Summary: | 碩士 === 國立交通大學 === 工業工程與管理系 === 88 === The process capability indices have been widely used in the manufacturing industry to provide numerical measures on process performance. The most useful four indices are Cp, Cpk, Cpm, and Cpmk. They combine two important factors- process spread and process centering in one number and make it easy to judge whether a process is acceptable. In this research the statistical properties of the four natural estimated indices will be presented, and the accuracy analysis of the estimated indices will be analyzed by the computational results using relative bias (BIASR), relative mean square error (MSER), and -level confidence relative error (CRER). The computational results are tabled, and the investigation is useful to the practitioners in determining the sample sizes required in their applications for the estimations good to the desired accuracy.
|