax: 82-53-8125702; E-mail:
[email protected] Seung-Jin Kim Textile Processing Laboratory Research staff: K.S. Park, S.B. Sim, S.Y. Kim, M.Y. Park
Development of easy-care worsted fabric using drawn worsted yarns Other partners: Academic None Project started: 1st March 2002
Industrial None Project ended: 28 February 2003
Research register
57
IJCST 15,6
Finance/support: US$ 50,000 Source of support: Ministry of Science and Technology Keywords: Easy-care, Drawn worsted yarns, Anti-shrinkable, Twist, Drawing temperature
58
This project covers development of easy-care worsted garment using drawn worsted yarns. For this purpose, various chemical treatment technology are applied for marking drawn worsted yarns. The chemical treatment technology includes anti-shrinkable technology and various chemical agent treatment recommended by WRONZ in New Zealand. The optimum processing condition such as yarn twist, drawing temperature, drawing time and yarn count on the drawing machine for easy care clothing are surveyed and analysed. Finally the physical properties of garment are measured and discussed with the various processing conditions of draw worsted yarns. Project aims and objectives . Development of antipilling worsted drawn yarns. . Development of easy-care worsted fabrics and garment. Academic deliverables . One or more graduate theses. . Presentation to seminar as a paper. Industrial deliverables . Manufacturing procedures for implant for making textile goods. Publication Lee, D.H., Kim, S.J. and Seo, O.K., ‘‘Changes in the properties of wool fibres, yarns and fabrics by finedrawing of worsted yarns’’, Extended Abstracts of the 31st Textile Research Symposium at Mt. Fuji, Textile Science Research Group in Text. Mac. Soc. of Japan, August, 2002, p. 24.
Kyeongsan, Korea School of Textile & Fashion, Yeungnam University, 214-1 Daedong, Kyeongsan, Korea Tel: 82-53-8102536; Fax: 82-53-8125702; E-mail:
[email protected] Seung-Jin Kim, Textile Processing Lab Research staff: B.K. Seo, S.D. Hong, S.B. Sim
Development of knit and woven fabrics using drawn worsted yarns and their drawing system Other partners: Academic Industrial None O.K. Seo Project started: 1 July 2001 Project ended: 30 June 2003 Finance/support: US$24,600 Source of support: Ministry of Commerce, Industry & Energy Keywords: Worsted drawing yarns, Silk-like worsted yarn, Drawing ratio, Heating temperature This project surveys manufacturing technology of the silk-like worsted yarns and fabrics, and includes development of the drawing system of worsted staple yarn. Using this drawing system, optimum drawing ratio and temperature are decided. Fine staple worsted yarns (100 Nm) are made from 66 Nm and 52 Nm staple worsted yarns using the drawing system. The optimum conditions in the drawing process such as drawing ratio and temperature for linen-like and silklike knitted fabrics are decided through various experiments. The physical and mechanical properties of the specimens of the yarns and knitted fabrics are measured and discussed with various processing conditions in the drawing system. The yarn physical properties measured are thermal shrinkage, snarl index, bending rigidity, torsional rigidity, and fabric mechanical properties are tensile, bending, shear, compression and surface. Project aims and objectives Objectives of this research are to develop the linen-like and/or silk-like worsted yarn for knitted fabric. Also this project aims at the development of drawing machinery for worsted yarns, which is available to control draw ratio and drawing temperature, and includes the determination of the optimum twist condition. Academic deliverables . one or more graduate theses; . presentation to seminar as a paper. Industrial deliverables . manufacturing procedures for implant for making textile goods. Publications None
Research register
59
IJCST 15,6
60
Kyoto, Japan Kyoto Institute of Technology, Faculty of Textile Science, Matsugasaki Sakyo-ku, Kyoto, 606 Japan Tel: (075) 724 7846; Fax: (075) 724 7800 M. Nakamura, T. Matsuo and M. Nakajima, Department of Polymer Science and Engineering
Analyses of tuft forming at bale opener Other partners: Academic None Project started: April 1994 Keywords: Simulation, Tuft, Yarns
Industrial None Project ends: To be continued
The purpose of the bale opener is to open the bale and to produce fine and uniform size tufts for the sequential process of yarn spinning. The opening mechanism at the tuft forming process was investigated, based on theoretical analyses of macroscopic mass balance and of tooth edge locus, and two experimental model tests. A theoretical model of microscopic mass balance was presented to simulate the opening process. Numerical calculations for several process conditions were carried out using the experimental results of model tufting studies. These simulation results have proved to be a good tool for better understanding of the process. The effect of good tuft forming by bale openers on the processibility of the sequential process of yarn spinning and the quality of yarn thus produced was also investigated through production tests of real mills. Project aims and objectives (1) To clarify the opening mechanism at the tuft forming process of the bale opener. (2) To establish simulation technology for the processing. (3) To investigate the effect of good tuft formings by bale openers on the processibility of the sequential process of yarn spinning, and the quality of yarn thus produced. (4) To pursue means towards the improvement of the bale opener. Industrial deliverables Refer to the publication. Publication Nakamura, M., Matsuo, T. and Nakajima, M. (1997), Journal of Textile Machinery Society.
Kyoto, Japan Kyoto Institute of Technology, Faculty of Textile Science, Matsugasaki Sakyo-ku, Kyoto, 606 Japan Tel: (075) 724 7846; Fax: (075) 724 7800 M.N. Suresh and T. Matsuo, Department of Polymer Science and Engineering
Development of total material design system of woven fabrics for apparel use Other partners: Academic M. Nakajima
Industrial T. Harada M. Inoue Project started: September 1994 Project ends: To be continued Keywords: Apparel, CAD, Fabric, Woven fabrics Although much research has been done on colour/pattern designing related to fabric appearance and fashion, material design technology for apparel fabrics is still in the developmental stage. A review shows that the past 20 years have witnessed the efforts towards trial construction of partial design systems and conceptualization of total material design logic. The main aim of this part in the series of our studies is to construct a fundamental logical structure of the computer-assisted total material design system for general apparel woven fabrics. The main components of the structure thus constructed and their functions are defined. The system consists of three sections: a user interface, the five design stages starting from conceptual design up to the detailed manufacturing design, and different types of databases which support the design stage. The format and contests of important system components are explained with examples. The executional logic of the system and its flow is also presented with a methodology to find a suitable design solution. Utilization of a ‘‘reference sample’’ has been introduced to simplify the design procedure. Some detailed case studies to illustrate application of this system have been carried out. The frame of the computer system is also being developed. Project aims and objectives To develop a ‘‘computer-assisted total material design system of woven fabrics for apparel use’’. Academic deliverables Refer to the publications.
Research register
61
IJCST 15,6
62
Publications Matsuo, T. and Suresh, M.N. (1997), Textile Progress. Suresh, M.N., Matsuo, T. and Nakajima, N. (1997a), Proceedings of IV Congress ATC, Taipei. Suresh, M.N., Matsuo, T. and Nakajima, N. (1997b), Journal of Text. Mac. Soc., Vol. 50, T146. Suresh, M.N., Matsuo, T. and Nakajima, N. (1997c), Proceedings of 25th Textile Research Symposium, Mt Fuji, Japan. Suresh, M.N., Matsuo, T. and Nakajima, N. (to be published), Journal of Text. Mac. Soc.
Kyoto, Japan Kyoto Institute of Technology, Faculty of Textile Science, Matsugasaki Sakyo-ku, Kyoto, 606 Japan Tel: (075) 724 7846; Fax: (075) 724 7800 D.A. Alimaq and T. Matsuo, Department of Polymer Science and Engineering
Sensory measurements of fabric hand/mechanical properties: part I – worsted fabrics; part II – knitted fabrics Other partners: Academic Industrial M. Nakajima T. Harada Project started: April 1993 Project ends: To be continued Keywords: Fabric, Knitwear, Sensory measurement, Worsted Systems of instrumental method for measuring fabric hand have been fairly successfully developed like KES and its basic way has been well established. On the contrary, systems of sensory method have remained controversial. In this paper, a practical sensory method is proposed on the basis of analogy to sensory colorimetry. Measurement of two kinds of worsted fabrics was conducted by making use of this sensory method. The effective range and the accuracy of this method are discussed based on the data of the above measurement. It is shown that, if a suitable control (temporary standard) sample is chosen, the instrumental values of bending rigidity, thickness and compressibility of worsted fabrics can be estimated by this sensory method with an error of around 20 per cent. The sensory measurement of main mechanical properties of knitted fabrics is now being conducted. Very good results have been obtained so far on these points.
Project aims and objectives (1) To develop handometry for fabrics by sensory method. (2) To investigate the effectiveness of sensory measurement of fabric mechanical properties. Academic deliverables Refer to the publications.
63
Publications Alimaa, D., Matsuo, T., Nakajima, M. and Takahashi, M. (1997), Proceedings of the 25th Textile Research Symposium, Mt Fuji, Japan. Matsuo, T., Harada, T., Saito, M. and Tsutsumi, A. (1995), Journal of Textile Machinery Society, Vol. 48, T244.
Kyoto, Japan Kyoto Institute of Technology, Faculty of Textile Science, Matsugasaki Sakyo-ku, Kyoto, 606 Japan Tel: (075) 724 7846; Fax: (075) 724 7800 T. Matsuo and Ryuichi Akiyama, Department of Polymer Science and Engineering Research staff: Fumitaka Okamoto
Surface mechanical properties of fabrics in terms of hand: part I - Shingosen fabrics Other partners: Academic M. Kinoshita
Project started: April 1993 Keywords: Fabric, Woven fabrics .
.
Research register
Industrial None S. Mukhopadhyay K. Izumi Project ends: To be continued
A measurement method for surface mechanical properties (especially frictional properties) of fabrics in the relation with their surface hands has been developed. The effects of the friction probe form, probe velocity, probe weight and the selection of suitable parameters representative of frictional properties are investigated.
IJCST 15,6
.
.
64
The relationships between surface hand and surface mechanical properties for Shingosen fabrics have been clarified, in comparison with silk-like fabrics and silk fabrics. An attempt is also made to simulate frictional properties by certain theoretical structure models of woven fabrics.
Project aims and objectives (1) To develop a measuring method for surface mechanical properties (especially frictional properties) of fabrics. (2) To find the features of these properties and the relationships between hand and these properties. (3) To simulate frictional properties theoretically on the basis of fabric structure. (4) To analyze Shingosen fabrics from the viewpoint of (2). Academic deliverables Refer to the publications. Publications Akiyama, R. et al. (1995), Journal of Textile Machinery Society, Vol. 48, T153. Kinoshita, M. et al. (1997), Journal of Textile Machinery Society, Vol. 50, T187.
Kyoto, Japan Kyoto Institute of Technology, Faculty of Textile Science, Matsugasaki Sakyo-ku, Kyoto, 606 Japan Tel: (075) 724 7846; Fax: (075) 724 7800 K. Kawabe and T. Matsuo, Department of Polymer Science and Engineering
Tow opening of reinforcing fibre and its application for thermoplastic composites Other partners: Academic Industrial None S. Tomoda Project started: June 1994 Project ends: To be continued Keywords: Fibre, Pneumatics, Thermoplastics Impregnation of matrix resin into fibre is further facilitated by using opened tow rather than compacted tow. A new processing system for spreading tow which is composed of plural rolls and a pneumatic device was introduced. Preliminary opening is conducted by threading it on plural fixed rolls under a suitable initial
tension. Transverse air of suitable flow velocity is then applied to the tow of steadily sagged form. Some experimental results for carbon fibre and glass fibre, and theoretical analysis on these opening mechanisms were presented. The roles of roll part and pneumatic part were also discussed. Thus opened tows have been applied to the impregnation of thermoplastic composites. Significant effect of the tow opening on the facilitation of matrix impregnation has been proved. Project aims and objectives (1) To develop tow opening of reinforcing fibre with high efficiency and low cost. (2) To analyze the opening mechanism. (3) To apply the tow opening technology to the production of thermoplastic composite prepreg. Academic deliverables Refer to the publications. Industrial deliverables At present, laboratorial scale. Publications Kawabe, K., Matsuo, T. and Tomoda, S. (1997), Proceedings of 42nd International SAMPE, Vol. 42, p. 65. Kawabe, K., Tomoda, S. and Matsuo, T. (1997), Journal of Textile Mac. Soc., Vol. 50, T68.
Leeds, UK School of Design, Leeds University, Leeds, LS2 9PR, UK Tel: 0113 2333711; Fax: 0113-2333704; E-mail:
[email protected] CTT, Mechatronics Research Group Farzad Jahanshah, Duncan Borman Research staff: H. Gaskell (Mechanical Eng. Dept), Dr A.A. Dehghani (Textiles Dept), Professor T. King (Honorary Professor)
Mechatronics and machine vision for online fault detection and rectification in inkjet printing Other partners: Academic Industrial None Dorma, Zaphire Flag and Banners, M & S,. . . Project started: September 2000 Project ended: August 2003
Research register
65
IJCST 15,6
66
Source of Support: EPSRC Keywords: Inkjet printing, Vision system, Machine vision, Textiles inkjet printing Implications for inkjet technology to be modified to print high-quality colour images on a wide range of surfaces have currently been recognized. This is particularly apparent for wide format designs. Using specialist inks, short run designs can be inkjet printed onto everything from floor coverings to textiles and ceramics. Speed and reliability are two important factors that can be developed to improve production printer results. Nozzle blockage can be a serious problem when using exotic inks and media. Imperfect prints mean wasted time, materials and energy and are very expensive. This has been seen particularly in the textile industry where attempts to inkjet print textiles at high speed with specialist inks have proven problematic. Current research addresses these problems with the emphasis being on the development of a vision and control system that can detect and rectify faults online. Using two scanners at either side of each colour printhead and an appropriately tuned illumination source, live images can be processed and analysed to detect the blocked nozzles. Results can be reported to a control system for online rectification. Colour line scan technology is still expensive in comparison to the technology used in desktop scanners. Work is being undertaken to develop hybrid-scanning devices that use the low cost modular technology, incorporated in scanner design, for monitoring inkjet printing. The developed system uses a novel approach to detect faults in real time. Lowcost modular scanner technology makes it possible for each printhead to have its own localized detection system. Project aims and objectives The approach taken is based on the concept of checking the printed textile during the printing process and using a flexible machine topology that allows errors to be recovered in real time without wastage. Parallel developments of low cost vision systems for digitally printed textiles have provided a means to check textiles during printing and to take corrective measures when necessary. The potential for a high-speed high-reliability system is greatly increased with the inclusion of a vision and control system Research deliverables (academic and industrial) . Printing industries; . Micro detecting applications;
. .
