Comparison of artificial neural network and linear regression models for prediction of ring spun yarn properties. II. Prediction of yarn hairiness and unevenness

Abstract

The objective of this second part of the study is to develop artificial neural network models for the prediction of yarn hairiness and unevenness and to compare the performance of ANN models with our previous statistical models based on regression analysis. Besides HVI properties, yarn count, twist and roving properties were also selected as input variables. Part 1 provided detailed description of experimental procedure of the study. Yarn hairiness and unevenness tests were performed on Uster Tester 3. Following the developed ANN models, sensitivity analysis results and coefficient of multiple determination (R-2) values of ANN and regression models were compared. Analyses are showed that ANN models improve the prediction performance with regards to regression models.