由於資訊科技的進步以及智慧型手機及行動裝置的消費需求增加，全球半導體市場持續快速成長，亦使得半導體產業競爭激烈，如何提升製程的良率與機台效能、減少製程風險及降低生產成本，是晶圓廠提升競爭力的重要因素。半導體的製程有數千道複雜的程序，若其中一道製程或其中一機台腔體(chamber)產生的異常都會造成半導體產品不良率的增加，故製程及機台腔體的異常問題則成為一重要的課題。本研究以某半導體機台即時錯誤偵測與分類系統(Fault detection and classification system, FDC)所蒐集的大量即時製程狀態變異偵測值(Status variable identification, SVID)作統計分析，並評估其機台腔體的健康狀況。本研究以蝕刻製程機台參數為研究方向，透過漢佩爾辨識(Hampel identifier)離群值判別方法從眾多相同型號的機台中找出有差異之機台腔體，為幫助半導體產業工程人員提早判斷機台腔體的狀況，及早做機台腔體的預防保養(Preventive Maintenances, PM) 以改善製程良率。除了判斷機台腔體的健康狀況外，本研究亦利用生產機台腔體之參數來預測晶片製程品質。透過實證分析的結果，本研究之建構方法使工程人員有較佳的異常機台腔體判斷能力，並能找出實際有異常的機台腔體，且透過區別分析法來預測晶片的製程品質，得以幫助工程人員在蝕刻製程中，了解機台腔體之健康狀況以及預測蝕刻製程之量測結果。

Owing to the developement of information technology and the increase of the demand of smart phones and mobile devices, the global semiconductor market has been grown fast which results in intensive competition. Thus, decreasing the risk of manufacturing processes has become essential factors. There are thousands of complicated sequential steps in the semiconductor manufacturing process, and even one abnormal step in chambers could lead to yield loss, which makes the abnormality of process a serious issue. This study analyzes the Status variable identifications (SVID) which are collected from the Fault Detection and Classification (FDC) system in one of the fabrications, and utilizes these data to evaluate states of the production equipment.
This study uses the Hampel identifier outlier detection method in which various etching processes with the equipment parameters data are combined into a single index, to detect abnormal chambers among numerous etching ones. The aim of this study is to use the collected data to determine the appropriate preventive maintenance (PM) schedule for equipments, which can raise the yield rate of the manufacturing process. In addition, this study investigates the proper parameters setting to anticipate better wafer quality during the etching process by using the multi-group discriminant analysis.
We validated the proposed method with a field empirical study, and the results demonstrate the practical viability of this approach. This approach can assist engineers in actively noticing the proper setting of chamber parameters and distinguishing the abnormal chambers during the etching process.

中文部分
白賜清、房克成、張有成與張定昌，SPC統計製程管制，中衛發展中心，民83
周子勤與蕭鉢，應用虛擬量測技術於半導體製程簡介-以CVD/ALD為例，
機械工業雜誌，智慧系統工程技術專輯，291期，108-109，民96
菊地正典(陳連春譯)，透視半導體，建興出版社，民93
張國棟，SPC統計製程管制技術手冊，中國生產力公司，民81
張庭嘉，「建構半導體黃光製程異常升溫現象之模糊監控技術」，碩士論文，國立清華大學工業工程與工業工程管理學系研究所，民96
梁惠姿，「建構半導體製造過程產品異常資料挖礦技術及其雛型系統」，碩士論文，國立清華大學工業工程與工業工程管理學系研究所，民94
梁志翔，「虛擬量測自參數之篩選法則」，碩士論文，國立成功大學製造工程研究所，民 95
陳順宇，多變量分析，華泰書局，民93
曹永誠與徐明照，Run-to-run control簡介與APC framework架構設計，機械工業雜誌 258期，191-199，民93
曹永誠，先進製程控制技術(APC)導論，電機月刊，174，202-215，民94
游欽宏，半導體與平面顯示器製造自動化技術，機械月刊第三十一卷第十二期，
民94
戴鴻恩，利用半導體製成即時參數值建立缺陷特徵化監控模型，國立台灣科技大學管理學研究所EMBA碩士班，民94
譚克平，極端值判斷方法簡介，台東大學教育學報第十九卷第一期，131-150，
民97
羅正忠與張鼎張，半導體製程技術導論，台灣培生教育出版，民95
英文部分
Alwan, L. C. & Roberts, H. V. (1988). Time-series modeling for Statistical Process Control. Journal of Business & Economics Statistics, 6(1), 87-95.
