| Summary: | 博士 === 國立臺灣科技大學 === 工業管理系 === 97 === The generally weighted moving average (GWMA) control chart was developed in the 2003s. Due to the added adjustment parameter, the weight function of the GWMA is more flexible and the GWMA control chart performs substantially better than the exponentially weighted moving average (EWMA) control chart for monitoring small shifts in the process mean. Besides the process mean shifts, it is also important to monitor either the process variance or both mean and variance simultaneously. The main problem addressed in this study is to develop the new control chart by GWMA techniques for monitoring the process variance, and combine the GWMA mean chart and variance chart for detecting small shifts in the process mean and variance at the same time. Finally, use an easy way to improve the performance of the GWMA control chart for monitoring median shifts in the process mean.
This thesis is divided into three major parts.
Firstly, some GWMA variance charts are applied to the sample statistics. Simulation is utilized to evaluate the average run length (ARL) of the EWMA and GWMA control charts. Numerous comparisons of ARLs indicate that the GWMA control chart is more sensitive than the EWMA control chart for detecting small shifts in the process variance.
Secondly, a combined scheme consisting of the GWMA mean chart and variance chart for detecting process variability is developed. Numerous comparisons of ARLs indicate that the combination of the GWMA charts is more sensitive than using two single GWMA control charts when the process mean and variance shifts occurred at the same time. The combined GWMA control chart is also more sensitive than the combination of the EWMA charts for detecting small shifts in the process mean and variance.
Thirdly, this part extends the GWMA control chart by the double weighted technique. The proposed chart is called the double generally weighted moving average (DGWMA) control chart. A simulation result indicates that the DGWMA control chart with time-varying control limits is more sensitive than the GWMA and the double exponentially weighted moving average (DEWMA) control charts for detecting medium shifts in the mean of a process.
|
|---|