Real time machine vision; Textiles inkjet printers manufacturer.
Publications Borman, D., Jahanshah, F., Dehghani, A.A., King, T. and Dixon, D. (2002), ‘‘Mechatronics system topology and control for high-speed, high-reliability textile inkjet printing’’, Mechatronics 2002, Holland. Borman, D.J., Jahanshah, F., Dehghani, A.A., King, T. and Gaskell, P. (2002), ‘‘Online vision system for ink-jet printed media’’, IS & T’s NIP18, International Conference, October 2002, San Diego, California, USA, pp. 562-7. Jahanshah, F., Borman, D., Dehghani, A.A., King, T. and Dixon, D. (2002),‘‘Mechatronics and machine vision for online fault detection and rectification in inkjet printing’’, Mechatronics 2002, Holland. Jahanshah, F., Borman, D., Dehghani, A.A., King, T. and Gaskell, P. (2002), ‘‘Real-time detection and rectification for ink-jet printing of specialist wide format surfaces’’, Machine Automation International Conference, September 2002, ICMA Finland, pp. 259-66.
Liberec, Czech Republic Technical University of Liberec, Faculty of Textiles, 461 17 Liberec, Ha´lkova 6, Czech Republic Tel: 00420 48 25441/25462 Professor Stanislav Nosek, Department of Weaving Technology (newly renamed Department of Mechanical Technologies in Textiles) Research staff: Ingolf Brotz, Petr Tumajer, Ales˘ Cvrkal and Jaroslava Richterova´
Research of shocks (impacts) and vibrations excited by technological processes in weaving and other textile machines Other partners: Academic Industrial None None Project started: 1998 Project ends: – Finance/support: Kc˘1,900,000 (estimated) Source of support: Applied with the Grant Agency of the Czech Republic (GACR) (or will be worked out as an internal project of TU Liberec) Keywords: Textiles, Weaving Many textile technological processes, especially the weaving process, produce during each working cycle a row of force impulses which affect the processed textile material as well as the machine. The impact of these impulses causes the
Research register
67
IJCST 15,6
68
propagation of delayed deformation of both media – textile material and machine parts – so that the deformation can return to the source of impulse through several paths. The result is that the next impulse changes with respect to the previous one and the technological process may become unstable or steadied in a different regime to that originally set on the producing system, etc. That can affect the quality of the produced good. At the same time, the excited shocks and vibrations in the system material – working machine – can be emitted in the air or into the floor as noise or vibrations on a wide band of frequencies and can affect the workers as well as the environment and the building. The problems of arising propagation and damping of technologically affected impulses and vibrations will be studied first on the process of fabric forming of the loom as the effect of beat-up, of shedding, back rest motion, functioning of fabric take-up and warp let-off devices. Later, the research should be widened to further textile processes – winding, warping, etc. Project aims and objectives The aim of the research is mainly to explain why and how the produced goods on textile machines often differ from the structure and quality of the goods originally (theoretically) set on the machine. One possible reason may be the deviated motions of textile material and machine parts caused by impulses and vibrations in these compliable and massive media, which impulses result from the technological process itself. The research will start with the weaving machines. Academic deliverables A new theoretical view on the stability of technological processes as processes of propagation and returning (feedback) of impulses and vibrations in compliable textile material and machine parts in textile technologies. The research will also be explored as the source of problems for training of PhD students. Industrial deliverables Results will be applied in textile machines design. Results concerning the propagation of vibrations into the air and into the floor will be used to research the protection of persons as well as buildings against damage by noise and vibrations. Publications Hanzl, J. (1995), ‘‘The behavior of the back rest on the loom’’, Poster and Book of Transactions, International Conference of Young Textile Science, TU Liberec. Nosek, S. (1994a), ‘‘The dynamics of fabric forming at high weaving rates’’, Industrial Journal of Fibre & Textile Research, Vol. 19 No. 3. Nosek, S. (1994b), ‘‘The dynamics of fabric forming on the loom and problems of weavability at high weaving rates’’, World Textile Conference, Huddersfield.
Nosek, S. (1995a), ‘‘Dynamics and stability of beat-up’’, Fibers & Textiles in Eastern Europe, Vol. 1 No. 1, Lodz. Nosek, S. (1995b), ‘‘Feedback phenomena in textile processes’’, International Conference of Young Textile Science, TU Liberec. Nosek, S. (1995c), ‘‘Mechanics and rise of stop marks and structural bars in fabrics’’, Poster and Book of Transactions, IMTEX ’95, Lodz, Polsko. Tumajer, P. (1995), ‘‘Dynamics of start of a weaving loom and the possible rise of transition marks’’, Poster and Book of Transactions, International Conference of Young Textile Science, TU Liberec.
Liberec, Czech Republic Technical University of Liberec, Halkova 6, 461 17 Liberec, Czech Republic Tel: +420 48 5353498; E-mail:
[email protected] Zdene˘k Ku˚s, Head of Department, Department of Clothing Research staff: Jir˘´ı Militky´, Otakar Kunz, Antonı´n Havelka, Dagmar Ruz˘ic˘kova´, Vladimı´r Bajzı´k, Jana Zouharova´, Andrea Halasova´, Viera Glombı´kova´, Blaz˘ena Musilova´, Petra Koma´rkova´, Jaroslav Beran, Josef Olehla, Miroslav Brzezina
Organoleptic properties of three-dimensional textile objects Other partners: Academic Industrial Other departments of the university Project started: 1 January 1999 Project ends: 31 December 2004 Finance/support: £70,000 Source of support: Ministry of Education, Technical University of Liberec Keywords: Fabric properties, Comfort, Handle, Thermal The research will be performed in the following areas: Surface properties of textile formations, non-linear deformations of fabric . Computer simulation of impact surface parameters of the planispheric textile fabric to their chosen macroscopic properties. This is made with the aim of predicting and optimising these properties. Provide evolution for new measurement methods in this area. . Computer simulation of non-linear deformation of the planispheric textile fabric with ballast, for example, by means methods of final elements.
Research register
69
IJCST 15,6
70
Physiological properties of comfort textile formations, fabric handle . Development of new methods for evaluation of physiology comfort and fabric handle. The following application progressive computers method. For example, the neural networks or the artificial intelligence for comparing objective new parameters with empirical find out values. Thermal properties of textiles sandwich materials . Development of new devices for measurements of thermal properties of textile composites and textiles sandwich materials, with special regard to these materials applied in extreme conditions. . Objectification of the property evaluation of textile materials from the point of view of comfort and hygienic properties. Project aims and objectives New methods of measurement, computer simulation of fabric deformation, etc. Publications Halasova, A. and Glombı´kova´, V. (2000), ‘‘Problem of simulation working breakdown apparel production in program accessories witness’’, Proceedings of Textile Science 2000, Liberec, Czech Republic, 12-16 June, ISBN 80-7083-409-9, p. 375. Hes, L., Li, Y. and Kus, Z. (1999), Ochranna´ textilie proti sa´lavemu teplu a ochranny´ oblek z te´to textilie, Czech Republic Patent PV1673-99. Kus, Z. (1999), ‘‘Investigation of seam pucker with help of image analysis’’, Proceedings of the 5th Asian Textile in the 21st Century Conference, Kyoto, Japan, p. 333. Kus, Z. and Koma´rkova´, P. (2000), ‘‘Computer simulation of apparel production’’, Vla´kna a Textil, Vol. 7 No. 2, ISSN 1335-0617, pp. 113-16. Kus, Z., Glombı´kova´, V. and Brada´cova´, H. (2000), ‘‘Application of image analysis and neural network for the evaluation of seam pucker’’, Proceedings of Textile Science 2000, Liberec, Czech Republic, 12-16 June, ISBN 80-7083-409-9, pp. 391-3. Trung, N.C. and Kus, Z. (1999), ‘‘Computer simulation of sewing needle heating’’, Progress in Simulation, Modeling, Analysis and Synthesis of Modern Electrical and Electronic Devices and Systems, World Scientific and Engineering Society Press, Athens, Greece, ISBN 960-8052-08-4, pp. 166-70.
London, UK King’s College London, Department of Mechanical Engineering, King’s College, University of London, Strand, London WC2R 2LS, UK Tel:+44(0)2078482321; Fax: +44(0)2078482932; E-mail:
[email protected]
Jian S Dai Research Staff: Honghai Liu
Research register
Feasibility study into robotic ironing Other partners: Academic Industrial None Paul M. Taylor Project started: June 2002 Project ended: April 2003 Finance/support: £65k Source of support: EPSRC Keywords: Robotic ironing, Handling and manipulation, Folding and unfolding, Garment handling, Gripping devices This is a collaborative project bringing two universities for a period of 8 month investigation with expertise in textile and garment handling and in flexible material manipulation and robotic gripper technology. The research programme looked into the dullest domestic chore which has not been changed for many hundreds of years, examined the existing techniques and required disciplines for automating the ironing process, established the detailed functionality of a robotic ironing device, and determined in detail the research needed to carry out the work. Successful results have been produced with four accepted international conference papers, six journal papers in preparation and 20 internal reports. A workshop with potential industrial partners was held at the end of the project and a number of presentations were given with technology studies into robotic ironing. The research further resulted in three proposals in preparation for submission to EPSRC and European Commission, and generated substantial impact on the advancement of robotics application to domestic tasks and aroused huge media interest including articles in The Engineer and in national newspapers. Project aims and objectives The aims and objectives of this are: (1) establish the detailed functionality of a robotic ironing device; (2) examine the existing techniques and required disciplines to solve the technical problems; (3) determine in detail the research needed to carry out the work, the team to do it and the resource required; (4) identify potential industrial collaborators who might provide commercial follow-through; and (5) draft research proposal(s) to fund the research programme.
71
IJCST 15,6
72
Research deliverables (academic and industrial) (1) Establish the detailed functionality of a robotic ironing device. (2) Examine the existing techniques and required disciplines to solve the technical problems. (3) Determine in detail the research needed to carry out the work, the team to do it and the resource required. (4) Identify potential industrial collaborators who might provide commercial follow-through. (5) Draft research proposal(s) to fund the research programme. Publications Dai, J.S., Taylor, P.M. and Sanguanpiyapon, P. (2003a), ‘‘Folding and ironing motion analysis in robotic ironing’’, Proc. INTEDEC 2003, Fibrous Assemblies and the Design and Engineering Interface, September 2003, Edinburgh. Dai, J.S., Taylor, P.M., Liu, H. and Lin, H. (2003b),‘‘Modelling, analysis and control issues in robotic ironing’’, Proc. INTEDEC 2003, Fibrous Assemblies and the Design and Engineering Interface, September 2003, Edinburgh. Dai, J.S., Taylor, P.M., Liu, H. and Lin, H. (2004), ‘‘Garment handling and corresponding devices – technology in robotic ironing’’, Proc. 11th IFToMM World Congress on Mechanisms and Machine Science , Tianjin, China. Taylor, P.M, Dai, J.S., Lin, H. and Liu, H. (2003), ‘‘Technologies for automated ironing’’, Proc. INTEDEC 2003, Fibrous Assemblies and the Design and Engineering Interface, September 2003, Edinburgh.
London, UK London College of Fashion, The London Institute, 20 John Princes Street, London W1G 0BJ, UK Tel: 020 7514 7690; Fax: 020 7514 7672; E-mail:
[email protected] or
[email protected] Zane Berzina Research staff: Supervisory Team: Dr Frances Geesin, London College of Fashion – Director of the Studies Prof. Norma Starszakowna, Director of Research Development, The London Institute (First supervisor) Kay Politowicz MA (RCA) Course Leader, Textile Design, Chelsea College of Art and Design (Second supervisor) External Advisors: Dr Klaus Hausmann, Institute for Biology and Zoology, Freie Universita¨t Berlin, Germany Colin Dawson, Material Scientist, ex-MOD
Skin stories – charting and mapping the skin: research using analogies of human skin tissue in relation to my textile practice
Research register
Other partners: Academic Industrial University of Arts, Berlin, Germany Supporters: Colbond Nonwovens Institute for Biology and Zoology, Cornelius Outlast Technologies Freie Universita¨t Berlin, Germany Hallcrest Project started: October 2000 Project ends: March 2004 Finance/support: £7.5000,-/year Source of support: The London Institute Keywords: Intelligent textiles, Smart materials, Textile design, Interior design, Art and science, Bio-mimetic design
73
My practice led research project reflects on the current developments within material research and technologies, their properties and the possibilities of their application in design. The research contributes to multidisciplinary activities in science, art and design and demonstrates how they might influence the cultural, social and economic environment. It is proposed that the principle aim of this body of work is to investigate new design possibilities for interactive and functional mixed-media textile surfaces for interiors by using analogies of human skin tissue in relation to my textile practice. Key aspects of my research project are: . to explore the potential of textiles as a latent heating system to control room temperature (the analogy of skin being a thermo-regulator); . to examine the thermochromic and photochromic properties of textiles as indicators of fluctuating interior conditions (the analogy used is that of human skin reactions to physical and psychological stimuli – skin as a sensor and biochemical mechanism); . to investigate the interactive and decorative potential of thermochromic and touch-sensitive surfaces to exploit transient images and patterns of the skin (the analogy used is skin as a sensor); . to explore the olfactory and filtering potentials of textiles as a deodorising, anti-microbial and curative surface (analogy – the skin as immunological surveillance and biochemical mechanism). Project aims and objectives Aims of the investigation are:
IJCST 15,6
74
to examine the technical and practical processes within textile design by focusing my practice-led research on biological and medical aspects of human skin and body surface, in particular scientific imaging, magnified anatomical structures and textures (cells, fibres); its physical characteristics; . to translate these phenomena into a textiles vocabulary and following the principles of bio-mimetic design, combine aesthetic and functional aspects of skin characteristics; . to develop and test new applications and technologies within textile design, which arise from the research, particularly focusing on aspects of membrane and display, which embodies protection, identity, communication, camouflage; . to develop mixed media textile surfaces as a result of the process of scanning and mapping the surface of the body, using practical simulation and re-making of the skin’s physical and functional characteristics into fabrics by means of bonding, coating, layering, 3D-moulding, dyeing and various print processes; and . to produce a body of textile work accompanied by a written thesis. The overall intent of this research project is to develop functional textile membranes, which enable individuals to experience a dynamic polysensual and interactive environment. It is anticipated that the new intelligent design concept should respond to peoples needs, enable them to enhance their sense of wellbeing and offers them the possibility to interact with their surroundings by creating an ambience according to their own requirements at that particular moment. In order to do this, I am investigating the biological human skin tissue as a ‘‘technology’’ using biomedical research methods to examine the properties and functions of the epidermis. By doing so I am selecting relevant skin properties for translation into my design work. The focus is on criteria for inclusion in the interactive and polysensual textile installation for interiors, in addition to the regular interior functions anticipated. A series of technical experiments are executed to facilitate the incorporation of selected aspects of functionality. .
