Prediction of Yarn Properties Using Evaluation Programing

This article proposes prediction approaches for the determination of the breaking I strength of the yarn properties by using evaluation programing. Gene expression programing (GEP) and neural networks are the evaluation programings that are used for the prediction of physical properties of yarn. In addition to these methods, multiple linear regression analysis is also used to examine the predictive power of the evaluation programings in comparison to classical statistical approach. The implementation of the genetic programing technique in GEP to the prediction of physical properties of yarn is indicated for the first time in this paper. The results obtained from the computational tests clearly show that GEP is a promising technique in terms of precision and computation time for the prediction of yarn properties (98.88%).