統計製程管制是品質管制領域中重要的技術之一，其中以利用機率原理為基礎，在製程為穩定、常態且生產數據間維持獨立的前提下所建構的修華特管制圖為目前應用最普遍的製程監控工具。然而，製程日益精密與自動化的情況下，實務上連續性製程已不符傳統管制圖常態、獨立及穩定之前提假設，使用傳統統計機率為主取最近一筆樣本的異常判讀法則常導致過多的錯誤警訊，造成管理階層過度干預反使製程無法維持穩定狀態。本研究以萃取歷史生產數據的圖形特徵值，使用k-最鄰近法進行大量實務資料類別值標定，將標定後的資料進行決策樹的訓練與學習，訓練後的決策樹分類正確率高達93%，同時亦分析歸納相對重要屬性供未來生產穩定之判定。藉由此模型建立的方法與流程，期可協助業界做為製程異常分類參考，進一步縮減可歸屬原因的調查範圍與降低反應時間，供管理者有效資訊進行及時調整與矯正。

Statistical process control（SPC） techniques are widely used in manufacturing and other processes to monitor mean and variation in quality characteristics. Shewhart control chart based on probability, stable process, normal distribution and independent is a common tool to identify if the abnormity occurs. However, the process is getting more precise and automotive, continuous process does not fit in with traditional control chart assumption nowadays. More and more false alarms are caused by using the latest sample to judge whether the process is out of control. In practice, management based on these alarms to adjust the process often leads to the process is getting unstable. In this study, at first we extracted the 8 features from the historical process data and label the class of some instances as the standard, then k-NN is based on the standard instances to label the class of remain instances. After that, these labeled instances are utilized to decision tress growth. The effectiveness of the methodology has a success rate up to 93% in CCP recognition. By way of this model establishment, we hope to narrow down the scope of possible causes and speeds up the troubleshooting process.

中文

孫靜、張公緒，常用管制圖標準及原理，中華民國品質學會，台灣，2002。

英文

Chen, Z., Lu, S., and Lam, S., A hybrid for SPC concurrent pattern recognition. Advanced Engineering Informatics, 21(3), 303-310, 2007.

Demirli, K. and Vijayakumar, S., Fuzzy assignable cause diagnosis of control chart patterns. Proceedings of the 27th Annual Meeting of North American Fuzzy Information Processing. 1-6, New York: New York, 2008.

Gauri, S.K. and Chakraborty, S., A study on the various features for effective control chart pattern recognition. International Journal of Advanced Manufacturing Technology, 34(3-4), 385-398, 2007.

Gauri, S.K. and Chakraborty, S., Improved recognition of control chart patterns using artificial neural networks. International Journal of Advanced Manufacturing Technology, 36(11-12), 1191-1201, 2008.

Gauri, S.K. and Chakraborty, S., Recognition of control chart patterns using improved selection of features. Computers & Industrial Engineering, 2008, in press.

Gullbay, M. and Kahraman, C., An alternative approach to fuzzy control charts：Direct fuzzy approach. Information Sciences, 177(6), 1463-1480, 2007.

Guh, R.S. and Shiue, Y.R., Effective pattern recognition of control charts using a dynamically trained learning vector quantization network. Journal of the Chinese Institute of Industrial Engineers, 25(1), 73-89, 2008.

Guh, R.S. and Shiue, Y.R., An effective application of decision tree learning for on-line detection of mean shifts in multivariate control charts. Computers & Industrial Engineering, 55(2), 475-493, 2008.

Kusiak, A. and Smith, M., Data mining in design of products and production systems. Annual Reviews in Control, 31(1), 147-156, 2007.

Kusiak, A. and Song, Z., Combustion efficiency optimization and virtual testing：A data-mining approach. IEEE Transactions on Industrial Informatics, 2(3), 176-184, 2006.

Montogmery, D.C. Introduction to Statistical Quality Control. (5th ed.). New York : John Wiley, 2005.

Nelson, L.S., The Shewhart control chart-test for special causes. Journal of Quality Technology, 16(4), 237-239, 1984.

Quinlan, J.R., C4.5: Programs for Machine Learning. Amsterdam: Morgan Kaufmann, 1993.

Wang, C.H., Guo, R.S., Chiang, M.H., and Wong, J.Y., Decision tree based control chart pattern recognition. International Journal of Production Research, 46(17), 4889-4901, 2008.

Wang, C.H. and Kuo, W., Identification of control chart patterns using wavelet filtering and robust fuzzy clustering. Journal of Intelligent Manufacturing, 18(3), 343-350, 2007.

Western Electric Company, Statistical Quality Control Handbook. (2nd ed.). New York : The Company, 1958.

Yang, J.H. and Yang, M.S., A control chart pattern recognition system using a statistical correlation coefficient method. Computers and Industrial Engineering, 48(2), 205-221, 2005.