Research deliverables (academic and industrial) . Demonstration of textiles concept for interiors which is based on the analogies of human skin tissue and that by involving new technologies and innovative materials, enhances people’s well-being as well as enabling individuals to experience a polysensual and responsive environment by interacting with their biological senses. I hope that this will provide extensive evidence of the potential and the huge resources of opportunities still to be recognised by the textile industry: in the fashion and clothing sector, interiors sector, as well as in the technical textiles sector.
.
.
.
Demonstration of how scientific thought influences contemporary art and design practice and vice versa: how art and design practitioners have the potential to influence scientific processes. Organisation of an exhibition featuring the outcomes and the process of my cross-disciplinary research project ‘‘Skin Stories – Charting and Mapping the Skin’’ which emphasises the future developments in research, in general, and will be the result of an active collaboration between professionals from a whole variety of backgrounds and disciplines: engineering, science, design, process development, and business and marketing. Audience will be able to literally follow the ‘‘red thread’’ which led me from the science of biology and material science to the actual textile concept. Moreover, they will be encouraged to explore their biological senses by interacting with artworks, to test the responsive skin-like properties of certain materials by touching, smelling and viewing them. My hands-on research project is a tangible try to bridge the gap between design, materials science and technology.
Publications Berzina, Z. (2003a), ‘‘Inteligentais tekstila dizains (Intelligent Textile Design)’’, Latvijas Architektura, Latvia, Vol. 45 No. 1 pp. 58-61 Berzina, Z. (2003b), ‘‘Topography of skin’’, Ballettanz, Germany, pp. 78-9 Hausmann, K. (2003), ‘‘Hautgeschichten – skin stories. Zane Berzina’’, Mikrokosmos, 4 July, Germany, pp. 241-3.
Louisiana, USA School of Human Ecology, Louisiana State University, LSU, Baton Rouge, LA 70803-4300 Tel: (225)578-2407; Fax: (225)578-2697; E-mail:
[email protected] Yan Chen and Jianhua Chen, LSU School of Human Ecology, LSU Department of Computer Sciences Teresa Summers, Jackie Robeck, Al Steward, and Ramesh Kolluru
Online fabric sourcing database with data mining and intelligent search Other partners: Academic Industrial University of Louisiana-Lafayette None Project started: 1 June 2001 Project ends: 30 June 2004 Finance/support: $119,822 Source of support: LA Board of Regents
Research register
75
IJCST 15,6
76
Keywords: Fabric properties, Fabric sourcing, Tailorability, Drapability, Online database With rapid development of Internet for business applications and e-commerce, textile manufacturers, garment makers, and clothing retailers are eager to go online for fashion tracking and material sourcing. The goal of this project is to establish an online intelligent database that will help the industry users to locate desired fabrics that match fashion trends in color, drape, and style; to narrow fabric selections to fabrics possessing good physical properties that insure high garment quality; to find better-buy fabrics; and to locate fabric manufacturers and determine earliest shipping date. Major objectives include: (1) construction of a database server using a PC and the Oracle software; (2) establishment of a dynamic database composed of fabric structural parameters, mechanical properties, drape images, tailorability, and manufacturers’ contacting information; (3) development of an intelligent search engine allowing clients to scour the database for their own priorities; and (4) investigation of new search patterns that relate client’s fashion requirements to fabric properties using the new data mining techniques of fuzzy clustering and decision tree approach. The new database will merge an existing database created by faculty at the Apparel-Computer Integrated Manufacturing Center (ACIM) at UL Lafayette. This existing database is providing information of the Louisiana Textile, Apparel, and Retail Consortium that aids the state economic development. With accessibility to this database, the new data resources, new search engine, and new search patterns developed in this research can be applied to this existing database. This will greatly enhance its functionality and help form a united textile and clothing sourcing database in the state and US. Research deliverables (academic and industrial) Online intelligent database and web service. Publications Jianhua Chen, Yan Chen, Bin Zhang and Ayse Gider (2002), ‘‘Fuzzy linear clustering for fabric selection from on-line database’’, 2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings, IEEE System, Man and Cybernetics Society, New Orleans, LA, pp. 518-22.
Manchester, UK Department of Clothing Design and Technology, Manchester Metropolitan University, Old Hall Lane, Manchester M14 6HR, UK Tel: 01612472632; Fax: 01612476329; E-mail:
[email protected] Dr J.E. Ruckman, Prof. M.K. Song
Heat and water vapour transfer through high performance clothing systems Other partners: Academic Industrial None None Project started: April 2002 Project ends: March 2004 Finance/support: £15,000 Source of support: British Council Keywords: High-tech fabrics, Layered clothing system, Mass transfer When technical fabrics are used in a clothing system it is to be expected that the performance of a fabric itself is not the only factor which contributes to thermophysiological comfort. In real life technical fabrics that are developed to suit outdoor activities are rarely worn on their own, but are incorporated into a layered clothing system, especially that incorporating a waterproof breathable fabric as an outer shell. For this reason, the characteristics and properties originally developed for specific end-uses (and evaluated using testing methods based upon Fick’s Law) may not be as decisive as originally anticipated. The aims of this project are to investigate the heat and water vapour transfer through high performance clothing systems and to identify the optimum combination of technical fabrics for each layer of a high performance clothing system.
Maribor, Slovenia Faculty of Mechanical Engineering, University of Maribor, Smetanova ulica 17, SI-2000 Maribor, Slovenia Tel: +386 2 220-7960; Fax: +386 2 220-7990; E-mail:
[email protected] Associate Prof. Dr Sc. Jelka Gersˇak, Department for Textiles, Institute of Textile and Garment Manufacture Processes Research staff: Research Unit Clothing Engineering, Research Unit Textile Technology
Clothing engineering and materials Other partners: Academic Industrial None None Project started: 1999 Project ends: 2003 Finance/support: 15.052.943,00 SIT or 65.840,62 ECU for 2001
Research register
77
IJCST 15,6
78
Source of support: Ministry of Education, Science and Sport Keywords: Clothing, Fabric, Mechanical properties, Behaviour, Prediction The research programme was based on three main activities: basic research on fabric mechanics regarding the non-linear mechanical fabric properties at low stresses and search for model of a fabric as shell; study of response of a fabric against acting stresses in garment manufacturing processes; and study of a behaviour between fabric’s mechanical properties and quality of a produced garment. In frame of the first group of activities, the research was focused on study of a fabric as shell. We have studied the fabric behaviour from the point of view of continuum mechanics. This work resulted in mechanical model of a fabric that was described with rheological coefficients, i.e. elastic and shear module and Poisson’s number. Furthermore, the fabric was modelled using the finite elements method and programme package ABAQUS. The second group of activities referred to the study of fabric response to acting stresses in garment manufacture processes and contained: (1) Study of relationship between tensional stresses and deformations of fabrics. Based on the study of the relationship between the parameters of mechanical properties of analysed fabrics and their response to acting tensional stresses, resp. resulted deformation, it was stated that deformation degree and relaxation time directly depended on mechanical properties of fabrics, acting load and length of the fabric layers. (2) Study of fabric behavior in garment manufacturing processes, which was focused above all on fabric response to acting stresses during cutting, fusing and finishing. The research resulted in: . definition of relationships between fabric mechanical properties and their behaviour in different garment manufacturing processes, . definition of potential problematic spots in manufacturing processes and limit values of particular mechanical properties of fabrics, . set-up of a model ‘‘NAPOVED1.1DZ’’ for prediction of fabric behaviour in garment manufacturing processes. The model was designed in Microsoft Access in such a manner that all respective data, i.e. parameters of mechanical properties were joined together using appropriate relation functions Study of relationship between the mechanical properties of fabrics and quality of produced garments was carried out in the frame of the third group of research activities. The aim was to design the model for qualitative prediction of garment appearance quality using the principles of objective evaluation of the quality level of garment appearance and comparable estimation of garment suit.
Project aims and objectives Project aims are: definition of principles of fabric behaviour in garment manufacturing processes regarding the non-linear mechanical properties of a fabric, set-up of a model for prediction of fabric behaviour in garment manufacturing processes, definition of relationships between fabric mechanical properties and quality of a produced garment, design of a model for prediction of garment appearance and set-up of a model of fabric as shell. Research deliverables (academic and industrial) Achieved knowledge in a field of fabric mechanics and fabric response to acting stresses, resp. fabric behaviour in garment manufacturing processes, as well as defined limit/critical values of fabric mechanical properties represent an important contribution of the research from the scientific as well as applied point of view. Designed model for prediction of fabric behaviour in garment manufacture processes, ‘‘NAPOVED1.1DZ’’, which comprehends achieved cognitions of basic research on fabric mechanics, and designed knowledge base in Microsoft Access, can be also stated as important applied results of the research. The other important applied achievement is designed simulation of a fabric with the help of a programme package ABAQUS. Fabric model is designed on the basis of numeric modelling as a starting point for study of fabric draping and formability. Furthermore, using appropriate programming tools it will be possible to carry out the simulation of a garment suit. Publications Gersˇak, J. (2002a), ‘‘Development of the system for qualitative prediction of garments appearance quality’’, International Journal of Clothing Science and Technology, Vol. 14 No. 3/4, pp. 169-80. Gersˇak, J. (2002b), ‘‘A system for prediction of garment appearance’’, Textile Asia, Vol. 33 No.4, pp. 31-4. Gersˇak, J. and Zavec, D. (2000), ‘‘Creating a knowledge basis for investigating fabric behaviour in garment manufacturing processes’’, Annals of DAAAM for 2000 and Proceedings of the 11th International DAAAM Symposium Intelligent Manufacturing and Automation: Man-MachineNature, DAAAM International, Vienna, pp. 155-6. Jevsˇnik, S. and Gersˇak, J. (2001), ‘‘Use of a knowledge base for studying the correlation between the constructional parameters of fabrics and properties of a fused panel’’, International Journal of Clothing Science and Technology, Vol. 13 No. 3/4, pp. 186-97. Zavec, D. and Gersˇak, J. (2001a), ‘‘Modular development of prediction knowledge base’’, Annals of DAAAM for 2001 and Proceedings of the 12th International DAAAM Symposium Intelligent Manufacturing and Automation: Focus on Precision Engineering, DAAAM International, Vienna, pp. 519-20. Zavec, D. and Gersˇak, J. (2001b), ‘‘Prediction of fabric behavior as an input information for garment manufacturing process (Napoved obnasˇanja tkanin kot vhodna informacija za proces izdelave oblacˇil)’’, Tekstilec, Vol. 44 No. 9/10, pp. 271-9.
Research register
79
IJCST 15,6
80
Michigan, USA Manchester Metropolitan University, Hollings Campus, Old Hall Lane, Manchester, M14 6HR Tel: 0161 247 2636; Fax: 0161 247 6354; E-mail:
[email protected] Department of Clothing Design and Technology David J. Tyler Research staff: Julie D. Wilson
Mass customisation for the interior textiles sector Other partners: Academic None
Industrial Acton and Acton Ltd Direct Textile Imaging Ltd John Clegg and Bros Ltd Project ends: 31 January 2004
Project started: 1 September 2002 Finance/support: £28,000 Source of support: Department of Trade and Industry Keywords: Textile printing, Ink-jet technology, Cost modelling
Until recently, the market for the digital printing of textiles has been dominated by sampling. There have been some applications involving production for consumers (ties, flag, banners) but the volumes have not been large. Advances in machinery technology have created new business opportunities. Wide-width printers suitable for textile applications have been introduced to the market with significant productivity improvements. There are now opportunities for businesses to move from sampling into consumer products. The research involves the analysis of costs in the supply of digitally-printed textile products. Sampling businesses generally provide a design service to interpret the needs of the customer and translate it into printed products. Such businesses do not have a high utilisation of their printing machinery and much of the cost is associated with set-up. The productivity of the printer is not a primary consideration. As the core business moves from a sampling service towards the supply of printed textiles, an internal culture change is needed. The productivity of the machinery is an issue, and there are significant benefits from batch-printing without the need for operator supervision. The cost of the ink (dye) becomes a more important element in the price structure. Pre- and post-printing treatments are significant costs affecting the price to consumers.
This research examines these cost issues from the perspective of prioritising cost factors and identifying targets for technology innovation. The findings are relevant to the wider implementation of digital printing technologies in the textiles sector. Project aims and objectives . To analyse the cost structure of textile digital printing. . To develop an economic cost model. . To develop tools for evaluating technology developments. Research deliverables (academic and industrial) Validated economic cost model. Publications None
Newcastle upon Tyne, UK University of Newcastle upon Tyne, Stephenson Building, University of Newcastle, Newcastle Upon Tyne NE1 7RU, UK Tel: 0191-2227145; Fax: 0191-2228600; E-mail:
[email protected] Paul M Taylor, School of Mechanical and Systems Engineering Research staff: Hua Lin
A feasibility study into robotic ironing Other partners: Academic Industrial King’s College London None Project started: 1 August 2002 Project ended: 28 February 2003 Finance/support: £38,742 Source of support: EPSRC Keywords: Garment, Ironing, Robotics This is an adventurous research aiming at investigating a development in applying robotics techniques to one of the most demanding household activities, performing a feasibility study into robotic ironing. Customer market research will be carried out to establish the minimal functional requirements for a range of potential users. Technical requirements will be established for complete and decomposed ironing tasks. A preliminary
Research register
81
IJCST 15,6
82
technology study will then be made, covering the relevant technologies in the UK, Europe, Japan and the USA to establish the state of the art and to determine advances that must be made. These technologies will cover gripping, handling, folding and manipulation, ironing and relevant textile technologies. Potential machinery manufacturers will be approached to bring their perspective into the study. Gaps in theories and knowledge will be identified and these will be used to determine the further research that must be carried out to provide solutions to the technical problems in a way that should eventually lead to a product acceptable to the consumer. Finally, a consortium will be established which could carry out this research and proposal(s) will be prepared. Project aims and objectives . Establish the detailed functionality of robotic ironing devices. . Examine the existing techniques and required disciplines to solve the technical problems. . Determine in detail the research needed to carry out the work, the team to do it and the resource required. . Identify potential industrial collaborators who might provide commercial follow-through. . Draft research proposal(s) to fund the research programme. Publication These surveys will be published in appropriate journals, such as the International Journal of Clothing Science and Technology, Journal of Robotic Systems, and at the 11th IFToMM World Congress in Tianjing in 2003, and at the IEEE ICRA in 2003.
