MANUFACTURING QUALITY IMPROVEMENT BY NEURAL NETWORKS UPON TAGUCHI METHOD

摘要

In developing new manufacturing and material handling process the influence of process parameters upon the target object need to be evaluated in order to find the adequate operation settings. An efficient method is the Taguchi method which reduces the number of experiments and has the capability of finding the optimal setting of process parameters. This paper proposes the construction of neural network upon Taguchi results to further advance the quality of a new deep hole drilling process. It is shown that neural network produces process parameter setting better than the "optimal setting" obtained by Taguchi method.