An intelligent method for the evaluation and prediction of fabric formability for men’s suits

Abstract

Sixty-six commonly used suitings were selected as the experimental samples of the current study. The Kawabata Evaluation System was used to measure the mechanical properties of the samples. Each sample fabric was made into a shoulder-back as a part of a men’s suit. In order to study the appropriateness of the samples for making good shaped men’s suits, which is known as fabric formability, sensory evaluation methods have been applied to obtain panelists’ assessments on the shape of the shoulder-backs. During data analysis, principal component analysis was initially adopted to reduce the complexity of the system by extracting a small number of important mechanical properties. Then, a fuzzy neural network was developed to model the underlying relations between the samples’ formability and their mechanical properties. Finally, a number of testing samples were used to verify the effectiveness of the proposed predictive model.