Construction of Predictive Model on Fabric and Sewing Thread Optimization

Abstract

Along with the trend of automation in apparel industry, some key parameters are applied to establish a prediction model for the sewing optimization between fabric and thread, which has become an important issue in apparel quality control. Hence, the structural properties of fabric and thread are used in this study to establish the optimized scope and prediction model for the quality of the sewed item. It can be noticed from the research results that the sewing quality and the structural properties of the fabric and the thread are in fact highly correlated. Therefore, neural network method can be applied to establish a neural network model between the two. By testing the neural network model established in this study, it is found that the model has high correlation coefficient and low residuals, indicating that the model has good fitting ability. Besides, the optimized scope of sewing quality can be used as a basis for evaluation of sewing optimization between fabric and thread.