Bedeian, A. G., and Mossholder, K. W. (2000). On the use of the coefficient ofvariation as a measure of diversity. Organizational Research Methods, 3, 285-297.
Bendel, R.B., Higgins,S.S., Teberg,J.E. & Pyke,D.A. (1989). Comparison of skewness coefficient of variation,and Gini coefficient as inequality measures within population. Oecologia,78, 394-400.
Bersimis, S., Psarakis, S. and Panaretos, J. (2007). Multivariate statistical process control charts: an overview. Quality and Reliability Engineering International, 23(5),517-543.
Blue, J., Roussy, A. ,Thieullen, A. & Pinaton, J. (2012). Efficient FDC based on hierarchical tool condition monitoring scheme. Advanced Semiconductor Manufacturing Conference (ASMC), 23rd Annual SEMI, 359-364.
Breunig, M. M., Kriegel, H. P., Ng, R. T. & Sander, J.(2000). LOF: identifying density-based local outlier. Proceedings of SIGMOD Conference, 93-104.
Chen, A., R.S.Guo & Wu, G. X. (1999). Real-time equipment health Evaluation and dynamic preventive maintenance. in Proceedings of the 9th International Symposium on Semiconductor Manufacturing,375-378.
Chen, A. & Blue,J. (2009). Recipe-independent indicator for tool health diagnosis and predictive maintenance. IEEE Transactions on Semiconductor Manufacturing, 22(4) ,522-535.
Chien, C.F., Wang, W.C., and Cheng, J.C. (2007). Data mining for yield enhancement in semiconductor manufacturing and an empirical study. Export Systems with Applications , 33, 192-198.
Choi, B. & Kim, K. (2004). Certain multi sample tests for inverse Gaussian Populations. Communications in Statistics – Theory and Methods, 33, 1557-1576.
Dunnett, C. W. (1955). A multiple comparison procedure for comparing several treatments with a control. Journal of the American Statistical Association, 50, 1096-1121.
Goodlin, B. E., Boning, D. S., Sawin, H. H. & Wise, B. M. (2003). Simultaneous fault detection and classification for semiconductor manufacturing tools. Journal of the Electrochemical Society, 150, 778-784.
Halkidi, M., Batiskakis, Y. & Vazirgiannis, M. (2001). Clustering algorithm and validity measures. Proceedings of the Thirteenth International Conference on Scientific and Statistical Database Management, 3-22.
He, Q. P. & Wang, J. (2007). Fault detection using the k-nearest neighbor rule for semiconductor manufacturing processes. IEEE Transactions on Semiconductor Manufacturing, 20(4),345-354.
Jiang, M.F., Tseng,S.S. & Su, C.M. (2001). Two-phase clustering process for outliers detection. Pattern Recognition Letters, 22(6-7), 691–700.
Kalin, M. (2004). Benefits of APC at Infineon application examples from production fab Villach. in Advanced Semiconductor Manufacturing,ASMC '04. IEEE Conference and Workshop, 437-442.
Klein, M. (2000). Modified S-charts for controlling process variability. Communications in Statistics - Simulation and Computation, 29(3),919-940.
Lacaille, J. & Zagrebnov,M. (2007). An unsupervised diagnosis for process tool fault detection: the flexible golden pattern. IEEE Transactions on Semiconductor Manufacturing, 20(4), 355-363.
Leang, S.& Spanos,C. J. (1997). A General Equipment Diagnostic System and Its Application on Photolithographic Sequences. IEEE Transactions on Semiconductor Manufacturing ,10(3), 329-343.
Li, R., Lin, D. K. J. and Li, B. (2001). Statistical Inference on Large Data Set. Department of Statistics, The Pennsylvania State University Technique Report.
Liu, W. & Zhang, Y. (2004). Multiple comparison of several linear regression models. Journal of the American Statistical Association, 99, 395-403.
Lucas, J. M. & Saccucci, M. S. (1990). Exponentially weighted moving average control schemes: properties and enhancements. Technometrics, 32(1), 1-12.
