Abstract: (18785 Views)
Knowing about the real time of a change in the parameter(s) of a statistical process would enable users to identify root causes more quickly and precisely. Due to the sensitivity and importance of reaching zero defects in high quality processes, to be aware of the change time would be so precious. In this paper, we consider the performance of the Maximum Likelihood Estimator in comparison with built-in change point estimators of Cumulative Sum (CUSUM) and Exponential Weighted Moving Average (EWMA) control charts. Using likelihood function approach, we also proposed a confidence set for the change point. The results of a Monte-Carlo simulation have shown the priority of the MLE in identifying the real time of the change.