半導體晶圓測試廠設備機台非常昂貴，每台費用大致從數千萬到數億元之間不等。設備充份之利用對測試產業將是一非常重要的課題。每當晶圓測試廠發生機台當機、預防保養與更換事件時，將會造成維修費用的增加，設備的閒置使用率降低，機台壽命減短及產品不良損失等情況，進而造成對客戶交貨設定時間延遲之商譽影響。

整個晶圓測試廠，針對預防維護作業中較無探討分別運用TBM ( Time Based Maintenance) 與CBM(Condition Based Maintenance)策略下對整合生產與維護系統之效益之影響為何。TBM與CBM兩種維護策略下所造成每次維護的時間點皆不同，將對測試產品製造過程的時間點也將會不同，進而影響到各產品的生產週期時間及訂單的達交率。同時在整合生產與維護中也未探討到TBM、CBM方式、機台維護優先順序及維護人力配置情況下之維護組合策略下對系統績效影響，也是先前研究者較無探討的部份。因此本研究將考量TBM與CBM維護方式、維護派工及維護人員配置等三種組合變數下來建構晶圓測試廠生產與維護績效評估系統，可使公司生產與維護目標盡量強化，並增加該企業之競爭優勢。

研究結果顯示，本研究運用TBM與 CBM方法於生產與維護管理績效系統的模式，可有效模擬企業生產與維護安排之情況，預先瞭解各種維護組合策略下之較佳生產與維護之績效值，並可提供管理者事先安排生產與維護工作之參考依據。
例如針對維護人員平均使用效率中以目前該公司之狀況應選擇第四組組合策略(灰色系統,SMT,全能工)為68.56％為最佳。而本研究運用系統模擬的方法將可得知設備機台維護之標準人力應為6人與案例公司目前之人力為8人將有2人之差距，因此若運用本研究之方法將可節省該公司每年人力為2人之費用為100.8萬元(3.6萬＊14(月)＊2人=100.8萬元)。

Testing tools used by semiconductor foundries are very expensive, the cost in each ranges from tens of millions to hundreds of millions. Optimal use of the equipment for the test industry will be a very important topic. When testing tools used for testing in semiconductor foundries crash, or when preventive maintenance and parts replacements are required, the maintenance cost redundant, lowering of the machine usage rates, reduced lifespan, and losses due to product defects may occur as a result. This may further delay shipment of the product to the customer and have productive effects on the business reputation.

There has not been much in-depth discussion of preventive maintenance in Ic Testing house , especially the performance of Time Based Maintenance (TBM) and Condition Based Maintenance (CBM) strategies on the production and maintenance systems. The difference between TBM and CBM strategies causes different maintenance times, resulting in different product manufacturing overall cycle times; further affecting the product production lifecycles and order fulfillment rates. Also, TBM and CBM methods, order in which machine tools are maintained, human resource distribution, and other maintenance strategy issues are not investigated during integrated product manufacturing and maintenance for effects in system performance; it is also a topic not fully investigated.

Therefore, this research evaluated the TBM and CBM methods, maintenance work displacement, and assignment of maintenance personnel. With these three factors, a semiconductor foundry production and maintenance effectiveness performance system is established to enable strengthening of a company’s production and maintenance goals while increasing the competitiveness of the business.

For example, based on the average usage rates by maintenance personnel, the company should select the fourth strategic combination (grey system, SMT, fully rounded personnel) with 68.56% as the must optimal selection.

This research applies the system simulation method to learn that the standard human resource for machine tools should be 6 personnel, 2 personnel less than the 8 personnel used by the company used by this case study. So by apply this case study method, 2 personnel can be reduced, resulting in a savings of 1.008 million dollars (36 thousand * 14 (months) * 2 personnel = 1.008 million dollars).