Newcastle upon Tyne, UK University of Newcastle upon Tyne, Stephenson Building, The University, Newcastle upon Tyne NE1 7RU Tel: (0191) 222 7145; Fax: (0191) 222 8600 Professor P.M. Taylor, Department of Mechanical, Materials and Manufacturing Engineering Research staff: D. Pollet
Vibration of fabric panels and automated garment assembly Other partners: Academic None Project started: 1 October 1992
Industrial None Project ends: –
Keywords: Bending, Environment, Friction, Grippers The primary aim is to understand the way fabrics and garments interact with mechanical devices designed to hold them and move them around and how their behaviors are affected by environmental changes. Studies are being undertaken on an analysis of the behavior of fabric during the pinch gripping operation and on how fabric panels move on vibratory surfaces. To complement this, the relevant properties of fabrics are being studied, particularly buckling, bending and friction under zero and low applied normal forces. Friction and bending tests are also being undertaken over a wide range of environmental conditions to see the effects of humidity changes on handling processes. Project aims and objectives The primary aim is to understand the way fabrics and garments interact with mechanical devices designed to hold them and move them around and how these are affected by environmental changes. Academic deliverables Gripping analysis – vibration analysis, new instrumentation, results showing strong links between handling behavior and environmental conditions. Industrial deliverables None yet Publications Taylor, P.M. and Pollet, D.M. (1996), ‘‘Why is automated fabric handling so difficult?’’, 8th International Conference on Advanced Robotics (ICAR 97). Taylor, P.M., Pollet, D.M. and Griesser, M.T. (1994), ‘‘Analysis and design of pinching grippers for the secure handling of fabric panels’’, Proceedings of Euriscon ’94, Vol. 4, Malaga, Spain, 22-26 August, pp. 1847-56.
Ontario, Canada University of Guelph, Ontario, Canada, N1G 2W1 K. Slater, School of Engineering Research staff: various graduate students
Protective clothing design for agricultural uses Other partners: Academic Industrial None None Project starts: April 1999 Project ended: April 2003 Finance/support: Applications under development
Research register
83
IJCST 15,6
Source of support: Various groups to be approached Keywords: Agriculture, Protective clothing
84
The project depends on the ability of textile materials to be incorporated into designs of garments which can resist the ingress of harmful chemicals yet allow the escape of perspiration moisture. Preliminary design considerations have been established, but continuation of the work depends on the successful negotiation of adequate funding, a step which is currently in progress. Project aims and objectives The aim of the project is to develop a clothing system capable of providing agricultural workers with adequate protection from the various chemical and microbiological hazards which they continually encounter in their daily work. Academic deliverables One or more graduate theses. One or more journal articles. Industrial deliverables Protective clothing capable of preventing health deterioration in agricultural workers continually exposed to harmful chemical or microbiological hazards, with the added advantage of being comfortable enough for the workers to accept it without demur. Publication No publications stemming directly from this project have appeared to date, but some of my earlier work (e.g. protective clothing for operating room use) is relevant and has appeared in the past. I have also presented papers dealing with the need for protection of agricultural workers at several recent conferences.
Ontario, Canada University of Guelph, Ontario, Canada, N1G 2W1 K. Slater, School of Engineering Research staff: various graduate students
Protective clothing design for industrial use Other partners: Academic Industrial None None Project starts: April 1999 Project ended: April 2003 Finance/support: Applications under development
Source of support: Various groups to be approached Keywords: Clothing, Industrial clothing, Protective clothing Industrial accidents frequently cause injuries which could have been prevented by the use of appropriate protective clothing. Flying projectiles, violent contact with machinery, vehicles or the ground, and exposure to harmful chemical or biological materials are encountered regularly in accident reports. This project is intended to build on my previous research in the degradative changes occurring in textiles, and on my comfort research, to match the protective needs of clothing intended to safeguard human beings from the hazardous conditions to which they are exposed. A major need is to ensure that wearers are not likely to discard the protective garments for reasons of physical or mental comfort, so that protection is abandoned. As the work is still in the planning stage, it is not possible to provide any detailed synopsis of its course. Project aims and objectives The aim of the project is to use textile materials, in conjunction with other components, to prevent (or minimise injury from) industrial accidents. Academic deliverables One or more graduate theses. One or more journal articles. Industrial deliverables Protective clothing capable of reducing or eliminating injury, and hence reducing financial costs, arising from workplace accidents. Publication No publications stemming directly from this project have appeared to date, but some of my earlier work relates closely to the needs of this research.
Philadelphia, USA School of Textiles and Materials Technology, Philadelphia University, School House Lane and Henry Avenue, Philadelphia 19144, USA Tel: 215-951-2680; Fax: 215-951-2651; E-mail:
[email protected] Dr Mohamed Abou-iiana, Textiles Department Research staff: S. Youssef
Research register
85
IJCST 15,6
86
Knitting and knitting related projects, knitted composites, . . . etc on-line weight and shrinkage control of knits Other partners: Academic Auburn University
Industrial National Textiles, Greensboro, NC, USA Project ends: May 2004
Project started: May 2001 Finance/support: $200,000 Source of support: National Textile Center
An automatic fabric evaluation system has been developed to automatically analyze the knit structures and objectively evaluate fabric properties. Fabric images are captured by CCD camera and preprocessed by Gaussian filtering and histogram equalization. Fabric construction parameters such as courses per inch, wales per inch, fabric cover, weight per unit length are measured and evaluated. The structural changes occurred to the fabric at different levels of fabric relaxation were documented. It has been shown that the system is capable of capturing the structural changes during stress relaxation. This system can be used to on-line control of knit structures during processing by having this image quality acquisition probe determine the spatial characteristics of knitted loop before and after wet treatment. For years knitting has been considered more of an art than a science. Many attempts have been made over the past century to quantify the characteristics of knitted fabrics. The key to unlocking a knitted structure lies within its basic element, the single knitted loop. It has been shown that the length of yarn knitted into a single loop will determine such overall fabric qualities as hand, comfort, weight, extensibility, finished size, cover factor and most importantly fabric dimensional stability. Therefore, to gain control over the characteristics of the fabric performance, the single knitted loop must be controlled to meet certain performance criteria. The problem then arises of how to determine that a knitted loop is of the correct size and shape for a given set of fabric properties. The answer lies in the ability to objectively measure the knitted loop size/shape during processing. Once the loop shape in a fabric is measured, the loops of that size/shape can then be correlated with specific properties of that fabric. With the advent of computers, and more specifically of image analysis and processing, this age old problem of measuring a knitted loop size/shape has been solved [1,2,3,4,5,6,7,8]. Today during fabric processing, a loop can characterize in a matter of seconds with great accuracy instead of the traditional inaccurate techniques of measuring the course spacing or courses per unit length. Computers have not only provided an accurate method for characterizing the loop shape but also a
means for checking the loop shape to the required shape to achieve certain fabric properties. The knitted structure consists essentially of a yarn bent into the shape of a loop, and this basic element, the loop repeated across the width of the fabric and along its length. The distinctive property of a knitted fabric is its high extensibility in both length and width, which gives it the ability to take up the shape of the wearer and allows it to fit. Attempts to specify the dimensional properties of a knitted fabric in terms of length or width parameter (courses per inch and wales per inch) are subject to high degree of inaccuracy because of its inherent stretch at low loads and poor recovery. Nevertheless, the construction of a fabric is still today frequently described in terms of courses and wales per inch. It is the use of this unreliable and inaccurate parameter for specifying the tightness of a knitted construction, which is directly or indirectly responsible for many of the problems associated with the control of dimensions of knitted structures. This characteristic of a knitted fabric when strained in length or width is due to the fact that the loop shape is easily distorted under low strain conditions and is caused by a change in loop shape without any associated stretching of the yarn forming the loop. In an industrial setting, the techniques of counting the course per inch, wales per inch or unraveling the fabric to determine stitch length are subjected to human error and time consumption. An automatic structure analysis and objective evaluation of knit structures using image analysis techniques will determine the fabric construction parameters and eliminate the subjectivity of the human element.
Pisa, Italy Interdepartimental Research Center ‘‘E. Piaggio’’, Faculty of Engineering, University of Pisa, Via Diotisalvi 2, 56126 Pisa, Italy Tel: ++39-050-553639; Fax: 550650; E-mail:
[email protected] Prof. Danilo De Rossi Research staff: Ing. Alberto Mazzoldi, Ing. Enzo Pasquale Scilingo, Dr Federico Lorussi, Ing. Alessandro Tognetti, Ing. Federico Carpi
Wearable health care system, WEALTHY Other partners: Academic
Industrial Millior (I)
Research register
87
88
Institut National des Sciences Centre Suisse d’Electronique et de Appliquees de Lyon (F) Microtechnique (CH) Istituto Scientifico H San Raffele (I) Atkosoft (EL) Centre de Recherches du Service de Messe Frankfurt (D) Sante des Armees (F) Project started: 1 September 2002 Project ends: 28 February 2005 Finance/support: 3.6 10 6 Euro (Total costs) Source of support: EC Keywords: Wearable sensors, Healthcare, Interface V
IJCST 15,6
The WEALTHY activity is planned over 30 months. The specification of the system and the pilot application will be defined as the result of elaboration of user needs and technological requirements. The implementation of software modules will be conducted in parallel in order to assess the integration feasibility among submodules. For this reason the first achievement of the integration is to evaluate the development phase results by establishing some checkpoints in order to make adjustments, if needed, on the ongoing implementation. The following steps need to considered: . A new multifunctional fabric, integrating smart material in form of fiber or yarn, will be realised. Advance textile technology will be used in the component integration also considering comfort and fitting. The wearable garments will be tailored in different shapes. The position and the number of strain gauge sensors and electrodes will be determined in terms of the anatomical location producing the signal and the signal/noise ratio. Alternative sensors will be added, if needed, to increase the number of biophysical variables to be monitored. . The portable part incharge of medical signal acquisition as well as information and communication management will be realised. The implementation of the future UMTS standard will also be examined. . Data processing and representation module will perform the acquisition of physiological parameters, instantiation of patho-physiological model, diagnosis, feedback generation to the patient and to the medical team. Security mechanism will be implemented for controlling the authorisation to access and the manipulation of data and protecting sensitive information. Project aims and objectives The main objective of WEALTHY is to set up a wearable healthcare system that will improve patient or user autonomy and safety. WEALTHY building blocks are:
cost-effective, non-invasive system based on wearable and wireless instrumented garments, which are able to detect user specific physiological signals; . intelligent system for data representation and alert functions for creating intelligent feedback and deliver information to a target professional; . electronic devices for signals transmission by using 3G wireless network, allowing monitor the patient ‘‘anywhere’’; . advance telecommunication protocols and services; . effective and user-friendly data format. WEALTHY solution will be validate in 2 pilot sites, with an active participation of users and health care institutions. .
Research deliverables (academic and industrial) Intermediate results will generate mock-up versions of the WEALTHY platform, available in month 13. These comprise a mock-up interface, a mock-up device and a mock-up monitoring system. In month 19 the versions of the components and service will be available: beta interface, beta device and beta monitoring system. Publications De Rossi D., Della Santa A. and Mazzoldi A. (1999), ‘‘Dressware: wearable hardware’’, Material Science and Eng. C, Vol. 7, pp. 31-5. De Rossi D., Lorussi F., Mazzoldi A., Orsini P. and Scilingo E.P. (2002), ‘‘Active dressware: wearable kinaesthetic systems’’, Sensor and Sensing in Biology and Engineering (in press). Scilingo E.P., Lorussi F., Mazzoldi A. and De Rossi D. (2002), ‘‘Strain sensing fabrics for wearable kinaesthetic-like systems’’, IEEE Sensors Journal (in press).
Port Elizabeth, South Africa Centre for Fibers, Textiles and Clothing, CSIR, P.O. Box 1124, Port Elizabeth 6000, South Africa Tel: 0415 0 83273; Fax: 0415 8 32325; E-mail:
[email protected] Dr Rajesh Anandjiwala, Dr Illona Racz Research staff: Michael Mkhize, Gabriella S.
