Analysis of the modeling methodologies for predicting the strength of air-jet spun yarns

The scope, usability, limitations and predictive power of four modeling methodologies are analyzed in this paper: mathematical models, empirical models, computer simulation models, and artificial neural network models. The predictive power of each of the four is estimated by comparing predicted yarn strength with experimentally obtained strengths for yarns spun using different process conditions and material parameters.