Regression model to predict thread consumption incorporating thread-tension constraint: study on lock-stitch 301 and chain-stitch 401

Abstract Prediction of sewing thread consumption requires an accurate method of calculation since it relates to the cost of manufacturing and distribution of apparel products. Previous researchers highlighted problems in existing thread consumption calculation methods; i.e. limitations in existing formulae which cause inaccurate predictions of thread amount needed for sewing operations. The existing methods of consumption calculations exhibit significant error percentages due to the ignorance of important parameters which affect on thread consumption. This paper investigates on correlation of thread tension to thread consumption of lock-stitch 301 and chain-stitch 401. The existing thread consumption formulae are optimized by considering a new parameter; thread tension, using regression analysis and geometrical modeling techniques. For the chain-stitch 401, results indicate that the thread tension significantly affects in determination of the thread consumption. The error analysis of proposed formulae was performed to indicate that the proposed formulae more accurate compared to the available methods of predicting sewing thread consumption. In addition, there are combined effects of thread tensions together with parameters such as fabric thickness and stitch density which determines accurate consumption values considering the properties of the stitch. In comparison, inclusion of the proposed thread tension variable depicts reduction in error percentages, so that the proposed formulae are expected to be a better approach to calculate thread consumption of lock-stitch 301 and chain-stitch 401.