單邊規格下不同附加條件之製程能力量測

Abstract

製程能力指標經常被用來衡量製程製造產品符合規格的能力，不僅是提供品質保證的工具，也是提供品質改善方面的一個方針。本篇論文主要是探討如何應用單邊規格的製程能力指標 和 在不同附加條件之下，例如：製程平均值偏移、製程變異數改變、量測誤差、工具汰換、允收抽樣、供應商選擇等實務問題，藉由程式的撰寫與執行以得出不同的信賴下界和臨界值，其中，信賴下界為對製程的最小能力測量，臨界值為判斷製程是否合格的依據，此兩種準則是製造商與顧客間檢驗製程的標準，本篇論文更表格化在不同附加條件之下的信賴下界與臨界值，以便於製造商與顧客使用。

Process capability indices (PCIs) have been widely used in the manufacturing industry to provide yield measures on process capability, which are effective tools for quality assurance and guidance for process improvement. The well-known process capability indices and are one-sided capability specifications with many product characteristics. In this thesis, we study the processes with side conditions, process with mean shift, process with variance change, process with measurement error, tool wear, product acceptance deterioration, and supplier selection, which are some situations that the industries often faced with unilateral specification. Several criteria are used to judge the capability of a process, including process yield, the lower confidence bound, and the critical value. The lower confidence bound provides a measure on the minimum capability of the process based on sample data, and the critical value is as a standard for engineer/practitioner to determine whether the process is capable or not by some statistical testing. The Matlab computer programs of various capability measures for one-sided processes with side conditions are developed and we also tabulate the computational results of the program for the convenience.