具有多項異常原因之 X-bar管制圖經濟設計

Economic Design of X ̅ Control Chart for Multiple Assignable Causes

蘇珈儀(Chia-Yi Su)

指導教授 ： 彭文理

國立交通大學/管理學院/工業工程與管理系所/博士(2012年)

https://doi.org/10.6842/NCTU.2012.00074

摘要

本論文主要研究製程異常原因發生為多項異常原因之 X ̅ 管制圖經濟設計。研究中製程生產方式分為斷續工件生產及連續流動生產兩種；製程失效機構則考慮指數分配與韋氏分配兩種。根據生產方式及製程失效機構，在論文中共提出二項研究主題：(1)連續流動生產且製程失效機構屬指數分配之 X ̅ 管制圖經濟設計，(2)斷續生產且製程失效機構屬韋氏分配之 X ̅ 管制圖經濟設計。
對於這兩種經濟設計模式，我們利用抽樣方法和成本結構來建構損失成本函數，並在損失成本最小化下來找尋最佳的 (樣本大小)， (抽樣間隔時間)， (管制界限係數)。由於經濟設計模式進行敏感度研究，可以提供管理者或工程師了解輸入參數對模式的影響。因此，我們也將對最佳的 , , 值進行敏感度分析，藉此分析來了解時間參數或成本參數的變動後，對於最佳 , , 值之影響。最後，我們有提供數值結果並討論之。
除此之外，我們將對多項異常原因下僅一個異常發生之模式和多項異常原因下有二個異常發生之模式進行比較分析，藉由數值結果可以得知考慮多項異常原因下有二個異常發生之模式對於降低品質成本和增加其在斷續生產之競爭力是有一個有用的方法。

關鍵字

X ̅管制圖 ； 多項異常原因 ； 經濟設計 ； 韋氏衝擊模氏 ； 指數衝擊模式 ； 斷續生產 ； 連續流動生產

並列摘要

In this dissertation, we analyze the economic design of -control charts and extend the model for the case of multiple assignable causes to allow for the second occurrence of an assignable cause following the first occurrence. In addition, two process failure mechanisms are investigated in different manufacturing environments. One is the Exponential failure mechanism in a continuous flow process and another is the Weibull failure mechanism in a discrete part process.
For those two models, the expected loss-cost functions are established by the sampling scheme and cost structure. Optimal values of the economic design parameters including the sampling size( ), the sampling intervals ( ) and control limit coefficient ( ) are determined by minimizing loss-cost functions. Because of sensitivity investigation on the model with critical input parameters may provide some answers for the model analyst. A sensitivity analysis is provided to discuss how the model can be affected by the time parameters or cost parameters in the investigated model. For illustration purpose, numerical results are also presented.
Subsequently, we perform comparative analysis between the model that once an assignable cause occurs, no further assignable causes will occur and the modified model that allow for the second occurrence of an assignable cause following the first occurrence. Our numerical investigations showed that a modified model should be helpful in reducing the quality cost and increasing competitiveness in a discrete part process.

並列關鍵字

Continuous Flow Process ； Discrete Part Process ； Exponential shock model ； Economic design ； Multiple assignable causes ； Weibull shock model ； x-bar-control chart

參考文獻

2.Banerjee, P. K. and Rahim, M. A. (1987). The economic design of control charts: A renewal theory approach. Engineering Optimization, 12, 63-73.

3.Banerjee, P. K. and Rahim, M. A. (1988). Economic design of x-bar control charts under Weibull shock models. Technometrics, 30, 407-414.

4.Barker, K. R. (1971). Two process model in the economic design of an x-bar chart. AIIE Transactions, 3, 257-263.

5.Chen, Y. S. and Yang, Y. M. (1999). Economic design of x-bar control charts for continuous flow process with Weibull in-control times. The Asia Quality Symposium, 17-20.

6.Chen, Y. S. and Yang, Y. M. (2000). Design of economic x-bar control charts in a continuous flow process – Case for multiple assignable causes. The Asia Quality Symposium, 85-91.