應用類神經網路偵測製程平均值之變化：設計策略之研究

A Neural Network Approach for Detecting Changes in Process Mean : the Investigation of Design Strategy

王仁達(Jen Ta Wang)

指導教授 ： 鄭春生

元智大學/工程學院/工業工程與管理學系/碩士(1997年)

摘要

管制圖通常被用來監視一個製造程序之製程變化。當管制圖上之點
超出管制界限或出現非隨機性之變化時，則表示製程為管制外。在本研
究中，我們以類神經網路發展一個監視製程變化之管制程序。

本研究所使用之類神經網路為多層之倒傳遞網路。我們所考慮的品
質特性為樣本平均值或事件出現之比例。類神經網路之輸入為樣本統計
量及一些參考指標。一些類神經網路之重要參數是以田口實驗設計方法
來決定。在論文中，我們也討論整個系統之設計策略。模擬分析顯示，
類神經網路之平均連串長度特性優於傳統累和管制法。

關鍵字

類神經網路 ； 倒傳遞網路 ； 平均連串長度 ； 累和管制法 ； 田口實驗設計

並列摘要

Control charts are often used in manufacturing processes for
monitoring the process changes. A control chart may indicate
an out-of-control condition either when one or more points fall
beyond the control limits, or when the plotted points exhibit
some nonrandom pattern of behavior. In this research, we
consider an alternative process monitoring approach based on
artificial neural networks.

The artificial neural network used in this research is a
multi-layer network trained by back-propagation algorithm.
We consider the quality characteristics expressed in terms of
sample means or the rate of occurrences of events. The inputs
of artificial neural networks consist of sample statistics and
some heuristic indexes. Some critical design factors are
determined by Taguchi''s experimental design. A discussion of
design strategies is also provided. Extensive simulation
results show that the proposed neural network-based procedures
are superior to CUSUM control charts in terms of the average
run lengths.

並列關鍵字

CUSUM ； Artificial Neural Networks ； Back-propagation ； Average Run Length ； Taguchi''s experimental design

參考文獻

2.Anderson, T. W., An Introduction to Multivariate Statistical Analysis, John Wiley, NY, (1958).

3.Brook, D. and D. A. Evans, "An approach to the probability distribution of cusum run length," Biometrika, 59, 3, 539-549 (1972).

4.Chang, S. I. and C. A. Aw, "A neural fuzzy control chart for detecting and classifying process means shifts," International Journal of Production Research, 34, 8, 2265-2278 (1996).

5.Chang, T. C. and F. F. Gan, "A cumulative sum control chart for monitoring process variance," Journal of Quality Technology, 27, 2, 109-119 (1995).

6.Cheng, C. S., "Detecting changes in the process mean using artificial neural networks approach," Journal of Chinese Institute of Industrial Engineers, 11, 1, 47-54 (1994).