研究生: |
王李伊秀 Lee, I-Hsiu Wang |
---|---|
論文名稱: |
應用類神經網路於改善MCUSUM管制圖之績效 Apply Neural Networks Methods on Improving the Performance of MCUSUM Control Chart |
指導教授: |
王泰裕
Wang, Tai-Yue |
學位類別: |
碩士 Master |
系所名稱: |
管理學院 - 工業與資訊管理學系 Department of Industrial and Information Management |
論文出版年: | 2018 |
畢業學年度: | 106 |
語文別: | 中文 |
論文頁數: | 77 |
中文關鍵詞: | 多變量管制圖 、類神經網路 、支援向量機 、深度學習 |
外文關鍵詞: | multivariate control chart, neural networks, support vector machine, deep learning |
相關次數: | 點閱:104 下載:5 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
管制圖是企業常用來監控製程的一項工具,當製程要同時考慮多個品質特性且這些特性間具有相關性時,為了避免誤判的情形發生,便會使用多變量管制圖做為主要的監控工具。MCUSUM管制圖因考慮了製程的歷史資料,在製程平均向量為中小幅度偏移的情形下,偵測績效比最常被使用的Hotelling’s T2管制圖來得好,但當製程平均向量呈大偏移的情形下,偵測能力並不如Hotelling’s T2管制圖。此外,當MCUSUM管制圖偵測出製程發生異常時無法直接獲得偏移的相關資訊,必須透過計算繁複的分解法或是主成分分析法來解決,但這些方式都只能求出造成製程異常的品質特性,無法從中得知關於偏移的其他細節,因此許多學者利用類神經網路優異的學習能力來得到管制圖偏移的相關資訊。本研究主要針對MCUSUM管制圖進行績效改善,分別透過倒傳遞網路、支援向量機及深度學習網路建立兩個偵測模型。首先,將資料轉為MCUSUM統計量後,利用模型1來找出超出管制界限之資料點,若偵測到異常再將異常點放入模型2中以取得偏移資訊。當MCUSUM管制圖考慮愈多品質特性時,在相同偏移程度上,ARL1有明顯上升的情形,表示受品質特性增加的影響使偵測績效變差,而本研究所提出的類神經網路方法能有效改善MCUSUM管制圖之績效。本研究使用的類神經網路方法中,倒傳遞網路只於小分析視窗時能顯著改善MCUSUM管制圖績效,隨著分析視窗增加,改善程度逐漸降低。當製程為小幅度偏移時,支援向量機及深度學習的偵測績效都受到分析視窗增加影響,但改善程度都較倒傳遞網路大,其中深度學習受情境改變影響最小,因此在模型1及模型2都有較佳的表現。最後本研究利用模擬的製程數據來進行實例驗證,把資料放入模型當中並與傳統MCUSUM管制圖做比較,可發現加入類神經網路能更快地偵測到製程偏移,且透過模型2可進一步知道造成偏移的品質特性及其偏移量大小。
Control chart is one of effective tools to monitor manufacturing process. People use multivariate control charts as the major methods to monitor more than one correlated characteristics. The MCUSUM control chart has better performance than Hotelling’s T2 control chart when the mean vector of a process is with moderate or small shift. However, the MCUSUM control chart works poorly as the shift increases. When MCUSUM control chart detects the signal of process, we usually access the relevant information through the decomposition technique or principal analysis. Unfortunately, these methods can just find out which charaterestic(s) cause the signal, but can’t offer advanced information. Therefore, application of neural networks on multivariate control chart has been investigated by several researchers. In this study, we propose two models to improve performance of MCUSUM control chart by Back Propogation Neural networks(BPN), Support Vector Machine(SVM), and deep learing, respectively. First, the model 1 is used to detect whether there is any out-of-control event or not. If it happends, we apply the model 2 to recognize magnitude of shift for each variable simultaneously. The ARL1 of MCUSUM control chart will increase when the process includes more characteristics, so we use nerual network methods to improve it. BPN performs well only when the window size is small enough. With the increment of window size and quality characteristics, SVM and deep learning are better than BPN. Deep learning has the best performance in these two models. It means we can find out the signal more quickly and clarify the unusal characteristics accurately through deep learning methods.
Bersimis, S., Psarakis, S. and Panaretos, J. (2007). Multivariate statistical process control charts: an overview. Quality and Reliability Engineering International, 23(5), 517-543.
Bergstra, J. and Bengio, Y. (2012). Random search for hyper-parameter optimization. Journal of Machine Learning Research, 13, 281-305.
Chang, C. C. and Lin, C. J. (2011). LIBSVM: A library for support vector machines. ACM Transactions on Intelligent Systems and Technology, 2(3).
Chang, S. I. and Aw, C. A. (1996). A neural fuzzy control chart for detecting and classifying process mean shifts. International Journal of Production Research, 34(8), 2265-2278.
Chen, L. H. and Wang, T. Y. (2004). Artificial neural networks to classify mean shifts from multivariate chart signals. Computers & Industrial Engineering, 47(2-3), 195-205.
Cheng, C. S. (1994). Detecting changes in the process mean using artificial neural networks approach. Journal of Chinese Institute of Industrial Engineers, 11, 47-54.
