利用無母數方法來監控一般線性輪廓的EWMA管制圖

Exponentially weighted moving average control chart for monitoring general linear profiles based on nonparametric method

張智翔(Chang, Chih Hsiang)

指導教授 ： 黃榮臣

國立清華大學/理學院/統計學研究所/碩士(2015年)

https://doi.org/10.6843/NTHU.2015.00309

摘要

在許多工業製程中，產品或製程的品質特徵可以用一個反應變數以及一個或多個解釋變數的函數關係來界定，這種函數關係被稱為輪廓製程。本文主要是利用無母數的重複抽樣方法來建構監控線性輪廓製程的管制圖。我們用統計模擬來評估新管制圖的監控效果並說明傳統上假設隨機誤差項為常態分佈所構造的管制圖為何不具有良好的監控效果。最後我們利用一個例子來說明實務上如何操作與使用我們所提出的管制圖。

關鍵字

管制圖 ； 線性輪廓 ； EWMA ； 無母數重抽法

並列摘要

In many industrial processes, the quality of a product or a process can be represented by the relationship of a function between a response variable and one or more explanatory variables. This function is called a profile. This article uses a nonparametric resampling method to develop a control chart to monitor the linear profile process. We evaluate the effectiveness of the new and the traditional control charts by simulation studies. The traditional control charts assuming normal random error term is shown not having good performance. Finally, we use an example to illustrate how to complement our proposed control chart.

參考文獻

[1] Crowder, S. V., and Hamilton, M. D. (1992). “An EWMA for Monitoring a Process Standard Deviation”. Journal of Quality Technology 24, pp. 12-21.

[2] Eyvazian, M., Noorossana, R., Saghaei, A., and Amiri, A. (2011). “Phase II Monitoring of Multivariate Multiple Linear Regression Profiles”. Quality and Reliability Engineering International 27, pp. 281-296.

[3] Huwang, L., Wang, Y., and Shen, C. (2014). “Monitoring General Linear Profiles When Random Errors Have Contaminated Normal Distributions”. Quality and Reliablility Engineering International 30, 1131-1144.

[4] Kang, L., and Albin, S. L. (2000). “On-Line Monitoring When the Process Yields a Linear Profile”. Journal of Quality Technology 32, pp. 418-426.

[5] Kim, K., Mahmoud, M. A., and Woodall, W. H. (2003). “On the Monitoring of Linear Profiles”. Journal of Quality Technology 35, pp. 317-328.