A general methodology for analyzing fashion oriented textile products using sensory evaluation.

Abstract In a garment company, key features on garment products can generally be characterized using three information sources: (1) physical measures using appropriate devices, (2) normalized sensory evaluation data, (3) consumers’ perception on fashion styles of garments. The understanding of the relations between criteria at these three levels is important for designing new fashion oriented textile products. In this paper, we propose a method for evaluating fabric hand and similarity between fashion styles and fabric samples. Next, the relationship between these two sets of sensory data is analyzed using the data clustering algorithm and a number of linguistic similarity and relevancy measures we defined. The effectiveness of the proposed method has been validated using a set of fabric samples and a set of T-shirt products.