A Control Chart Combining Supervised Classifier and MEWMA

In the process of monitoring, it is still a challenging task to use the features of modern industrial system for accurate and effective monitoring. This article presents the OPE-MEWMA control chart based on the multivariate exponentially weighted moving average (MEWMA) strategy combined with a supervised classifier ('one plus epsilon', OPE classifier for short). We evaluate the ability of the control chart detecting the mean shifts by considering various factors. The average run length and several other indicators are used to measure the performance of the control chart. The simulation results show that the OPE-MEWMA control chart can detect the mean shifts quickly and the model has high sensitivity to process shifts.