In this study, a hybrid algorithm is presented to tackle multi-variables robust design problem. The proposed algorithm comprises neural networks (NNs) and co-evolution genetic algorithm (CGA) in which neural networks are as a function approximation tool used to estimate a map between process variables. Furthermore, in order to make a robust optimization of response variables, co-evolution algorithm is applied to solve constructed model of process. Results of CGA are compared with genetic algorithm (GA). This algorithm is tested in a case study of open-end spinning process.
arXiv: Computer Science::Neural and Evolutionary Computation
free text keywords: Co evolution Genetic Algorithm, Genetic algorithm, neural networks, Quality Engineering, Robust optimization, lcsh:Information resources (General), lcsh:ZA3040-5185