Online shopping has become prevalent because it is convenient and saves time. However, there are still many drawbacks to purchasing clothing online. One such problem is the discrepancy between information obtained from a photograph (visual information) versus the actual fabric hand (haptic information). One reason for this is that the haptic information may not sufficiently convey to consumers. The objective of this study was to examine the influence of fabric texture on the information transfer from visual to haptic and vice versa. A generalized linear mixed model (GLMM) was used to find the difficulties of memorizing or retrieving the fabric. The random effects of the model are the individual differences between participants or samples. The GLMM analysis showed that there was an individual difference among fabrics in subject’s performance for haptic and visuo-haptic recognition, but not for visual recognition. The subsequent analysis suggested that the cross-modal recognition failure mainly occurred in the estimation of fabric thickness. Surface texture was a very important factor for estimating the connection between visual and haptic information. When the surface texture was evaluated visually first and then haptically second, participants might evaluate the surface as rougher than the actual texture.

In the present study, a method of optimizing the design of laminated composite with new genetic algorithm techniques, selection strategy and mutation operator, is proposed. Experiments are designed to minimize thickness (or weight) of midplane-symmetric composite laminate subject to in-plane loading, with fiber orientation and ply thickness chosen as design variables. To avoid spurious laminate designs, both the Tsai-Wu and the maximum stress criteria are taken to determine whether load-bearing capacity is exceeded or not. The state-of-art experiment results are obtained for some loading cases, related numerical results are presented under different cases. The simulation results clearly show that the proposed algorithm provide better or equivalent solutions for the design of laminated composites than work in related literature.