Preparation and properties of b-nucleated polypropylene nanocomposites Other partners: Academic None
Industrial BAYATI, Hungary
Research register
89
IJCST 15,6
Project started: 1 April 2002 Project ends : 31 March 2004 Finance/support: R550,000 Source of support: CSIR – Parliament Grant+NRF Bilateral Funds Keywords: Nanocomposite, Polypropylene, Packaging, Composite sheets
90
Intensive research on nanocomposites started about 10 years back, when Toyota developed and used the first polyamide nanocomposite. The composite was prepared by in situ polymerization from caprolactone and layered silicate filler. The main advantages of the composite containing only a few percent of filler were its high stiffness, high strength and improved thermal resistance, which were naturally accompanied by a significantly lower mass. Since then, intensive research is carried out in many laboratories and at numerous companies all over the world in order to produce nanocomposites with the most diverse matrices. The most important projected application fields of the present proposal are the automotive and packaging industries. The packaging industry is interested in the new materials because of the significantly decreased permeability of films produced from the composite compared to simple polymeric matrix. Isotactic polypropylene (iPP) is a crystalline polymer, which is manufactured all over the world in great mass and applied for variety of end-uses. iPP is not in the range of the technical or engineering polymers traditionally, but functionally it is so widely used that it is allowed to label it as technical material. Hungary is one of the biggest polypropylene producers in the East-and Central-European region, therefore, both the application development and the research focus are essential. iPP is a polymorphic material, it can crystallize in three different crystal modifications, such as the monoclinic a-form, the trigonal b-form, and the orthorhombic g-form. Traditional types of the iPP forms predominantly aiPP under processing circumstances and the b-modification is observed only occasionally during the crystallization, and it appears as a minor constituent in the structure. The relative amount of the b-modification can be increased by special methods. g-form starts up in the case of small molecular mass or partially degraded iPP ingredients, or rather during the crystallization process of small co-monomer weight propylene random copolymers. Crystallization under high pressure also assists the inducement of g-form rich products. In the past decade the need for basic and applied research, which are related to the b-iPP increased greatly when research proved that several additive actuate as b-nucleator. In this manner, b-modification-rich or pure b-modification can be achieved. The industrial importance of b-iPP is the outstanding impact grade, which is multiple (up to four or fivefold) of the aiPP and may exceed the characteristics of the copolymers with the same
melt index. Ductility index, tensile and impact energy together with bursting strength values are better than that of the a-iPP. Beyond that b-iPP turned out to be satisfactorily good for some special applications, such as: . roughened surface films, . microporous membranes, . films with higher burst factor, . filaments with high adsorptive capability, . constructional units of impact-resistant piping systems, and . helps the deformability and stability of shape in the case of thermoforming. The improvement of these characteristic data affects deterioration in the strength of the b-modification (like Young’s modulus or necking stress), independently from the type of the iPP. These values are 10-15 percent lower when compared to the a-iPP. Project aims and objectives In the scientific literature – although several teams carry out research on polymer nanocomposites or b-polypropylene – no data were found on b-nucleation of polypropylene in nanocomposites. By developing nanocomposite with b -polypropylene matrices the advantages of each component is intended to combine. By using layered silicate depending on the type of nanocomposite – intercalated or delaminated – the strong improvement of thermal and/or mechanical and transport properties is anticipated. By the application of cellulose-based nanoparticles, the improvement of mechanical properties is expected, at the same time the composite will be at least partly biodegradable. Research deliverables (academic and industrial) The proposed project on developing nanocomposites with b -polypropylene matrices will enable us in developing new products for automotive applications such as: . injection moulded components, . door bracing, . thermoformable sheets (e.g. rear window shelf), and . presentation and publication of results. Publications None
Research register
91
IJCST 15,6
92
Port Elizabeth, South Africa Centre for Fibers, Textiles and Clothing, CSIR, P.O. Box 1124, Port Elizabeth 6000, South Africa Tel: 0415 0 83273; Fax: 0415 8 32325; E-mail:
[email protected] Dr Rajesh Anandjiwala
Buckling phenomenon in woven fabric Other partners: Academic Industrial None None Project started: 1 April 2003 Project ends: 31 March 2005 Finance/support: R300,000 Source of support: CSIR – Parliament Grant Keywords: Buckling, Woven fabric, Sewing automation The automation of labor intensive garment manufacturing operation is a key to enhance competitiveness of clothing and apparel industries, particularly in the developed nations where unit labor cost is very high. The problem of automation is attempted in some clothing research centers as modular manufacturing similar to what is applicable in the manufacturing of many engineering goods. However, the automatic feeding of the fabric is notoriously attributed to inherent the flexibility of the structures with difficulty. The fabric tends to buckle even under small forces while feeding to the sewing machine and therefore human intervention is unavoidable. The accurate estimation of critical buckling force is therefore necessary to understand the onset of buckling. Current fabric buckling models are deficient in this respect due to the hypothesis on which they are based. A more precise, computer driven model is required, which can provide information on critical buckling force to the feeding robot. In this proposal, we will develop a model that can be utilized for the automatic feed mechanisms in physical phenomenon such as sewing. We propose to modify the assumptions on which the current fabric buckling models are built. A more realistic assumption based on real phenomena of moment – curvature relationship and fabric frictional resistance will be considered to estimate the critical buckling load. The information can then be utilized to develop an automatic feeding mechanism for sewing machine. Project aims and objectives Development of new fabric buckling model was based on more realistic assumptions and then analyzing through computer driven packages.
The comparison of the model with real experimental behavior will be conducted to test the model.
Research register
Research deliverables (academic and industrial) . Papers in reputed journals. . A new approach for designing automatic feeding for the sewing machine.
93
Publications None
Port Elizabeth, South Africa CSIR Division of Manufacturing and Materials, Gomery Avenue, Summerstrand, Port Elizabeth, South Africa Tel:+27 41 5083282; Fax: +27 41 5832325; E-mail:
[email protected] Centre for Fibres, Textiles and Clothing S.A.Chapple Research staff: Dr F.A. Barkhuysen, Mr D. Qamse
Multi-functional natural fibres Other partners: Academic Industrial None None Project started: April 2002 Project ends: March 2004 Finance/support: Rand 300,000 (2003) Rand 170,000 (2002) Source of support: Internal funding – Government grant Keywords: Natural fibres, Cotton, Wool, Sol-gel, Microencapsulation, Grafting, Multi-functionality There has been an increasing demand for textile materials with special functionalities that can meet the existing and new application requirements. Such functionalities can be met by the development of new polymers, by using special fibre spinning processes and the modification of existing fibre surfaces by physical and chemical methods. Chemical methods include chemical modification (graft copolymerisation, alkalising, mercerisation, etc.), enzymatic treatments and application of chemicals (cross-linking, adsorption, inclusion, anchoring biopolymers, supramolecular chemistry, sol-gel coating etc.) Natural fibres have many unique properties, notably comfort and renewable/environmentally friendliness, however, it is not always easy to
IJCST 15,6
94
impart multi-functional or high performance characteristics, especially without affecting the inherent properties of the fibre. New technologies and new applications of technologies are emerging, for example, sol-gel coating, microencapsulation and grafting, which can be used to impart multi-functionality to fibres. These new technologies will be examined for use in imparting other functionalities, for example, retardency and repellency to natural fibres such as cotton and wool. Project aims and objectives The objective of the project is to develop novel products / processes for imparting additional functionalities to natural fibres. Research deliverables (academic and industrial) . To build up skills, knowledge and capacity in this field. . To develop at least one novel treatment for imparting additional functionality to natural fibres. Publications None
Riga, Latvia Institute of Textile and Clothing Technology, Riga Technical University, Latvia. Prof. Dr habil sc. ing, Austrums Klavins Research staff: Assoc. Prof. Dr habil sc. ing, V. Priednieks, Lecturer I. Ziemele, postgraduate (Master and doctoral study programs) students
Control, optimisation and monitoring of the stitch formation process in sewing machines Other partners: Academic None
Project started: 1969 Finance/support: N/A Source of support: N/A Keywords: Sewing machines, Stitch
Industrial Company ‘‘Promshveymash’’ Orsha, Byelorussia (1969-1991) Sewing Company ‘‘Latvia’’ from 1991 Project ends: No limit
Effective operation of sewing machines is of crucial importance for the qualitative production of sewn goods. It is possible to attain it by improving the mechanisms of the machine, controlling and monitoring them. These problems are being solved by applying mathematical methods of statistics, theory of probability and experimental design. The research is based on the unconformity of interaction of stitch formation tools (mechanisms) and needle thread as a complex parameter which permits one to estimate the process as a whole in each cycle and to simulate this process. It allows one to find out the impact of separate mechanisms, to improve the quality of the stitch formation process, so that it is possible to control, optimise and monitor them. On this basis the sewing machines are modernised or rationally used in mass production lines. Project aims and objectives . To work out the investigation methods of the sewing machine process. . To develop methods of the basic, complex parameters control, optimization and monitoring. . To develop practical methods for increasing the sewing machine’s serviceability, improving separate mechanisms of the machine. . To work out methods for the selective application of sewing machines for a rational organization of mass production lines in the garment industry. Academic deliverables . The following part of the study programs for Master’s has been worked out, namely, control and monitoring of the sewing machine process. . The results of the research have been used by doctoral students acquiring investigation methods. Industrial deliverables . Control, monitoring and improving the quality and serviceability of sewing machines. . Rational organization of garment mass production lines.
Publications Aizpurietis, A.V., Klavins, A.R., Poluhin, V.P. and Sharamet, U.I. (1988), ‘‘Raschot chetirjohzvennogo mechanizma nitepritjagivatelja shvejnih machin na EVM’’, Journal Tehnologia logkoi promishlennosti, Moscow, Vol. 6, pp. 94-7. Klavins, A. and Priednieks, V. (1998), ‘‘The quality improvement problems of the operation sewing machines and the prospects of the development of scientific research’’, Scientific Conference of ‘‘Technologies and Design of Consumer Goods’’, Kaunas University of Technology, 21-22 April.
Research register
95
IJCST 15,6
96
Klavins, A.R., Salenieks, N.K. and Rachok, V.V. (1974), ‘‘Diagramma ispolzovanija igolnoi niti’’, Machinostrojenie dija logkoi promishlennosti, Moscow, No. 3, pp. 3-7. Olshanskij, V., Fedoseyev, G. and Klavins, A. (1987), ‘‘Raschot parametrov prushinih kompensatorov shvejnih mashin’’, Journal Tehnologia logkoi promishlennosti, Moscow, Vol. 4, pp. 114-15. Priednicks, V. and Klavins, A. (1997), ‘‘The optimization of lockstitch formation system in sewing machines’’, The 78th World Conference of The Textile Institute in Association with the 5th Textile Symposium of SEVE and SEPVE, Vol. 11, Thessaloniki, Greece, pp. 195-204. Some Adjusting Techniques for Sewing Machines, Pat (USSR) 442252. IPCl D 05 B 69/24. The Method for Plotting Needle Thread Take-up Curve, Pat (USSR), Nr. 324322 IPCI. D. 05 B45/00. The Method for Plotting Needle Thread Take-up Curve, Pat (USSR), Nr. 461189 IPCl. D. 05 B45/00. Ziemele, I., Klavins, A. and Priednieks, V. (1997), ‘‘Selection of lockstitch sewing machine obtaining a high quality of thread joints in garment’’, abstract of the papers presented at the 26th Textile Research Symposium at Mt Fuji, Shizouka, Japan, 3-5 August, pp. 60-3.
Riga, Latvia Institute of Textile and Clothing Technology, Riga Technical University, Latvia Prof. Dr habil. sc. ing. Viktoria Kancevicha Research staff: Prof. Dr habil. sc. ing, V. Kasyanov, Assoc. Prof. Dr sc. ing. H. Vinovskis, postgraduate (Master and doctoral study programs) students
Development of new textile technology for manufacturing hybrid textile vascular grafts Other partners: Academic Latvian Medical Academy Project started: 1980 Finance/support: N/A Source of support: N/A Keywords: Technology, Textiles
Industrial None Project ends: No limit
In the fields of medicine and bioengineering extensive efforts have been directed to the development of various new types of vascular grafts using different technologies. The textile industry has a lot of practical experience in the production of different kinds of vascular grafts and allows a high production rate to be reached. Nevertheless there are practical needs for compliant grafts for patients with cardiovascular disease. Clinical implantation and chronic experiments on
animals with various grafts have indicated a fairly good correlation between their compliance and patency (especially for a diameter less than 6 mm) because the compliance vascular graft practically does not change the haemodynamics of the blood flow. Thus compliant vascular grafts having mechanical properties matching the human arteries are very promising for successful reconstruction operation and good patency. This problem of developing new textile technology and producing compliant vascular grafts is very important for Latvia because cardiovascular disease is very high – and not only in the Baltic states. The investigation of the peculiarities of the mechanical behavior and structure of human blood vessels is carried out at RTU. On this basis the new structure of the hybrid textile materials is developed. Using the system of two threads having substantially different modulus of elasticity, it is possible to model peculiarities of the biomechanical behavior of the arterial tissue. Project aims and objectives . Development of the new textile technology for manufacturing novel hybrid compliant vascular grafts using knowledge of the biomechanical properties and structure of human arteries. . Manufacturing of novel compliant hybrid vascular grafts with biomechanical properties matching the host arteries. . Establishing the new principles of the manufacturing of the hybrid material composed of two different types of threads for creation of the reinforced composite structure applicable to different engineering purposes. As expected, these structures will provide unique properties and will be characterised by improved reliability and durability. Academic deliverables . The part of the study Masters program in textile technology has been worked out. . The methods and results of the research have been used in Doctoral study programs. Industrial deliverables The new textile technology for manufacturing novel hybrid compliant vascular grafts. Publications Chnourko, M. and Kancevich, V. (1998), ‘‘Woven textile biomaterials international conference’’, Textiles Engineered for Performance, UMIST, Manchester, 8-11 July. Kancevicha, V. and Kasyanov, V. (1994)., ‘‘Small diameter blood vessel prostheses’’, Fibres and Textiles in Eastern Europe, Vol. 2 No. 3, pp. 32-3. Kancevicha, V. and Kasyanov, V. (1996), ‘‘Crimp vascular graft’’, Latvian Patent, Nr. 10836. Kancevicha, V. and Kasyanov, V. (1998), ‘‘Crimp vascular graft’’, Latvian Patent, Nr. 12175.
Research register
97
IJCST 15,6
98
Kasyanov, V., Purinya, B. and Kancevich, V. (1994), ‘‘Compliance of human blood vessels and novel textile vascular grafts’’, Abstracts of Second World Congress of Biomechanics, Amsterdam, The Netherlands, 10-15 July, Vol. 1, p. 28. Kasyanov, V., Kancevich, V., Purinya, B. and Ozolanta, I. (1996), ‘‘Design of biomechanically compliant vascular grafts’’, 10th Conference of the European Society of Biomechanics, Leuven, 28-31 August, p. 26. Kasyanov, V., Kancevich, V., Purinya, B., Izolanta, I. and Ozols, A. (1998), 1st Conference of the European Society of Biomechanics, Toulouse, 8-11 July.
Riga, Latvia Institute of Textile and Clothing Technology, Riga Technical University, Latvia. Associate Professor, Dr sc. ing. Ivars Krievins Research staff: Msc, Dipl. ing. V. Sorokins, Msc, Dipl. ing. S. Valaine, postgraduate (Master and doctoral study programs) students
Latvian clothing market product oriented research Other partners: Academic Terminology Committee of Academy of Sciences
Industrial Ministry of Light Industry of Latvia (1988-89) Ministry of Economics of Latvia (1997–98) Lauma Co., Liepaja, Latvia (1996–97) Project started: 1988 Project ends: No limit Finance/support: For particular objectives Source of support: N/A Keywords: Clothing, Fashion General long-term Latvian clothing market studies embrace common information areas met through market secondary (desk) research. Mainly it is based on the analysis of the Latvian 8000 household budget survey data and others available, e.g. pricing statistical data. The results of the Latvian clothing market general monitoring are oriented towards use in academic and industrial fields. In addition, the results are used for planning of primary clothing market research within a narrower in-depth range of clothing products. The morphological structure of ladies’ underwear demand has been determined
by the studies of catalogues, that of corresponding retail outlets and by administering questionnaires to 300 respondents on their actual and planned underwear wardrobe in 1997. In order to carry out the inquiry, Latvian/Russian terminology has been developed for naming illustrated product characteristics of the questionnaires. An elaborated thesaurus of clothing production terms is included in wider consumer education programmes as well as in the academic and professional ones. Conceptual analysis can be used for coordination and subordination of the educational contents within different levels and sectors of clothing education. Project aims and objectives . General monitoring of Latvian clothing market size and segmentation trends. . Distribution of consumer preferences by morphological attributes within in-depth analysis of clothing products. . Simulation of the garment quality evaluation based on consumer perception/satisfaction analysis. Academic deliverables . Morphological simulation of clothing product differentiation. . Comprehensive clothing quality evaluation methodology. . Mathematical simulation of clothing sizing. . Dictionary of Latvian/Russian clothing terms as the basis for product information processing. Industrial deliverables . General information on Latvian clothing market. . Particular information on the women’s underwear style preferences in Riga, 1997. . Feasibility of textiles and clothing standardization items in Latvia. Publications Blinkens, P., Be¯zina, V. and Krievins˘, I. (1989), Tekstilr & u¯pniecibas terminu v & a¯rdnica, Zina¯tne Riga, 855, 1p. (Dictionary of 14,000 textile terms). Der lettische Textil- und Bekleidungsmarkt ¼ LR tekstiliju un apgerbu tirgus – Riga (1997), 57 S. (German/Latvian: Latvian textiles and clothing market, Desk research). Krievins, I. (1996), ‘‘Systematization of clothing technology concepts for Latvian terminology’’, (Lengvsios pramones tehnologios ir dizainas), Kauno, pp. 221-8. ‘‘Rigas sievies˘u apaks˘g`e rbu pieprasi-juma struktu¯ra’’ (1997), gada¯, ZPD pa¯rskats; Riga, 97,1p (Structure of ladies’ underwear demand in Riga).
