A Simulation Study on the Average Run Length of Moving Range Control Charts

• Main Authors: Ding-Kai Hong, 洪鼎凱
• Other Authors: Chao-Yu Chou
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/7s9dy3

碩士 === 國立臺中科技大學 === 財務金融系碩士班 === 101 === Control chart is one of seven quality control tools which can be used on the process to give an appropriate monitoring and judgment. When the process is out of control, the control chart can detect the process variation such that some actions may be taken before producing much more defective products. In this study, we use the framework of the moving range control charts to probe the influence of expansion of standard deviation on average run length performance under different types of data. According to our simulation results, we find that there exist significantly different performances among average run lengths of moving range charts for different independent data. The performance of out-of-control average run lengths can be improved by increasing multiply of standard deviation. In the different single type of data, the right-skewed data of average run length with α3=0.75, α4=3 have the best performance, which means the normal distribution data are not the best. For the auto-correlated data, the out-of-control average run lengths of moving range chart is always longer than those for independent data.