Predicting sewing thread consumption for lockstitch using regression model

Abstract

Purpose This paper aims to develop a single regression model (instead of developing models separately for each thread type) to predict the sewing thread consumption for cotton and polyester staple spun threads. Design/methodology/approach A single regression model is developed for predicting sewing thread consumption for cotton and polyester threads. The polyester sewing threads have lower sewing thread consumption as compared to cotton threads because of their higher elongation behaviour. The model differentiates between the cotton and polyester sewing threads using their elongation values at peak levels of tensions experienced by the sewing threads during stitch tightening. By comparing the estimated thread consumption values with actual values, the effectiveness of model is evaluated with root mean square error and coefficient of determination (R2). Findings During the sewing process, by understanding the behaviour of different types of sewing threads, it is possible to develop a single regression model for all types of threads. Practical implications The sewing thread consumption can be easily calculated for cotton and polyester sewing threads using a single regression equation using the sewing assembly thickness, stitch density and elongation of thread at peak tension. The garment manufacturers need not depend on different charts for sewing thread consumption for stock management. Originality/value The sewing thread consumption is different for different types of threads, and garment manufacturers have to depend on different charts given by sewing thread manufacturers or use different equations for each type of threads. Using this single regression equation, sewing thread consumption for cotton and polyester sewing thread can be estimated accurately.