The prediction of cotton ring yarn properties from AFIS fibre properties by using linear regression models

Abstract

In this paper some models for predicting the most important ring yarn quality characteristics were built by using AFIS (Advanced Fibre Information System) data. Yarn count, twist and roving properties were also selected as predictors because of their great effect on the yarn properties. A total of 180 ring yarns were produced from 15 different cotton blends on the same ring spinning machine, on the same spinning positions and under the same conditions at Ege University Textile and Apparel Research-Application Centre. Each blend was spun in four yarn counts (29.53 tex, 23.63 tex, 19.69 tex, and 16.88 tex) at three different coefficients of twist (alpha(Tt) 3639, alpha(Tt)4022, and alpha(Tt) 4404). Linear multiple regression methods were used for the estimation of theyarn quality characteristics. The goodness of fit statistics showed that our equations had very large R-2 (coefficient of multiple determination) and adjusted R-2 values.