簡易檢索 / 詳目顯示

回結果列表
研究生： 鄭仁傑
Jen-Chieh Cheng
論文名稱： 以混合決策樹方法分析有相互關係之半導體製造資料
A hybrid decision tree approach for analyzing interrelated semiconductor manufacturing data
指導教授： 簡禎富
Chen-Fu Chien
口試委員:
學位類別： 碩士
Master
系所名稱： 工學院 - 工業工程與工程管理學系
Department of Industrial Engineering and Engineering Management
論文出版年： 2003
畢業學年度： 91
語文別： 英文
論文頁數： 67
中文關鍵詞： 資料挖礦 、決策樹 、交互影響檢測 、事故診斷 、半導體製造資料
外文關鍵詞： data mining, decision tree, interaction detection, defect diagnosis, semiconductor manufacturing data
相關次數： 點閱：110  下載：0 
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

    • 在數位資訊時代，大量的資料儲存於資料倉儲或資料庫中，從這些資料能翠取豐富資訊以供知識發現與決策分析。在半導體的製造過程中，大量的製程資料會收集到工程資料庫中以便進行製程監控、故障分析與製造管理。然而因為半導體的製程複雜，而且影響的變因眾多且通常具有相互關係，工程師往往難以藉著本身的專業知識或是經驗法則，從分析的資料中迅速且有效率的發覺導致製程異常的原因以及可能隱藏的資訊，並且迅速的處理事故問題。本研究建構半導體資料挖礦架構，並發展混合決策樹方法，其中包含Kruskal-Wallis檢定、卡方交互影響檢測、變異降低分支法則，以尋找可能造成製程變異的原因，做為工程師及領域專家解決問題的參考依據，並且協助縮短事故診斷的時間，進而提升半導體製程的良率。本研究運用真實資料比較混合決策樹方法與現行決策樹演算法的表現，並以某半導體廠之案例為實證，以檢驗本研究的可行性。

    In the age of digital information, large amounts of data are recorded in data warehouse or database. Such data may provide a rich resource for knowledge discovery from database and decision support. During the fabrication process, a large amount of process data will be automatically or semi-automatically recorded and accumulated in the engineering database for process monitoring, fault diagnosis and manufacturing management. However, in semiconductor industry, the yield of a silicon wafer is affected by many manufacturing factors that are often interrelated. It is difficult for domain engineers to find possible root causes rapidly and efficiently by own domain knowledge or rule of thumb. In this study, we construct a data mining conceptual framework for analyzing semiconductor manufacturing data, and propose a hybrid decision tree approach, including Kruskal-Wallis test, chi-square interaction detection, and variance reduction splitting criterion to explore the huge engineering data to analyze the semiconductor manufacturing data and infer the possible fault causes of manufacturing process variation. The information is helpful to engineers as the basis of the trouble shooting and defect diagnosis. In this study, we use real data to compare the performance of hybrid decision tree approach with current decision tree algorithms, and apply a real case from a semiconductor fabrication company as empirical study and the results show the practical viability of this approach.

    摘要 I ABSTRACT II ACKNOWLEDGEMENT III TABLE OF CONTENTS IV LIST OF FIGURES VI LIST OF TABLES VII CHAPTER 1 INTRODUCTION 1 1.1 Background, significance, and motivation 1 1.2 Research aims 1 1.3 Overview of this thesis 2 CHAPTER2 LITERATURE REVIEW 3 2.1 Semiconductor manufacturing process 3 2.2 Data mining 8 2.2.1 Data mining model functions 8 2.2.2 Data mining techniques 9 2.2.3 Data mining process 10 2.2.4 Data mining applications 12 2.3 Decision tree 13 2.3.1 Decision tree construction 14 2.3.2 CART 19 2.3.3 C4.5 20 2.3.4 CHAID 22 CHAPTER 3 A HYBRID DECISION TREE APPROACH FOR DIAGNOSING DEFECTS 25 3.1 Problem definition 27 3.2 Data selection 28 3.3 Data preprocessing 30 3.4 Decision tree construction 30 3.4.1 Step 1: Using Kruskal-Wallis test to indicate the significant attributes 33 3.4.2 Step 2: Detecting two-variable interaction among these pair significant attributes 35 3.4.3 Step 3: Using variance reduction criterion to split among the candidate attributes 38 3.4.4 Step 4: Iterative growing 40 3.5 Evaluation and interpretation 40 CHAPTER 4 AN EMPIRICAL STUDY 42 4.1 The real dataset “Servo” 42 4.1.1 Problem definition of real dataset “Servo” 42 4.1.2 Data selection of real dataset “Servo” 42 4.1.3 Data preprocessing of real dataset “Servo” 43 4.1.4 Decision tree construction of real dataset “Servo” 43 4.1.5 Evaluation and interpretation of real dataset “Servo” 46 4.2 A real case in semiconductor manufacturing fab 52 4.2.1 Problem definition of real case 52 4.2.2 Data selection of real case 53 4.2.3 Data preprocessing of real case 53 4.2.4 Decision tree construction of real case 55 4.2.5 Evaluation and interpretation of real case 59 CHAPTER 5 CONCLUSION 63 REFERENCES 65

