, Payam Noor
Control chart pattern (CCP) recognition techniques are widely used to identify the potential process problems in modern industries. Recently, artificial neural network (ANN) –based techniques are very popular to recognize CCPs. However, finding the suitable architecture of an ANN-based CCP recognizer and its training process are time consuming and tedious. In addition, because of the black box nature, the outputs of the ANN-based CCP recognizer are not interpretable. To facilitate the research gap, this paper presents a statistical decision making approach to recognize and classify the patterns of control charts. In this method, by taking new observations from the process, the Maximum Likelihood Estimators of pattern parameters are first obtained and then in an iterative approach based on the Bayesian rule, the beliefs, that each pattern exists in the control chart, are updated. Finally, when one of the updated beliefs becomes greater than a predetermined threshold, a pattern recognition signal is issued. Simulation study is performed based on moving window recognition approach, and the accuracy and speed of method is evaluated and compared with the ones from some ANN-based methods. The results show that the proposed method has more accurate interpretable results without training requirement.