摘要
科技發展來自於生活的需求，在大眾生活中各類電子商品，如手機、電視、平板電腦，已經不可獲缺。作為各類電子產品的重要組成，如顯示器朝向多色階、高速、快速反應與低耗能發展，產品生命週期越來越短、價格也越來越低。面對後端IC測試代工費用也逐年調降的趨勢，測試代工廠如何提昇效能與掌控毛利率以應對產業發展趨勢成為一個新興重要的議題。
本研究針對晶圓測試之探針卡主要耗材取得問題，基於紫式決策分析架構，採用影響圖與決策樹方法，建構探針卡取得之分析架構，並繪製風險輪擴圖與敏感度分析圖，從而能夠有效提供探針卡耗材取得模式選擇方案，以供決策者實務中進行決策。
為實際檢驗研究架構之效度，本研究以台灣某半導體晶圓測試廠為案例。藉由取得來源與特定客戶型號的所有報廢探針卡，依據決策層級架構分析方法，建立模型進行架構分析，作為決策者參考依據。同時依據敏感度分析圖之成功率，決策者可取得未來同型號探針卡生產平均效率之決策預估值。實證結果顯示，本研究具備效度與可行性。其分析手法與方式，亦將可提供其他探討有關物料取得之決策問題，作為參考流程與架構的依據。

Abstract
Technology developments come from the real life requirements. Various types of electronic goods, such as mobile phones, televisions, tablet PCs, are indispensable in the public life recently. As an important part of electronic products, the display develops toward multi-color, high-speed, rapid response and low energy consumption trend. Product life cycle in display industry becomes shorter and shorter while sales price getting lower and lower. Facing costs decreasing year by year in trendency, how to improve the efficiency of the foundry and control gross margin to cope with industrial development trends has been a new critical issue for back-end IC test foundries.
In this paper, to solve the main consumables acqusition problem of the probe card for wafer testing, a probe card acqusition decision-making framework is constructed based on the UNISON decision-making framework by using the influence graph and the decision tree method. Besides, risk profile and sensitivity analysis are used to obtain the mode of choice for probe card supplies to make decision in practical problem.
To test the validity of the research structure, this study takes a Taiwan semiconductor wafer test company as a case. All the discarded probe cards are obtained from the foundry source and the specific customer model. Based on the decision hierarchy analysis method, the proposed model is established to analyze probe card acquisition decision problem and serves as a reference for decision makers. Meanwhile, refer to success rate of the sensitivity of graph analysis, decision maker can aquire the implementation of future decision with average efficiency of the probe card production decision-making estimates. The empirical results show that this study is effective and feasible. Its analytical practices and methods can also be a basis of reference flow and architecture.provide for other decision-making issues related to material acquisition.

參考文獻
Bersn,A.,Smith,S.,and Thearling,K.,Building Data Mining Applications for CRM,McGraw-Hill(2000).
Berry,M.,and Linoff,G.,Data Mining Techniques for Marketing,Sales and Customer Support,John Wiley and Sons,New York(1997).
Borison, Adam, “Real Options Analysis: Where Are the Emperor's Clothes?” Journal of Applied Corporate Finance, vol. 17, no. 2, pp. 17-31, March 2005.
Breimen,L.,Friedman,J.H.,Olshen,R.J.,and Stone,C.J.,Classification and Regression Trees,Belmont,CA:Wadsworth(1984).
Chen-Fu Chien1,Hung-Ya Huang, Kuo-Hao Chang,Yu-Cheng,Tien-Hsian,Joint Symposium 2011 E-manufacturing & Design Collaboration and ISSM (2011).
Chen-Fu Chien,Yu-Shian Chiang,Chih-Chiang Chen,Real Option Analysis for Capacity Investment Planning for Semiconductor Manufacturing(2007).
Chen-Fu Chien and Ying-Jen Chen, Manufacturing Intelligence and Smart Production for Industry 3.5 and Empirical Study of Decision-based Virtual Metrology for Controlling Overlay Errors (2016 )
Chen-Fu Chien,Jei-Zheng Wu Analyzing Repair Decisions in the Site Imbalance Problem of Semiconductor Test Machines (2003).
Chen-Fu Chien, Chih-Wei Hsiao1, Cheng Meng, Kuo-Tong Hong, Szu-Tsung Wang Cycle Time Prediction and Control Based on Production Line Status and Manufacturing Data Mining (2005).
Chih-Chiang Chen,Yu-Shian Chiang,Chen-Fu Chien Real Option Analysis for Capacity Investment Planning for Semiconductor Manufacturing ISSM Paper: MS-P-126 (2007)
Chen-Fu Chien1, Hung-Ya Huang1,Kuo-Hao Chang1,Yu-Cheng Lin2, Tien-Hsian Liu2 Semiconductor Manufacturing Intelligence and Key Factor Control Mechanism for Managing Production Cycle Time (2011).
C.F. Chien, A. Hsiao and E. Wang, “Constructing semiconductor manufacturing performance indexs and applying data mining for manufacturing data analysis,”Journal of Chinese Institute of Industrial Engineers, vol.21, pp. 313-327, 2004.