Research register
99
IJCST 15,6
100
Riga, Latvia Institute of Textile and Clothing Technology, Riga Technical University, Latvia Prof. Dr habil sc. ing. Silvia Kukle Research staff: Assoc. Prof. Dr sc. ing. A. Vilumsone, Lecturer I. Vilumsone, Postgraduate (Master and doctoral study programs) students
The investigation of the geometry and composition of Latvian folk art designs Other partners: Academic Latvian Council of Science 1989-1996 Project started: 1989 Finance/support: N/A Source of support: N/A Keyword: Textiles
Industrial Latvian Crafts Chamber, 1993-present Project ends: No limit
Folk art is the form of ethnic consciousness to consolidate not only people of one nation but also of many generations. It is a means for manifesting and forming the specific face of the country and investment in the worldwide cultural heritage. We started our work on computerized collections in 1989, involving scholars of our university and many generations of students. As a result, material from museum and private collections and published works were collected together and systematized and presentations were prepared. The data address different users and applications, such as teaching materials to support school and university courses such as home economics, crafts technologies, ethnography ornamentation and composition; teaching materials for craftsmen for use in design studies libraries to support reproduction for local users – householders, artists and tourist markets; as a source of ideas – motifs and symbols, technologies, ways of material combination with different properties, fashions, placement of ornamentation, compositional solutions, creation of motifs, methods of designing double-face ornamented fabrics, pattern designing; methods of forming color ranges; leveraged space filling; fantasy for imagination. The other application is a well organized source of multipurpose scholastic studies, for example, to sort and classify, to study decoration methods and/or technologies, to find out rules; ethnographic studies, historical studies; regional studies; ethnoastronomy studies, linguistic studies.
Project aims and objectives Aim: Creation of the computerized knowledge basis of Latvian folk designs, crafts, technologies and tools. Objectives: Creation of the image and text libraries for different groups of folk textiles (mittens, table cloths, towels, girls’, women’s and men’s folk costumes, blankets), woodwork tools; investigation of basic rules followed in forming motifs, symbols, compositional groups and composition; investigation of the colour, symbol preferences in different regions and products; comparative analysis of ornamentation traditions in Latvian and Lithuanian folk art; analysis of the information structure and creation of the codification system; calculation of data leading system. Academic deliverables New knowledge supplementing basics of Latvian folk art; creation of new methods of investigation; creation of databases for further investigations; investigation of the use of prehistorical symbols and signs in Latvian border patterns, comparison with other ancient cultures; creation of the system of hypothesis; highly systematized teaching materials supporting different study courses. Industrial deliverables Methods of motifs’ creation, organization of rhythms, color ranges, border and panel type compositions; methods of creation of two face fabric designs; library of designs for reproduction (crafts companies, crafts people, students) and as an inspiration for new designs (artists, craftsmen, designers, students), knowledge of folk art basics (artists, craftsmen, designers, technologists). Publications Kukle, S. (1993), Geometry of Latvian Designs, thesis of Dr habil ing dissertation, Latvia. Kukle, S. (1995), Geometry of the Crosses and Diamond Type Signs Preferably Used in Latvian Folk Designs. The Investigation and Optimization of the Textile Technology, Riga, Latvia, pp. 67-78. Kukle, S. (1996), ‘‘Computer graphics as a tool giving unambiguous results’’, poster abstracts, 2nd World Congress on the Preservation and Conservation of Natural History Collections, University of Cambridge, 20-24 August. Kukle, S. (1997), ‘‘Latvian border patterns, Lengvosios pramones technology’’, Kauno tecgnologijos universitetas, Lithuania. Kukle, S., Vilumsone, A., Vilumsone, I. and Kikule, D. (1996), ‘‘Database of Latvian folk designs’’, poster abstracts, 2nd World Congress on the Preservation and Conservation of Natural History Collections, University of Cambridge, 20-24 August. Vilumsone, I., Kukle, S. and Zingite, I. (1995), ‘‘The ornaments and rhythms of sashes of Alsunga (town on Western Baltic seaside)’’, The Investigation and Optimization of the Textile Technology, Riga, Latvia, pp. 54-60.
Research register
101
IJCST 15,6
102
Riga, Latvia Institute of Textile and Clothing Technology, Riga Technical University, Latvia Tel: 00371 708 9333 Assistant Professor Ilze Baltina, Department of Mechanical Technology of Fibre Materials Research staff: Assistant Professor I. Brakch, Associate Professor H. Vinovskis and postgraduate students
Wool carbonizing in the radio frequency electromagnetic field Other partners: Academic None
Industrial Textile factory at Kustanai (Kazakhstan) 1992-present Textile factory at Cernigov (Ukraine) 1993-1995 Textile factory ‘‘Riga tekstils’’ (Latvia) 1991-1994 Project started: 1989 Project ends: No limit Keywords: Electromagnetics, Radio, Wool In wool carbonizing the baking process is usually carried out at very high temperatures, 120-125 C, but sometimes also at 130 C. At such temperatures wool decomposes and turns yellow. It is advanced by a high concentration of sulphuric acid solution. In the new method, instead of baking with hot air, there is inclusion of radio frequency electromagnetic field, which creates vegetable matter energy that is extracted as heat. Wool temperature does not exceed 100 C and end moisture is 10-15 per cent, but vegetable matter temperature is sufficient for hydrolysis. Sulphuric acid concentration in this case does not exceed 35–40 per cent. Wool fibre rapid hydrolysis and dissolution occurred when the acid concentration ranged from 40 per cent up. Project aims and objectives To work out new carbonizing technology in which vegetable matter can be removed maximally, but wool fibre damage is very low. Academic deliverables Practice and new knowledge for Master’s and postgraduate students.
Industrial deliverables New carbonizing technology which prevents wool damage during carbonizing.
Research register
Publications Baltina, I. and Brakch, I. (1997), ‘‘Wool carbonizing in a radio frequency electromagnetic field’’, World Textile Congress on Natural and Natural-Polymer Fibres, University of Huddersfield. Baltina, I. and Reihmane, S. (1998), ‘‘Use of cellulose production waste product lignosulphonate in carbonisation of wool’’, 7th International Baltic Conference on Materials Engineering, Jurmala, Latvia, pp. 161-5. Zarina, I. (Baltina, I.), Reihmane, S., Braksch, I. and Liepa, I. (1995), ‘‘Wool carbonizing methods’’, Progress in New Polymer Materials: Seminar Materials of TEMPUS Programme, Riga.
103
Selkirkshire, Scotland Heriot-Watt University, Netherdale, Galashiels, Selkirkshire, Scotland Tel: 01896 892136; Fax: 01896 758965; E-mail:
[email protected] School of Textiles George Stylios, Bert Mather, Bob Christie, Dean Robson, The School of Textiles, Heriot-Watt University
Engineering the performance and functional properties of technical textiles Other partners: Academic Industrial UMIST British Textile Technology Group University of Leeds Industrial Member companies Project started: 1 December 2002 Project ends: 31 November 2005 Finance/support: £1,000,000 Source of support: Department of Trade & Industry Engineering and Physical Science Research Council Keywords: Industrial textiles, Non-woven, Biomedical, Fibres, Yarns, Fabrics, Garments Technical textiles are defined as textile materials and products manufactured for their technical performance and functional properties rather than their aesthetic and decorative characteristics. Despite market predictions for technical textiles, and incremental advances in some companies, many fundamental problems relating to the engineering and development of technical textiles remain to be solved; these become more urgent due to fierce global competition. There are many areas that still employ subjectivity and tradition
IJCST 15,6
104
that hinder the leap forward the sector needs to enable the development of new products and applications of textiles as engineered materials. Structural mechanics of textiles has been researched extensively, focusing on the understanding of simple textile structures mainly for apparel applications. Whilst such research may have solved specific problems, there are serious limitations to the production of generic solutions for precision engineering and manufacture of technical textiles, due to the inherent complexity of technical textile materials and their structures. Limited research has been carried out on the engineering of other properties such as thermal and fluid, which are equally important for the engineering of technical textiles. Consultation within the academic community, and with industrial members of the TechniTex Partnership over the past 12 months has established three key themes of research required to enable the proposed underpinning platform of knowledge to be established. These mutually dependent themes are modelling (to enable 3D design, simulation, and visualisation), measurement (to define and understand the relationships between structure, performance, and functionality), and manufacture (to enable appropriate manufacturing conditions to create the engineered textile). Within each a fundamental understanding of materials is required. The three themes are reflected in the technical textile challenges. To ensure the industrial relevance and applicability of the theme-based research proposed, it is not possible to explore each theme in isolation. As shown above the level of mutual dependency requires that an integrated programme of research be conducted. Project aims and objectives The rapidly growing technical textiles industry draws ideas and expertise from a diverse range of academic groups and disciplines. The aim of the TechniTex core research programme is to formalise and extend this distributed generic body of knowledge relating to textiles. This extended and integrated body of knowledge will generate a platform for the creation of methodologies for specification, design and manufacture of technical textiles, and relate performance and functionality with manufacturing processes. The specific objectives are to: . establish databases of existing technical textiles, associated technical data, and the associated body of knowledge; . classify existing technical textile structures on the basis of their function, properties, and end use; . classify processing conditions for fibres, yarns, and fabrics; . identify missing data in terms of structural and mechanical detail, and fibre and yarn properties; . research and develop new and enhanced test methods and equipment for technical textiles;
.
.
.
.
. .
.
generate new data to further establish the body of knowledge on existing yarns, fibres and fabrics; establish the performance criteria necessary for technical textiles appropriate to their end use; create geometric and mechanical models of technical textile structures using the derived classification and data; create predictive models for these processes specific to the demands of technical textiles, and to optimise their manufacturing conditions to fulfil the specified performance criteria; conduct an experimental programme for the verification of the models; create interfaces between the models and generate an integrated suite for the engineering of technical textiles; conduct a programme of dissemination and technology transfer through established TechniTex Faraday practices.
Research deliverables (academic and industrial) . an integrated suite of databases encapsulating the body of knowledge on technical textiles; . geometric models for visualisation and input into performance modelling; . mechanical models for performance and manufacturability; . new and enhanced test methods and equipment; . new standards for the specification and manufacture of technical textiles; . methodologies for the creation of engineered technical textile structures; . methods for optimising manufacturing conditions, processes and materials geared to the specific needs of these engineered structures; . a technology transfer pathway through to the wider industrial and academic networks of the TechniTex Partnership. Publications Russell, S.J. and Mao, N. (2000), ‘‘Directional permeability in homogeneous nonwoven structures. Part 2: permeability in idealised structures’’, J. Text. Inst., Vol. 91, pp. 344-58. Bandara, P. and Islam, S. (1991), ‘‘Yarn spacing measurement in woven fabric with special reference to start-up marks’’, J. Text. Inst., Vol. 87, Part I, pp. 107-19. Finn, J.T., Sagar, A. and Mukhopadhyay, S.K. (2000), ‘‘Effect of imposing a temperature gradient on moisture vapour transfer through water resistant breathable fabrics’’, Tex. Res. J., Vol. 70 No. 5, pp. 460-6. Partridge, J.F., Mukhopadhyay, S.K. and Barnes, J. (1998), ‘‘Dynamic air permeability behavior of Nylon66 air bag fabrics’’, Text. Res. J., Vol. 68 No. 10, pp. 726-31. Potluri, V.V.P., Atkinson, J. and Porat, I. (1992), ‘‘Performance assessment of a robot for use in a fabric test cell’’, 29th International Matador Conference, Manchester, April.
Research register
105
IJCST 15,6
106
Selkirkshire, Scotland Heriot-Watt University, Netherdale, Galashiels, Selkirkshire, Scotland Tel: 01896 892136; Fax: 01896 758965; E-mail:
[email protected], Professor G.K. Stylios, School of Textiles Research staff: Ms Fan Han
HOMETEX: a virtual trading centre for textiles Other partners: Academic None
Industrial OCF Ltd Silicon Graphics Inc. Scottish Enterprise Borders Scottish Textiles Manufacturers Association Borders Textile Forum Project ends: 31 August 2004
Project started: 1 September 2001 Finance/support: £1,000,000 Source of support: EU ERDF Objective 2 Keywords: Drape, Augmented reality, Virtual trading, Home shopping, 3D simulation, Dynamic draping In recent years we have witnessed a revolution in networking of information on a global scale via the Internet. Many companies have capitalised on this provision and have used it in many diverse ways, from electronic mailing to marketing, selling and trading of products and services. Marketing and selling of limp products such as textiles and, particularly, garments using new multimedia techniques would be extremely beneficial to the industry, since it would enable companies to reduce product to market, to enhance product development through 3D visualisation and to trade directly without the intervention of retailers. But selling of garments is not as easy as selling other commodities; garments are made of limp materials which take up the configuration of the wearer; customers would in most cases like to wear the garment, or, in the case of buyers, see the garment worn by a model. For the effective exploitation of these possibilities, we should, therefore, develop a multimedia environment to enable the simulation of drape behavior of garment designs on virtual models who may resemble real customers.
