Modeling and Prediction on Productivity Performance of Sewing Machine Operation Using Dimensional Analysis and Artificial Neural Network

Abstract

The productivity performance of standalone sewing machine operation is analyzed in the present work by using dimensional analysis approach & artificial neural network (ANN). The models are formulated to correlate the input parameters such as Anthropometric data, personal data, environmental data at workstation, workstation specification, specifications of parts of pedal sewing machine and their condition with  the dependent parameter productivity through design of experiments (DOE) plan. From the field data based study findings, it has been observed that the anthropometric parameters of operator, his environmental conditions, sewing machine parameters are the most influencing parameters. In order to find outthe accuracy of the formulated dimensional analysis approach and ANN models, correlation coefficient (R²) was calculated. From theR² values, it was clear that both dimensional analysis and ANN approaches are competent to predict the productivity performance of the sewing machine operator. In addition, the models formulated by using ANN approach were found tobe more perfect than the dimensional analysis approach. The higher values of R² (87.89%) and lower value of variouserror based parameters shows the adequacy and reliability of the dimensional analysis and ANN models. Comparativestudy of dimensional analysis and ANN models disclosed the accuracy of ANN models hence recommended.