Modelling of Colour Yield in Polyethylene Terephthalate Dyeing with Statistical and Fuzzy Regression.

The aim of this study is to investigate and apply two approaches of fuzzy regression as well as statistical regression for modelling the colour yield in polyethylene terephthalate (polyester) high temperature (HT) dyeing as a function of disperse dyes concentration, temperature, and time. The two approaches of fuzzy regression modelling were triangular coefficients and exponential coefficients. The statistical models obtained, did not meet the required conditions to be accepted. However, after applying modification to the original fuzzy regression models with triangular coefficients, acceptable models were obtained. The modification was carried out by means of enabling the centre of coefficients to take positive and negative values as well as dispersing fuzziness in the coefficients. The models obtained with exponential fuzzy coefficients showed weak interaction for the above mentioned variables and hence it seems that for modelling cases such as the one considered in this paper, fuzzy regression with triangular coefficients may be preferred.