C.F. Chien, Y.C. Shih, C.M. Lin and J. Chen,Semiconductor Manufacturing Technology &Management, Tsing Hua University Press, 2005.
C. Peng and C.F. Chien, “Data Value Development to Enhance Yield and Maintain Competitive Advantage for Semiconductor Manufacturing”, International Journal of Service Technology and Management, vol. 4, no. 4-6, pp. 365-383, 2003.
C.-F. Chien, C. Hsu, H. Chou and C. Lin, “Overall Wafer Effectiveness (OWE): A Novel Industry Standard for Wafer Productivity,” Proceedings of International Symposium of Semiconductor Manufacturing, pp. 317-320, 2006.
Chien, C. F., Chen, Y., and Peng, J., 2010, “Manufacturing Intelligence for Semiconductor Demand Forecast based on Technology Diffusion and Product Life Cycle,”International Journal of Production Economics, Vol. 128, No. 2, 496-509.
"Defining Failure:What Is MTTR,MTTF,and MTBF?"Stephen Foskett,Pack Rat. Retrieved(2016).
Galligan,D.T.,Marsh,W.E.,and Madison,J.(1987),”Economic Decision Making in Veterinary Practice:Expected Value and Risk as Dual Utility Scales,”Preventive Veterinary Medicine,Vol.5,No.2,79-86.
Keeney,Raiffa(1981),”Analysis Of Preference Dependencies Among Objectives”, Operations Research,Vol. 29,pp. 1105-1120.
Keeney, R. L. and Raiffa, H., 1993, Decisions with Multiple Objectives: Preferences and Value Tradeoffs, New York: Cambridge University Press.
Kuo, C., Chien, C. F., and Chen, C., 2011, “Manufacturing Intelligence to Exploit the Valueof Production and Tool Data to Reduce Cycle Time,” IEEE Transactions onAutomation Science and Engineering, Vol. 8, No. 1, 103-111.
Markowitz,H(1952), ”Portfolio Selection,”Finance,Vol.7,No.1,77-91.
M. Gardner and J. Bieker,“Data mining solves tough semiconductor manufacturing problems, ”presented at the Conference on Knowledge Discovery in Data, Boston, United States, 2000.
Nakajima,S. (1988) , Introduction to Total Productive Maintenance ,Productivity Press : Cambridge, MA.
Platts, K.W., Probert, D.R. and Canez L.(2002), “Make vs. buy decisions: A process incorporating multi-attribute decision-making”, International Journal of Production Economics 77.
Quinlan,J.R.(1983),”Learning Efficient Classification Procedures and Their Application to Chess End Games,”In Michalski, R.S., Carbonell, J.G.,Mitchell, T.M.(Eds.),Machine Learning : An Artificial Intelligence Approach,Morgan Kaufmann,Los Altos.
Quinlan,J.R.,”Induction of decision tree,”Machine Learning(1986),pp.81-106.
Shin-Chung Chen,Shan-Ming Lan,The Effect of Probing Contact on Chip Test(2011).
S. Bahadur,S. Mohanraj,and D. Tirupati, “Test head scheduling in a semiconductor test facility,” in Proc. IEEE/CPMT Int. Electronics Manufacturing Technology Symp., 1998, pp. 131–135.
Shih, W., Chien, C. F., Shih, C., and Chang, J., 2009, The TSMC Way: Meeting
Customer Needs at Taiwan Semiconductor Manufacturing Co., Boston: Harvard Business School Publishing.
T. Chen, T. Chang, C. Chen, and J. Kao, “Scheduling for IC sort and test with preemptiveness via Lagrangian relaxation,” IEEE Trans. Syst.,Man, Cybern., vol. 25, pp. 1249–1256, Aug. 1995.
Tom Copeland and Vladimir Antikarov, “Real Options: A Practitioner’s Guide,” TEXERE, New York, NY, 2001.
簡禎富、許嘉裕 (2014)，資料挖礦大數據分析，前程文化事業有限公司，頁140-177。
簡禎富（2013），決策分析與管理，雙葉書廊，台北。
簡禎富、胡志翰（2011），全面資源管理架構－以晶圓廠為實證，品質學報，18卷5期，頁1-26。
胡志翰(2011)，旺宏電子公司-製作策略決策，ISSN 1023-2842 中山管理評論第十九卷第一期,P1-27。
張期閔(2011)發展設計多層探針卡之能量法解析謨型與探針卡優化分析
許嘉裕（2009），半導體可製造性設計之資料挖礦架構及其實證研究，國立清華大學工業工程與工程管理學系博士論文。
簡禎富、游智閔、徐紹鐘（2009），紫式決策分析以建構半導體晶圓廠人力規劃決策模型，管理與系統，16卷2期，頁157-180。
李駿恆(2006)，藍芽Dongle外包或自製決策個案探討，清華大學科技管理學院高階主管經營管理碩士專班論文。
賴彥中(2005)，發展主幹式決策樹法則以提昇半導體良率之研究-以DRAM廠為實證，國立清華大學工業工程與工程管理學系碩士論文。
張木榮(2005)，無線監控產品自製或外包決策之個案研究，清華大學清華大學科技管理學院高階主管經營管理碩士專班論文。