A Combined MLE and Generalized P Chart Approach to Estimate the Change Point of a Multinomial Process

Correct and quick identification of the change point of a process shift is very important for process improvement. Although several studies have been devoted to the estimation of the change point of the process shifts for a univariate process, little research has been done on a multivariate process, particularly for a multivariate proc ess where the quality characteristics cannot be measured in a numerical scale. In this study, an effective approach that combines th e method of maximum likelihood and the generalized p control chart is developed to estimate the starting time of a process disturbance for a multinomial process. An illustrative example is provided to show how to apply the proposed approach in practice. The positive results with the use of the proposed approach are also demonstrated by a series of simulation studies. It is found that the proposed approach has better performance than the original generalized p control chart.