| Objective measurement techniques have been used since the 1930's to provide a more accurate method of gauging fabric quality than subjective hand evaluations. In addition to predicting subjective hand preferences, they also provide data on fabric development and ease of garment manufacture. The main techniques researched are the Kawabata Evaluation System and the Fabric Assurance by Simple Testing. The KES has been used to evaluate suiting, women's dresses, knitted fabrics and nonwoven fabrics. Research with the FAST has been limited to mainly suiting with some shirting. This has left a large area of work to be explored; whether the FAST can be used to assess fabrics for a wider range of end-uses. In this thesis, the investigation focuses on women's dress fabrics, specifically weight reduced polyester fabrics. The FAST system was modified in order to accurately differentiate between these fabrics. The Cusick drape tester was also used and further investigations focussed on other methods of testing drape, including the Aldrich method and one developed by the author. Although early indications were that this new test was repeatable, reproducible and correlated well with the Cusick method, it required more engineering work than was possible in the scope of this thesis and therefore was not used for the empirical work. The KES equipment for tensile and shear was also used and comparison made between the results of previous studies and with results found from the FAST. Statistical analysis was used throughout to establish both the effect of the weight reduction process on the fabric and its relationship to the problems of ease of manufacture, distortion of garment measurements and garment appearance. The modifications made to the FAST procedure and apparatus proved valid during the analysis. The effect of the weight reduction process was to soften the fabric, reducing the drape coefficient and bending and shear rigidity results and increasing the weft extension results. The appearance data could not be statistically analysed but it seemed that for the garment silhouette chosen, low or no weight reduction was required. Equations were established that predicted ease of manufacture and correlation factors were found between actual and predicted grades ranging from 0.71 - 0.87. Interestingly, one of the most powerful combinations was the basis data of weight and number of warp ends. High correlation between predicted and actual grades was also found for the distortion problem (0.86). |
