Applying genetic-fuzzy approach to model polyester dyeing

Abstract

Genetic-Fuzzy (GF) approach is used to model color yield in the high temperature (HT) polyester dyeing as a function of dye concentration, temperature, and time. The proposed method is compared with statistical regression model for their prediction powers using the Mean Square Errors (MSE). The results show that the GF model is preferable to the statistical regression model in at least two ways. First, it enjoys a better predictive power regarding to its minimum MSE. Second, the statistical regression model fails to satisfy the standard requirements. © 2007 Springer-Verlag Berlin Heidelberg.