Magidson, J. & Vermunt, J. K. (2002). Latent class models for clustering: A comparison with K-means. Canadian Journal of Marketing Research, 20, 36-43.
Mirza, A. I., Donoghue, G.O., Drake, A. W. and Graves, S. C. (1995). Spatial Yield Modeling for Semiconductor Wafers, IEEE/SEMI Advanced Semiconductor Manufacturing Conference, 276-281.
Nairy, K. S. & Rao, K. A. (2003). Tests of Coefficients of variation of normal population.Communications in Statistics – Simulation and Computation, 32(3),641-661.
Pearl, J. (1981). A Space-Efficient On-Line Method of Computing Quantile Estimates. Journal of Algorithms, 2, 164-177.
Pinto Da Costa, J. & Soares, C. (2005) . A weighted rank measure of correlation. Aust. N. Z. J. Stat. 47(4), 515–529.
Qin, S. J., Cherry ,G., Good, R. ,Wang, J. & Harrison,C.A. (2006). Semiconductor Manufacturing Process Control and Monitoring: A Fab-Wide Framework. Journal of Process Control, 16, 171-191.
Roes, K. C. B., Does, R. J. M. M. & Schuring, Y. (1993). Shewhart-type control charts for individual observations. Journal of Quality Technology, 25(3),188-19.
Rousseeuw, P.J. and Bassett, G.W. (1990). The Remedian: A Robust Averaging Method for Large Datasets. Journal of the American Statistical Association, 85, 97-104.
Royall, R. M. & Tsou, T.S. (2003). Interpreting statistical evidence using imperfect models:robust adjusted likelihood functions. Journal of the Royal Statistical Society Series B, 65(2), 391-404.
Safavian, S. R., and Landgrebe, D. (1991). A Survey of Decision Tree Classifier Methodology. IEEE Transactions on Systems, Man, and Cybernetics, 21(3), 660-674.
Schirru, A., Pampuri,S. & Giuseppe De Nicolao (2010). Multilevel statistical process control of asynchronous multi-stream process in semiconductor manufacturing.
in proceedings of 6th IEEE Conference on Automation Science and Engineering,57-62.
Schirru, A.,Pampuri,S., DeLuca,C. & DeNicolao,G. (2011). Multilevel kernel methods for virtual metrology in semiconductor manufacturing. in 18th IFAC World Congress, 11614-11621.
Sim, C. H. (2000). S-Chart for Non-Gaussian Variables. Journal of Statistical Computation and Simulation, 65(2),147-156.
Spanos, C. J., Guo ,H. F., Miller, A. & Parrill ,J.L. (1992). Real-time statistical process control using tool data. IEEE Transactions on Semiconductor Manufacturing,5(4),308-318.
Taagepera, R.& Ray, J.L. (1977). A Generalized index of concentration. Sociol. Methods Res. 5, 367–384.
Tierney, L. (1983). A Space-Efficient Recursive Procedure For Estimating A Quantile of An Unknown Distribution. SIAM Journal on Scientific And Statistical Computing, 4(4),706-711.
Tsou, T. S. (2003). Comparing two population means and variances - a parametric robust way. Communications in Statistics - Theory and Methods”, 32(10), 2013-2019.
Wang, H., Wang, W., Yang, J., & Yu, P. S. (2002).Clustering by Pattern Similarity in Large Data Sets. Proceeding of ACM Special Interest Group on Management of Data, 394–405.
Wilcox, R. (2003). Applying contemporary statistical techniques.Amsterdam: Academic Press.
Woodall, W. H. (1997). Control Charts Based on Attribute Data:Bibliography and Review. Journal of Quality Technology,29(2), 172-183.
Yamanishi, K., Takeuchi, J.I., Williams,G. & Milne,P.(2004). On-line unsupervised outlier detection using finite mixtures with discounting learning algorithms. Data Mining and Knowledge Discovery, 8(3)275-300.
Yeh, A. B., Huwang, L.& Wu C.W. (2005). A multivariate EWMA control chart for monitoring process variability with individual observations. IIE Transactions, 37,1023-1035.
Yue, H. H. and Tomoyasu, M. (2004). Weighted principal component analysis and its applications to improve FDC performance. in Proc. 43rd IEEE Decision and Control Conference, 4(14-17), 4262–4267.