Control chart pattern recognition using back propagation artificial neuralnetworks

In this paper, control chart pattern recognition using artificial neural ne tworks is presented. An important motivation of this research is the growin g interest in intelligent manufacturing systems, specifically in the area o f Statistical Process Control (SPC). Online automated process analysis is a n important area of research since it allows the interfacing of process con trol with Computer Integrated Manufacturing (CIM) techniques. Two back-prop agation artificial neural networks are used to model traditional Shewhart S PC charts and identify out-of-control situations as specified by the Wester n Electric Statistical Quality Control Handbook, including instability patt erns, trends, cycles, mixtures and systematic variation. Using back propaga tion, patterns are presented to the network, and training results in a suit able model for the process. The implication of this research is that out-of -control situations can be detected automatically and corrected within a cl osed-loop environment. This research is the first step in an automated proc ess monitoring and control system based on control chart methods. Results i ndicate that the performance of the back propagation neural networks is ver y accurate in identifying control chart patterns.