Summary: | 碩士 === 國立雲林科技大學 === 工業工程與管理研究所碩士班 === 92 === When control charts find that a special cause is existed, it cannot provide process engineers an indication of the source which caused the unanticipated variation. The signal of the control chart only present us that process has changed, but it doesn’t show us what is the special cause, nor does it show us when the process change really changed. Therefore, knowing when a process has really changed can simplify the search for the special cause. This is good for lowering the probability of incorrectly identifying the special cause.
This research uses the MLE method to find the estimator of the change point of the process mean and process variance. About the process change in the mean, combine the maximum likelihood change point estimator with three kinds of different control charts and compare this estimator’s performance when it is used with the control chart. About the process change in the variance, compare the performance of the estimator used with S2 control chart and individually used the S2 control chart. To know that which one can find the actual change time quickly and correctly.
|