Cheng, C. S. (1995). A multi-layer neural network model for detecting changes in the process mean. Computers & Industrial Engineering, 28(1), 51-61.
Cheng, C. S. and Cheng, S. S. (2001). A neural-based procedure for the monitoring of exponential mean. Computers & Industrial Engineering, 40, 309-321.
Crosier, R. B. (1988). Multivariate generalizations of cumulative sum quality control schemes. Technometrics, 30(3), 291-303.
Du, S., Lv, J. and Xi, L. (2012). On-line classifying process mean shifts in multivariate control charts based on multiclass support vector machines. International Journal of Production Research, 50(22), 6288-6310.
El-Midany, T. T., El-Baz, M. A. and Abd-Elwahed, M. S. (2010). A proposed framework for control chart pattern recognition in multivariate process using artificial neural networks. Expert Systems with Applications, 37(2), 1035-1042.
Guh, R. S. (2007). On-line identification and quantification of mean shifts in bivariate processes using a neural network-based approach. Quality and Reliability Engineering International, 23(3), 367-385.
Hinton, G. E., Osindero, S. and Teh, Y. W. (2006). A fast learning algorithm for deep belief nets. Neural Computation, 18(7), 1527-1554.
Hotelling, H. (1947). Multivariate quality control-illustrated by the air testing of sample bombsights. Techniques of Statistical Analysis, Eisenhart, C., Hastay, M. W. and Wallis, W. A., McGraw-Hill, New York, 11-184.
Hsu, C. W., Chang, C. C. and Lin, C. J. (2003). A practical guide to support vector classification. Technical Report, Department of Computer Science and Information Engineering, National Taiwan University.
Kanellopoulous, I. and Wilkinson, G. G. (1997). Strategies and best practice for neural network image classification. International Journal of Remote Sensing, 18(4), 711-725.
Kumar, S. (2004). Neural networks : a classroom approach. Tata McGraw-Hill, New Delhi.
Lecun, Y., Bottou, L. and Bengio, Y. (1998). Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11), 2278-2324.
Lowry, C. A. and Montgomery, D. C. (1995). A review of multivariate control charts. IIE Transactions, 27(6), 800–810.
Lowry, C. A., Woodall, W. H., Champ, C. W. and Rigdon, S. E. (1992). A multivariate exponentially weight moving average control chart. Technometrics, 34(1), 46-53.
Lucas, J. M. (1982). Combined Shewhart-CUSUM quality control schemes. Journal of Quality Technology, 14(2), 51-59.
Mason, R. L., Tracy, N. D. and Young, J. C. (1997). A practical approach for interpreting multivariate T2 control chart signals. Journal of Quality Technology, 29(4), 396-406.
Montgomery, D. C. (1996). Introduction to Statistical Quality Control, Wiley, New York, USA.
Patel, H., Thakkar, A., Pandya, M. and Makwana, K. (2017). Neural network with deep learning architectures. Journal of Information and Optimization Sciences, 39, 31-38.
Pignatiello, J. J. and Runger, G. C. (1990). Comparisons of multivariate CUSUM charts. Journal of Quality Technology, 22(3), 173–186.
Psarakis, S. (2011). The use of neural networks in statistical process control charts. Quality and Reliability Engineering, 27(5), 641-650.
Pugh, G. A. (1989). Synthetic neural networks for process control. Computers and Industrial Engineering, 17, 24-26.
Rumelhart, D. E., Hinton, G. E. and Williams, R. J. (1986). Learning internal representations by error propagation, Parallel Distributed Processing. MIT Press, Cambridge, Massachusetts, 318-362.
Salehi, M., Kazemzadeh, R. B. and Salmasnia, A. (2012). On line detection of mean and variance shift using neural networks and support vector machine in multivariate processes. Applied Soft Computing, 12(9), 2973-2984.
Seide, F., Li, G. and Yu, D. (2011). Conversational speech transcription using context-dependent deep neural networks. Interspeech, 437-440.
Sun, R. and Tsung, F. (2003). A kernel-distance-based multivariate control chart using support vector methods. International Journal of Production Research, 41, 2975-2989.
Vapnik, V. (1995). The Nature of Statistical Learning Theory, Springer-Verlag, New York.
Wang, T. Y. and Chen, L. H. (2002). Mean shifts detection and classification in multivariate process: a neural-fuzzy approach. Journal of Intelligent Manufacturing, 13(3), 211-221.
Woodall, W. H. and Ncube, M. M. (1985). Multivariate CUSUM quality-control procedures. Technometrics, 27(3), 285-292.
Wu, C., Liu, F. and Zhu, B. (2015). Control chart pattern recognition using an integrated model based on binary-tree support vector machine. International Journal of Production Research, 53(7), 2026-2040.
Yu, J. B. and Xi, L. F. (2009). A neural network ensemble-based model for on-line monitoring and diagnosis of out of control signals in multivariate manufacturing processes. Expert Systems with Applications, 36, 909–921.
Zorriassatine, F. and Tannock, J. D. T. (1998). A review of neural networks for statistical process control. Journal of Intelligent Manufacturing, 9(3), 209-224.