The textile industry chain, being a traditional industry, is very conservative in the use of multimedia for manufacture, advertising and/or sales. The reason is firstly because the industry consists of small companies which do not have the resources to use multimedia technologies effectively without training, and secondly there is no technological infrastructure available to realistically visualise new products from home. This project aims to pilot such possibilities which have other benefits to this industry in terms of ‘‘just-in-time’’ manufacture, 3D visualisation of new products and better communication with their customers. Project aims and objectives The main objectives of this scheme are as follows: . To develop a virtual Home Trading Centre for the textile, clothing and retailing industries: ‘‘HomeTex’’. . To enable the production of virtual fashion shows for buyers through CD-ROM and Internet presentations. . To network 40 companies with 500 homes (directly) via the technology and to regularly upgrade and manage company trade data, for piloting the technology. . To establish and provide through ‘‘HomeTex’’ other trade data, real-time electronic mail, Tele Trading and, possibly, banking. . To enable textile and clothing companies to interface with this technology so that new products can be made much faster and to minimise energy, raw materials and other resources. . To network with other services, such as Cyber Tex and SPIN. Research deliverables (academic and industrial) A Virtual Trading Centre in Textiles operating from the Borders of Scotland. Publications Stylios, G.K. and Wan, T.R. (1998), ‘‘A new collision detection algorithm for garment animation’’, International Journal of Clothing Science and Technology, Vol. 10 No. 1 pp. 38-49. Stylios, G.K. and Zhu, R. (1998), ‘‘The characterization of static and dynamic drape of fabrics’’, Journal of the Textile Institute, Vol. 88 No. 4, pp. 465-75. Stylios, G.K., Wan, T.R. and Powell, N.J. (1996), ‘‘Modeling the dynamic drape of garments in synthetic humans in a virtual fashion show’’, The International Journal of Clothing Science and Technology, Vol. 8 No. 3, pp. 44-55. Stylios, G.K., Wan, T.R. and Powell, N.J. (1995), ‘‘Modeling the dynamic drape of fabrics on synthetic humans: a physical lumped parameter model’’, International Journal of Clothing Science and Technology, Vol. 7 No. 5, pp. 10-25.
Research register
107
IJCST 15,6
108
Tampere, Findland Tampere University of Technology, Institute of Fibre Materials Science, P.O. Box 589, FIN-33101 Tampere, Finland Tel: +358-3-311511; Fax: +358-3-31152955; E-mail:
[email protected] Dr Ali Harlin Research Staff: Pirjo Heikkila¨, Esa Leppa¨nen
Self-organized functional materials Other partners: Academic Industrial Helsinki University of Technology None University of Helsinki University of Art and Design Helsinki Project started: None Project ended: 31 December 2003 Source of support: Tekes, the National Technology Agency of Finland The aim of the project is to combine the expertise of block copolymer RAFTsynthesis expertise, nanostructure expertise, textile fibre expertise and processing and filter expertise to assess novel types of nanostructured fibres. The hypothesis is that such fibres could be added within the supporting fibre mat to: (1) add surface area for absorptive filters; (2) act as a delivery media in specific applications; (3) act as catalyst support; and (4) act as templates for sensors functions. In all these cases, response to environmental changes is an additional very attractive goal. The goal is to achieve enhanced surface area for nanofilters. Project aims and objectives The technical target for the project is to demonstrate nanostructured fibre filter and special web containing self-organized polymer structures. Technically feasible products and processes are aimed. The products should be environmentally friendly and incinerable. The products should lead to acceptable mechanical, chemical, and heat resistance. Publications None
Tampere, Finland Tampere University of Technology, Institute of Fibre Materials Science, P.O. Box 589, FIN-33101 Tampere Tel: +358 - 3 - 3115 11; Fax: +358 - 3 - 3115 2955 docent Eija Nieminen Research staff: Pa¨ivi Talvenmaa, Auli Sipila¨
Kitex Other partners: Academic None
Industrial SOK, Kesko, Virke FINATEX & Tekstiili- ja jalkinetoimittajat ry Project started: 1 May 2003 Project ended: 31 December 2003 Source of support: Tekes (National Technology Agency of Finland) and the partners Keywords: Textiles, Recycling, Reducing the waste amount The aim of the waste policy in European Union and Finland is to reduce the waste amount and increase the recycling of all materials. In the first part of this project, the information of the finnish and international textile recycling systems is collected with the information of volumes, used technology and business operation models. The information is also collected concerning the preconditions and goals of the producers, importers and trade in the area of developing a recycling system for the textile products. Project is realized in the co-operation with partners (producer, importer and trade). Project aims and objectives The aim of this project is to prevent the waste attached to the consumers’ textile products and rise to challenge the future by developing the ways of actions to the situations when textile waste is not allowed to take to the dumping place. Publications None
Tampere, Finland Tampere University of Technology, Institute of Fibre Materials Science, P.O. Box 589, FIN-33101 Tampere, Finland Tel: +358-3-311511; Fax: +358-3-31152955; E-mail:
[email protected]
Research register
109
IJCST 15,6
Dr Heikki Mattila Research Staff: Auli Sipila¨, Nina Ojala
Smart store 110
Other partners: Academic Designium / Helsinki University of Art and Designs (leader) Institute of Software Systems Tampere University of Technology
Industrial Lectra Finland Major Blue Ltd Turo Tailor Luhta/L-Fashion Group Sokos/SOK Project started: 1 August 2002 Project ends: 31 January 2004 Source of support: Tekes, National Technology Agency of Finland Industrial partners Keywords: Concept, Customer study, Mass customization, Software integration The goal of the project The goal of the Smart Store project is to create a concept of a new intelligent garment store that takes advantage of the new available technologies. In the first phase of the project, till the end of year 2003, focus is in studying the possibilities and background technologies to achieve this goal. The technologies available open new possibilities to the customer by providing support in choosing, ordering, and paying the garment – a new kind of shopping experience, in general. The technologies examined include 3D measurement and visualization of the human body, virtual 3D try-on of the garment, automated decision of the appropriate and fitting size of the garment, and – on the other hand – enhancing the whole logistic chain. Actions (1) University of Art and Design Helsinki (Uiah) Industrial Design Department . Use of consumer decision-making models from economics; . Information gathering from defined target groups with probes method; . Observation of target groups; . Evaluation of similar concepts; . User centric design process from the gathered information; . Platform-based concept; . Usability testing of the concept’s interactive prototype; . Scenarios of use from the main features of the concept;
(2) Tampere University of Technology, Fibre Materials Science . Offer information from the apparel industry and trade to the other academic partners; . Gather information from other concepts and their working methods (benchmarking); . Apply the technological know-how when comparing the options e.g. different body scanners or mass customization methods; . Find out the way the garment producers (who are partners in this project) work and define how the new concept will fit in their working methods and what kind of changes are needed; . Find the solutions for the logistic chain behind the concept; . Take part in concept building. (3) Tampere University of Technology, Institute of Software Systems . Designing an architecture for software integration; . Studying the systems of the industrial partners for the above, including current data flow, data formats and protocols employed; . Studying the promise and applicability of 3D visualization and virtual try-on systems. Project aims and objectives (1) Boost the efficiency of supply chains and networks of the textile and clothing industry and trade by creating a new e-business concept. (2) ‘‘Semi-virtual’’ buying experience: body scanning – virtual fitting. (3) Mass customisation: best fit / made to measure = instant delivery / order delivery. (4) More efficient logistics – smaller forecast errors – faster response to demand. (5) New kind of buying experience – boosted sales – better visibility for alternative articles – repeat orders possible, etc. Publications None
Tampere, Finland Tampere University of Technology, Institute of Fibre Materials Science, P.O.Box 589, FIN-33101 Tampere, Finland Tel: +358-3-311511; Fax: +358-3-31152955; E-mail:
[email protected] Arja Puolakka Research staff: Pertti Nousiainen, Arja Puolakka, N.N.
Research register
111
IJCST 15,6
Biotechnical quality improvement of synthetic textile fibres
112
Other partners: Academic TU-Graz, Austria VTT Biotechnology, Finland University of Minho, Portugal
Project started: 1 October 2001 Finance/support: 1739220 e Source of support: EC
Industrial Sattler AG, Austria Textil Alberto de Sousa SA, Portugal Rhodia Industrial Yarns AG, Switzerland FISIBE, Portugal Inotex, Czech AB Enzymes Project ends: 30 September 2005
Publications None
Terrassa, Spain INTEXTER (Instituto de Investigacion Textil), UPC, Colon,15, 08222-Terrassa, Spain Tel: 34-937398270; Fax: 34-937398272; E-mail:
[email protected] Laboratory of Physical-Chemistry of Dyeing and Finishing Riva, Ascensio´n Research staff: Riva, Ascensio´n, Algaba, Ine´s, Prieto, Reme
Ultraviolet radiation protection proportioned by textiles: study of the influence of the most significant variables and application of specific products for its improvement Other partners: Academic Industrial Yes None Project started: 1 January 2000 Project ended: 31 December 2002 Finance/support: 88.854 e Source of support: Spanish Ministry of Science and Technology
Keywords: Ultraviolet radiation, Diffuse spectral transmittance, Solar protection factor (SPF, UPF), Textiles, UV-absorbers, Dyestuffs, Optical brightening agents, Finishing products, Colorimetry, Spectrophotometry, Comfort parameters, Ecological parameters The main objective of the project is to study the protection factor to the UV radiation achieved by the fabrics in function of the most significant variables that have influence on it, as they are: type of fiber, structural parameters of the fabric, color, UV-absorbent finishing products, and optical brightening agents. For this aim, fabrics have been elaborated with different fiber composition (including new fibers that incorporate absorbent products of the radiation) and structures. Specific finishing and optical brightening products have been applied. The diffuse spectral transmittance in the UV wavelength of the different fabrics has been determined by means of the technique that has been put onto point in this project. Other complementary objective of the project is the influence that finishing specific products, applied to improve the UV protection factor, can have on some characteristics of quality and comfort of the fabrics, as well as their permanency after repeated laundries and the ecological incidence of the application baths. The results of the project can directly benefit our textile industry contributing to their know-how and facilitating their specialization in high value added products. They can also be of great utility as contribution to the public health. Project aims and objectives To establish the influence of the main parameters of fabrics on the ultraviolet protection factor (UPF) values and propose the mathematical models that permit to predict the behavior of the new fabrics with predetermined characteristics. To give support to the textile industry in the knowledge and possibilities to produce new protective articles of great value. To conscious industrialists and people of the importance of the textile UV protection. Academic deliverables Scientific publications Industrial deliverables Application of deduced mathematical equations as models to predict the influence of several fabric parameters. To impart courses and training. Publications Algaba, I. and Riva, A. (2002), ‘‘In vitro measurement of the ultraviolet protection factor of apparel textiles’’ Coloration Technology, Vol. 118, pp. 52-8. Algaba, I., Riva, A. and Crews, P. (2002), ‘‘Influence of fiber type and fabric porosity on the ultraviolet protection factor provided by summer fabrics’’, Proceedings of the 2002 International Conference and Exhibition of the American Association of Textile Chemists and Colorists, October 2002, Charlotte, North Carolina (USA), AATCC Review (in press).
Research register
113
IJCST 15,6
114
Gueguen, V., Riva, A. and Prieto, R. (2001), ‘‘Effect of UV absorber on photoyellowing of wool’’, Proceedings of the 2001 International Textile Congress, June 2001, Terrassa, Spain. Riva, A. (2000), ‘‘¢Que´ es el UPF de un tejido?’’, Revista de Quı´mica Textil, Vol. 144, pp. 72-8. Riva, A., Algaba, I. and Prieto, R. (2002a), ‘‘Action of UV protective auxiliary products applied in the wool dyeing with metal complex dyes’’, Proceedings of the 71 IWTO Congress, May 2002, Barcelona, Spain. Riva, A., Algaba, I. and Prieto, R. (2002b), ‘‘Influence of color on the UV protection supplied by fabrics of cellulose fibers’’, Proceedings of the VI National Congress of Color, September 2002, Sevilla, Spain.
Villeneuve d’Ascq., France Technical Sciences University of Lille (USTL), UFR Mathe´matiques pures et applique´es, 59655 Villeneuve d’Ascq. France Tel: 33 3 20 43 67 83; Fax: 33 3 20 43 67 74; E-mail:
[email protected] Pure and Applied Mathematics Department USTL Research staff: Dr Jean-Jacques DENIMAL, Dr Franc¸ois BOUSSU
Development of a decision aid software for textile retailer firms Other partners: International Data Processing and Engineering firm Academic Industrial None International Textile Retailer Project started: 1 March 2003 Project ends: 1 January 2005 Source of support: French National Agency for Innovation (ANVAR Lille) Keywords: Decision aid software, Data mining tools, Data visualization tools The research program tends to achieve a decision aid software mainly oriented in the understanding of past sales behaviors and forecasting of the new textile items collection. Different tools will be provided to each of the supply chain actors (marketing executive, textile collection responsible, logistic manager and chief executive). All of these tools are designed to be as concise as possible and precise to give an efficient aid for the decision process. At the end, the prototype software will be adapted to any data processing environment, any users categories and levels.
Project aims and objectives The project aims at providing different decision aid tools to each of the supply chain actors involved inside a textile retailer firm. Different organizational scheme of stock delivery and management are considered in the methodology. So, different models can be adapted to these different textile items managements. The challenge is to propose tools based on complex mathematical and statistical methods. However, these tools will have to remain accessible for unskilled statistical persons. Research deliverables (academic and industrial) Two main functions are to be present in the prototype software.The first one deals with the understanding of the nearest past sales seasons. Data mining tools helps to highlight the significant parameters influencing the sales behaviors. The second one treats the two stages of the forecasting process. Before the season, different decision tools allow to provide estimated future sales. During the season, an efficient and fast algorithm provides forecasted sales values including the recent changes. Publications Boussu, F. and Denimal, J.J. (2002a), ‘‘Statistical modeling and data mining to identify the consumer preferences. Application to a Textile sales data set’’, in Warren, C. (Ed.), Neel Conference on the New Frontiers of Statistical Data Mining and Knowledge Discovery, 22-25 June 2002, Marriot Knoxville, Tennessee, USA, p. 30. Boussu, F. and Denimal, J.J. (2002b), ‘‘The data-mining : a competitive advantage to the textile firms’’, World Textile Conference – 2nd AUTEX Conference, Textile Engineering at the Dawn of a New Millennium: An Exciting Challenge, Poster session, 1-3 July 2002, Bruges, Belgium, p. 589. Boussu, F. and Denimal, J.J. (2003), ‘‘Data visualization tools to highlight sales and products relationships’’, Application of a Data Mining Method to Textile Sales, Las Vegas, Nevada, USA, CSREA Press, 23-26 June, 2003, Vol. II, pp. 475-81. Boussu, F., Denimal, J.J. and Ousmane Wora DIALLO (2002), ‘‘Reduction of the estimation uncertainty of the textile sales profiles by using a data-mining technique’’, Paper 54, Special Session on Information Processing and Management of Uncertainty in Textile/Garment Industry, IPMU 2002, 1-5 July 2002, Annecy, Vol. 3, pp. 1895-1900.
