A Generic Identification Strategy to Determining the Change Point of an Attribute Process

• Main Authors: Lin, Keshan, 林可姍
• Other Authors: Shao, Yuehjen
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/21622543329888331625

碩士 === 輔仁大學 === 統計資訊學系應用統計碩士班 === 100 === Statistical process control (SPC) chart is one of the most important monitoring tools in detecting process disturbances. When an SPC signal is triggered, it is crucial to identify the change point since it is able to effectively help process personnel to quickly identify the root causes of problems. Accordingly, the change point determination has become a promising research issue. The typical maximum likelihood estimate (MLE) approach may be challenged for its deficiencies; particularly, the known prior process distribution and the mathematical difficulties. These deficiencies are quite commonly encountered. In addition, since the attribute process is commonly seen in practice, this study proposes a generic approach to resolve the difficulties of the typical MLE approach in determining the change point of an attribute process. Specifically, the performance among the statistical process control chart alone, the typical MLE approach and the proposed generic mechanism are investigated for the following general cases: (1) a known attribute process distribution and the associated MLE is available to use; (2) an unknown attribute process distribution and the MLE is unable to use, and (3) an unknown attribute process distribution but the MLE is misused. The superior results and the performance of the proposed approach are reported and discussed.