    Allen, B. (1994), “Case-based reasoning: business applications”, Communications of the ACM 37 (3), pp.40-42.
    Barr, D. S. and G. Mani (1994). “Using neural nets to manage investments”, Al Expert, pp.16-21.
    Biggs, D., Vallie, B. D., and Suen E., (1991), “A method of choosing multiway partitions for classification and decision tree”, Journal of applied statistics, 18,1, pp.49-62.
    Bose, I. and R. K. Mahapatra (2001), “Business data mining—a machine learning perspective”, Information & Management 39, pp.211-225.
    Braha, D. and A. Shmilovici (2002), “Data mining for improving a cleaning process in the semiconductor industry”, IEEE Transactions on Semiconductor Manufacturing, Vol.15, No.1, pp.91-101.
    Brachman, R. J., Khabaza, T., Kloesgen, W., Piatetsky-Shapiro, G., and Simoudis, E. (1996), “Mining Business DataBase”, Communication of ACM, vol.39, no.11, pp.42- 48.
    Breiman, L., Friedman, J. H., Olshen, R. J., and Stone, C. J. (1984), Classification and Regression Trees, Belmont, CA:Wadsworth.
    Chien, C., Lin, T., Peng, C., and Hsu, S. (2001), “Developing data mining framework and methods for diagnosing semiconductor manufacturing defects and an empirical study of wafer acceptance test data in a wafer fab”, Journal of the Chinese Institute of Industrial Engineers, Vol.18, No.4, pp. 37-48.
    Esposito, F., Malerba ,D., and Semeraro, M. (1997), “a comparative analysis of methods for pruning decision trees”, IEEE Tractions on pattern analysis and machine intelligence, Vol.19, No.5, May, pp.476-491.
    Fan, C., Guo, R., Chang, S., Wei, C. (2000), “SHEWMA: an end-of-line SPC scheme using wafer acceptance test data”, IEEE Transactions on Semiconductor manufacturing, Vol.13, No.3, August 2000, pp.344-358.
    Fayyad, U. (1997), “Data mining and knowledge discovery in database: implication for scientific database”, Proceedings of Ninth International Workshop on Scientific and Statistical Database Management, pp.2-11.
    Fayyad, U., Piatetsky-Shapiro, G., and Smyth, P. (1996), “ The KDD Process for Extracting Useful Knowledge from Volumes of Data ”, Communication of ACM, vol. 39, no.11, pp.27-34.
    Feelders, A., Daniels, H., and Holsheimer, M. (2000), “Methodological and practical aspects of data mining”, Information & Management 37 pp.271-281.
    Fu, Y. (1997), “Data mining”, IEEE Potentials, vol. 164, pp.18-20.
    Han. J. and M. Kamber (2001), Data Mining: Concepts and Techniques, Morgan Kaufmann Publishers.
    Kass, G. V. (1980), “An exploratory technique for investigating large quantities of categorical data”, Applied Statistics, 29, 2, pp.119-127.
    Montgomery, D. C. (1997), Design and Analysis of Experiment, John Wiley & Sons, INC.
    Nanopoulos A., Y. Manolopoulos (2001), “Mining pattern from graph traversals”, Data & Knowledge Engineering 37, pp.243-266.
    Lim, T. S., Loh, W. Y., and Shih, Y. S. (2000), “A Comparison of Prediction Accuracy, Complexity, and Training Time of Thirty-three Old and New Classification Algorithms”, Machine Learning, Vol.40, pp.203-229.
    Loh W.-Y (2002), “Regression tree with unbiased variable selection and interaction detection”, Statistica Sinica, vol. 12, pp.361-386.Quinlan, J. R. (1993a), C4.5: Programs for Machine Learning. Morgan Kaufmann, San Francisco, California.
    Quinlan J. R. (1993b), “Combining instance-based and model-based learning”, Proc. ML'93 (ed P.E. Utgoff), San Mateo: Morgan Kaufmann.
    Shortland, R. and R. Scarfe (1995), “Digging for gold”, IEE Review 41 (5), pp.213-217.
    Zant, P. V. (1997), Microchip Fabrication, McGraw-Hill.
    Zhou, C., Nelson, P. C., Xiao, W., Tripak, T. M., and Lane, S. A. (2001), “An intelligent data mining system for drop test analysis of electronic products”, IEEE Transactions in electronics packaging manufacturing, Vol. 24, No.3, July, pp.222-231

    無法下載圖示 全文公開日期 本全文未授權公開 (校內網路)

    全文公開日期 本全文未授權公開 (國家圖書館：臺灣博碩士論文系統)
    QR CODE