Improvement of process capability through neural networks and robust design: A case study

摘要

Conventional experimental design studies for process improvement are difficult in continuous production processes such as manufacture of silicon filler. For such processes, an integrated method using neural networks and robust design methods for improving process capability are presented in this article. Initally, neural network is used to provide a nonlinear relationship between process parameters and different responses, obtaining the optimum parameter settings. Then, a computerized simulation employing Taguchi's approach with orthogonal array (OA) is performed to help identify critical parameters with respect to multiple responses in order to exercise process control. The results demonstrate many fold improvement in process capabilities. (7 refs.)