Zagreb, Croatia Faculty of Textile Technology, University of Zagreb, Pierottijeva 6, 10000 Zagreb, Croatia Prof. Dr Ruz˘ica C˘unko, Department of Textile Chemistry and Testing of Materials Research staff: Dr Maja Andrassy, Dr Emira Pezelj, Dr Mirjana Gambiroz˘a-Jukic´, Vera Fris˘c˘ic´, Biserka Vuljanic´, Antoneta Tomljenovic´, Marija Kovac˘evic´
Research register
115
IJCST 15,6
116
Ecological aspects of fiber properties and quality of textiles Other partners: Academic Industrial None None Project started: 1 June 1997 Project ends: – Finance/support: N/A Source of support: Ministry of Science and Technology of the Republic of Croatia Keywords: Ecology, Textiles Research activities proposed are related to the environmental aspects of textile materials and will include investigations of the impact of some environmental parameters on fiber properties, as well as the investigations of a possibility of evaluating textile products on the basis of their ecological safety. Textile products interact with their environment, which influences the changes occurring in them, covered by the term ‘‘ageing’’. The changes are complex and varied and in most cases quite specific and sophisticated investigations are necessary to understand them. The investigation of fibre ageing under the influence of UV-radiation, ozone and pollutants, is supposed to contribute to understanding the mechanism of the changes on molecular, structural and morphological levels. The other research task deals with the problem of a possible harmful influence of textiles on human health, covered by the expression of ‘‘human-environmental safety’’. This kind of safety has become one of the basic prerequisites, when speaking about the quality of textile products, and the most important one when trying to sell on the European market. The investigations are supposed to create basic prerequisites for laboratory evaluation of humane-environmental safety. Part of the research task concerns modification of fibre properties using ultrasound waves, as an environmentally very acceptable solution. The effect of ultrasound waves will be investigated on cellulose fibres, polypropylene, polyamide, polyester and wool. The results will be published in scientific research periodicals and conferences, making possible their usage, checking and evaluation. Project aims and objectives In the area of polypropylene and aramide fibres ageing the aim is to complete the investigations started by the previous project, especially concerning the impact of atmospheric pollutants, sunlight, weather conditions and ozone concentration in various stress conditions. The second aim is to investigate possible modifications of fibre properties through the application of ultrasound, and the third aim is to create a scientific and expert basis for testing and objective evaluation of human-environmental safety as a basic prerequisite for textile quality assurance.
Publications C˘unko, R. and Pezelj, E. (1997), ‘‘The ageing of polypropylene through environmental agency’’, The 78th World Conference of Textile Institute, Thessaloniki. C˘unko, R., Andrassy, M. and Pezelj, E. (1998), ‘‘Elimination of polyester fibre oligomers using ultrasound waves’’, Proceedings TEXSc ’98, Liberec. C˘unko, R., Pezelj, E. and Andrassy, M. (1997), Tekstil, Vol. 46, pp. 677-83. Pezelj, E., C˘unko, R. and Andrassy, M. (1997a), ‘‘The effect of global climatic change on the fibre ageing’’, Proceedings Slovenia Chemical Days. Pezelj, E., C˘unko, R. and Andrassy, M. (1997b), ‘‘The influence of repeated maintenance treatments on properties of PP fibers’’, Proceedings The 78th World Conference of Textile Institute, Thessaloniki.
Zagreb, Croatia University of Zagreb, Faculty of Textile Technology, HR-10000 Zagreb, Croatia Tel: ++385 (1) 37 03 153; Fax: ++385 (1) 37 74 029; E-mail:
[email protected] Edita Vujasinovic, Department of Textile Chemistry and Material Testing
Sorption characteristics of medullated wool fibres Other partners: Academic None
Industrial None
Project started: 1999 Project ended: 2001 Finance/support: 3,000 DEM/year Source of support: Croatian Ministry of Science and Technology Keywords: Textiles, Wool, Sorption, Medullated wool From 500 to 700 tons of greasy wool are sheared in Croatia every year (annual statistics of Croatia 1990-1995). Because of the widely heterogeneous character of the wool sheared, the number of different breeds of sheep, inadequate shear preparation and extremely high content of medullated fibres (Tekstil, Vol. 41 (1992), 591; Stocarstvo, Vol. 48 (1994), 443), coarse wool of domestic sheep is not used in the textile and garment industry. Under these conditions, the quantities of wool stated cease to be a useful raw material and become an ecological hazard. Some developed European countries are faced with the same problem (Nuova Sel Tess, Vol. 5 (1996), 28), and, as coarse wool, due to high content of medullated fibres, can be a useful absorbing and insulating material (both sound and heat insulating), investigations were started to establish the possibilities of using
Research register
117
IJCST 15,6
118
domestic wool as raw material in the manufacture of technical textiles for a wide range of applications (e.g. in civil engineering, building and construction, agriculture and some other branches of industry). Investigations of the physicochemical (and especially absorptive) properties of medullated wool fibres will show how they can be used in the manufacture of technical textiles, such as various filters, agro-, geo- and thermo-textiles. In this way, coarse domestic waste wool will be used as an environmentally and economically acceptable product, which is in tune with European and global trends of more a rational managing of natural resources, with the purpose of preserving and protecting the environment. Project aims and objectives The quality of most of the domestic wool does not meet the technological requirements stated by the Croatian wool industry, meaning that it is an industrial raw material that cannot be utilised. The aim of the investigation proposed is to explore the possibilities of using coarse domestic wool as a raw material in the manufacture of technical textiles. Fine wool fibres are appropriate, ecologically acceptable and a cheap natural raw material for the textile garment industries. Coarse wool fibres are most often a by-product of sheep breeding, which cannot be properly used. The results of investigating the physio-chemical, and especially absorptive, properties of medullated wool fibres will indicate the feasibility of using such fibres as a proper absorbing material (primarily for liquid and solid waste), in the manufacture of a wide range of technical textiles, such as filters, geo-, agro-textiles, sound and heat insulators, etc. In this way, coarse domestic wool, which has become harmful waste through burning without control and years of depositing, will be used as an environmentally acceptable and economically profitable product. Publications Raffaelli, D., Dosen-Sver, D. and Vujasinovic, E. (1999), Kemija u Industriji, Vol. 48 No. 5, pp. 189-96. Vujasinovic, E. and Andrassy, M. (2000a), Proceedings of the 4th International Conference TEXSCI 2000, Liberec, Czech Republic, 12-14 June, pp. 84-8. Vujasinovic, E. and Andrassy, M. (2000b), Tekstil, Vol. 49 No. 6, pp. 277-86.
Research index by institution Institution Aitex, Asociacion de la Investigacion de la Industria Textil, Alcoy-Alicante, Spain Bolton Institute, Bolton, UK
Index by institution
Page 5-9 9-10
Budapest University of Technology and Economics, Budapest, Hungary Centre for Fibers, Textiles and Clothing, Port Elizabeth, South Africa
28-35 89-94
Ege University, Izmir, Turkey
45-46
Ghent University, Ghent, Belgium
41-45
Heriot-Watt University, Selkirkshire, Scotland Hong Kong Polytechnic University, Kowloon, Hong Kong INTEXTER (Instituto de Investigacion Textil), Terrassa, Spain
103-108 53-57 112-114
Kaunas University of Technology, Lithuania
47-50
Kyoto Institute of Technology, Japan
60-65
Leeds University, UK
65-67
Louisiana State University, USA
75-76
Manchester Metropolitan University, USA
80-81
Manchester Metropolitan University, UK
76-77
National Institute for Textile and Leather, Bucharest, Romania
10-28
Philadelphia University, USA
85-87
Riga Technical University, Latvia
94-103
Satra Technology Centre, Kettering, UK ¨ BI¨TAK TAM The Scientific and Technical Research Council of Turkey TU ¨ BI¨TAK Tekstil Arastirma Textile Research Center, Ege U¨niversitesi TU ¨ Merkezi, Bornova, Izmir TURKEY
50-53
Tampere University of Technology, Finland
46-47 108-112
Technical University of Liberec, Czech Republic
67-70
University of Guelph, Ontario, Canada
83-85
University of Lille, France
114-115
University of London, UK
70-75
University of Maribor, Slovenia
77-79
University of Newcastle upon Tyne, UK
81-83
University of Pisa, Italy
87-89
University of Twente, The Netherlands
35-41
University of Zagreb, Croatia
115-118
Yeungnam University, Korea
57-59
119
Research index by country
IJCST 15,6
120
Country
Page
Belgium
41-45
Canada
83-85
Croatia
115-118
Czech Republic
67-70
Finland
108-112
France
114-115
Hong Kong
53-57
Hungary
28-35
Italy
87-89
Japan
60-65
Korea
57-59
Latvia
94-103
Lithuania
47-50
Romania
10-28
Scotland
103-108
Slovenia
77-79
South Africa
89-94
Spain The Netherlands Turkey
5-9, 112-114 35-41 45-46, 46-47
UK
9-10, 50-53, 65-67, 70-75, 76-77, 81-83
USA
75-76, 80-81, 85-87
Research index by subject
Index by subject
121
Subject Aeronautics Agriculture, protective clothing Apparel, design material, CAD, folk art Asbestos substitutes Automation, mechatronics, robotic ironing, stitch formation, garments Cardiovascular surgery, medical devices, biomaterials, tissue engineering, wound dressing, wearable health system Centres, nanotechnology, textile centre, virtual trading Comfort clothing Cotton fabrics, cellulose, mercerisation, bio-scouring Data mining, virtual trading Dimensional stability, anti-shrinkable yarn Environmental protection, health, pollution, ecology, reduction of waste filtering Enzyme technology European norms Fabric sourcing, clothing market Feathers and down Flame returdancy, flammability Hand evaluation, sensory measurement Heat insulation Hypothermia protection Image analysis, machine vision Individual protection equipment, diving suit, inflatable equipment, UV protection, parachute system
Page 22 83 61, 73, 100 16 65, 71, 81, 82, 94 24, 27, 29, 34, 42, 44, 88, 96 55, 86, 106 51, 69, 77 8, 9, 29, 32, 34, 37, 38, 45, 47, 93 75, 106, 114 9, 58 10, 11, 16, 109, 116 37, 39 14 75, 98 5 7, 8 49, 62, 63, 69 12 21 48, 46 12, 18, 19, 21, 112
Ink-jet printing Intelligent textiles Knitwear, knitting Laundering, washing, easy care Lycra knitted fabric, elastane Manufacturing technologies, wet textile process, coating, surface modification, dyeing and printing, wetting, UV, ultrasonic, ink jet printing, wool carbonising, bio-scouring, intencification, novel yarns Near infra red analysis, UV light, ultrasonics, radio frequency, UV radiation Nono-technology, nano-composites Objective measurement, mechanical properties, appearance, handle, biaxial, geometry Optimisation of process Paper fibres Recycled fibres Risk factor Robotic ironing Sewing machines Shingosen fabrics Simulation of process, tuft forming, CAD, 3D simulation Soil covering Spider silk, synthetic fibres Spinning, novel yarns, drawn worsted Swimwear
65 55 62, 86 9, 47, 57 18, 52 27, 35, 37, 39, 40, 41, 45, 46, 53, 59, 65, 67, 71, 80, 81, 102, 103
5, 46, 102, 112 41, 55, 89 5, 48, 49, 62, 63, 69, 78, 82, 103, 106 7, 35, 103 37 37, 109 12 71 94 63 25, 60, 61, 103, 106 14,29 42, 112, 116 25, 53, 57, 60 52
IJCST 15,6
122
Technical textiles, protective clothing, composite materials, microcellular plates, protective clothing, implantable medical device, PVC, plastic, rubber, antimicrobial, non-woven, UV radiation, paper, nanotechnology, biotechnology, tissue engineering, scaffold, wound dressing, intelligent textiles, multi functional
16, 18, 19, 21, 24, 27, 29, 34, 37, 41, 44, 48, 55, 64, 70, 73, 77, 83, 84, 86, 87, 89, 93, 96, 103, 108, 112
Testing performance Tissue engineering, scaffold, cartilage, Virtual trading, customisation, interior textiles Wearable system Weaving, worsted fabric, machines Wool, medulated, sorption Wrinkling, anti-shrinkable
18,19, 21, 22, 37 42 80, 106, 110 87 24, 25, 54, 57, 59, 61, 67, 92 117 9, 58
Research index by principal investigator
Index by principal investigator 123
Principal investigator
Page
Kiekens, P
44-45
Abou-iiana, M.
85-87
Kim, S.-J.
57-58, 58-59
Agrawal, P.B.
38-39
Labil. sc. ing. viktoria kancevicha
96-98
Alimaq, D.A.
62-63
Langenhove, L.V.
41-43
Lopez-Lorenzo, M.
37-38
Matsuo, T.
63-64
Mattila, H.
110-111
Anandjiwala, R.
89-91, 92-93
Baltina, I.
102-103
Berzina, Z.
72-75
Borsa, J. Botı´, R. C˘unko, R.
29-32, 32-33 34-35 5-7 115-117
Moholkar, V.S.
35-36
Nakamura, M.
60
Nosek, S.
67-69
Puolakka, A.
111-112
93-94
Riva
112-114
Chen, Y.
75-76
Ruckman, J.E.
76-77
Dai, J.S.
71-72
Sc. ing. Ivers krievins
98-99
De Rossi, D.
87-89
Simmons, A.
51-52
Carbonell, A.
7-9
Chapple, S.A.
docent Eija Nieminen
109
Slater, K.
83-84, 84-85
Duran, K.
46-47
Strazdiene, E.
Ene A.
23-25
Stylios, G.
103-105
Gersˇak, Sc. J
77-79
Stylios, G.K.
106-108
Gutauskas, M.
49-50
Suresh, M.N.
94-96
Tao, X.
habil sc. ing habil sc. ing. Silvia Kukle
100-101
Tarakc¸iogˇlu, I.
48-49
61-62 53-55, 55-57 45-46
Harlin, A.
108
Taylor, P.M.
81-82, 82-83
Higgins, L.
9-10
Topalovic, T.
39-40 80-81
Jahanshah, F.
65-67
Tyler, D.J.
Ku˚s, Z.
69-70
Vujasinovic, E.
Kawabe, K.
64-65
Wilford, A.
117-118 50-51, 52-53
