摘要：When a control chart signals that an assignable cause is present, process engineers are required to identify a time point of process changes and then search for the assignable cause of the process disturbance. In the statistical process control literature, there is a research area called change point detection. As one of change point detection problems, it has been considered how to identify the time of a step-change in the process fraction nonconforming using the maximum likelihood theory. However, the process fraction nonconforming may be changed multiple times until a chart signals. Such a multiple change-points model for the process fraction nonconforming has not been considered yet. In this study, we consider a multiple change-points model of the process fraction nonconforming under apchart. Then, a method of tracking the transition of process fraction nonconforming is proposed using the maximum likelihood theory and information criterion.
关键词：KeywordsAkaike information criterion (AIC)Change point detectionControl chartDynamic programmingMaximum likelihoodpchartProportion of nonconforming itemsQuality controlstatistical inference
