Applying Filter design Approach to Statistical Process Control for Lewis’s ARMA Model

• Main Authors: Jhe-hung Yeh, 葉哲宏
• Other Authors: Ying-chieh Yeh
• Format: Others
• Language: en_US
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/32261600602820359382

碩士 === 國立中央大學 === 工業管理研究所 === 99 === In this study, we apply filter design to statistical process control (SPC) and discuss the impact of different process distributions. Instead of using conventional autoregressive moving-average processes, we assume Lewis’s autoregressive moving-average (ARMA) processes as data processes. The control chart we used in this study is the exponentially weighted moving average (EWMA) control chart. We will apply linear filter on the observations generated from our data process to obtain our control chart statistic. With the control chart statistic, we can calculate the out-of-control ARL by Markov chain method. And our research objective is to reduce the out-of-control ARL with a predetermined in-control ARL. In the final, we adjust parameter and transform distribution to propose a relatively simple algorithm. Therefore, we can avoid complex and time-consuming calculation.