A MULTILAYER NEURAL-NETWORK MODEL FOR DETECTING CHANGES IN THE PROCESS MEAN

Control charts are an important tool in statistical process control (SPC). They are useful in determining whether a process is behaving as intended or if there are some unnatural causes of variation. In this paper we present an alternative approach to SPC using artificial neural network technology and compare its performance with that of the combined Shewhart-CUSUM schemes. The neural network developed here is concerned with the detection of gradual trends and sudden shifts in the process mean. The performance of the neural network was evaluated by estimating the average run lengths (ARL's) using simulation. An extensive comparison shows that the proposed approach has 20-40% faster detection of small process changes than the combined Shewhart-CUSUM control schemes. The neural network approach presented in this paper appears to offer a competitive alternative to existing control schemes.