參考文獻
中文部份：
[1]江益璋(1999), “設備診斷與技術實用手冊”, 中衛發展中心.
[2]鄭達才(2000), “設備維護管理:現在與未來”, 中國生產力.
[3]蔣瑞祥(2000),“退化製造系統維修策略之研究”, 國立清華大學工業工程與工程管理學系研究
所博士論文.
[4]王清平(2000), “退化製造系統下多種動作的動態維護策略之研究”, 國立清華大學工業工程與
工程管理學系研究所碩士論文.
[5]聶錦楨(2004), “異質馬可夫式衰退系統之多動作動態預防維護策略研究”,國立清華大學工
業工程與工程管理學系研究所碩士論文.
[6]沈育樹(2002), “維護管理系統設計與模式分析”, 中原大學工業工程研究所碩士論文.
[7]顧尚芳(2003), “生產系統中利用製程不良率評估設備預防維護之研究”, 中原大學工業工程研
究所碩士論文.
[8]莊木貴(1998) , “最佳置換策略之貝氏分析與圖解法”,國立台灣科技大學碩士論文.
[9]施育仁(1999), “半導體後段製程生產設備維修決策支援系統之研究”, 國立台北科技大學生產
系統工程與管理研究所碩士論文.
[10]陳穆臻(2004),“維修備用零件之預防性替換與比例政策存貨管理”,第一屆台灣作業研究學會
學術研討會,台北,國立台北科技大學.
[11]蔡燕純(2004), “應用馬可夫預防維護模型於維修保養策略之探討”, 中央大學機系研究碩碩士
論文.
[12]江錦宗(2002) , “應用馬可夫決策過程進行台灣股票投資分析之研究”,東海大學工業工程與資
訊研究碩碩士論文.
[13]楊明松(2003) ,“人民幣匯率預測之研究－灰馬可夫鏈應用”, 大葉大學經營研究碩碩士論文.
[14]洪穎怡(2002), “應用模糊馬可夫決策發展自由化電業市場多狀態競價策略”, 中原大學電機工
程研究碩碩士論文.
[15]顧志遠(2001), “服務復原策略效益評估模式建立之研究-以馬可夫鏈為分析工具”,東海大學工
業工程與資訊研究碩碩士論文.
[16]林則孟(2002), “系統模擬理論與實務”,滄海書局.
[17]陳耀茂(1998) , “可靠性方法與應用”,雙葉書廊.
[18]張偉哲(2000) , “灰關聯模型方法與應用”,高立書局.
[19]高孔廉(1969) , “馬可夫鏈鎖之理論與應用”,國立政治大學企業管理研究所碩士論文.
[20]溫坤禮(2006) , “.MATLAB在灰色系統理論之應用”，全華科技圖書股份有限公司.

英文部份
[1] M.E. pate-cornell, H. L. LEE (1987), ”Warnings of Malfunction : The decision to inspect and
maintain production processes on schedule or on demand”, Management science , 33, pp-1277-1290.
[2] A. Chen, R.S. Guo, G.S. Wu (2000), ”Real-time Equipment Health Evaluation and Dynamic
Preventive Maintenance”, The Ninth International Symposium on Semiconductor Manufacturing,
pp-375-378.
[3] A.C. Marquez, A.S. Heguedas (2002), “Models for Maintenance Optimization: a Study for
Repairable Systems and Finite Time Periods”, Reliability Engineering and System Safety, 75,
pp-367-377.
[4] D. Chena, K. S. Trivedib (2005), “Optimization for condition-based maintenance
with semi-Markov decision process”, Reliability Engineering and System Safety 90 pp-25–29
[5] Tsai, Y. T. , Wang, K. S. and Teng, H. Y(2001)., “Optimizing Preventive Maintenance for
Mechanical Components Using Genetic Algorithms”, Reliability Engineering and System Safety 74,
pp-89-97.
[6]C. V. Flores, R. M. Feldman(1989), “A survey of preventive maintenance models for stochastically
deteriorating single unit systems” , Naval. Res. Logist. Quart., 36, pp-419–446.
[7]Love, C.E. and R. Guo(1991), “Using proportional hazard modelling in plant maintenance” , Quality
And Reliability Engineering international, 7, pp- 7-17.
[8] D. I. Cho and M. Parlar(1991), “A survey of maintenance models for multi-unit systems” , Eur. J.
Oper. Res, 51, pp-1–23.
[9]Jette Paulsen, R. Cooke and R.Nyman(1997), “Comparative Evaluation of Maintenance
Performance Using Subsurvival Functions” , Reliability Engineering and System Safety, 58,
pp-157-163.
[10] M. Ben-Daya, A.S. Alghamdi(2000) , “On an Imperfect Preventive Maintenance Model,”,
International Journal of Quality & Reliability Management, 17,pp-661-670.
[11] Li. C.Hong , C. S.Chao(2002), “ A Preventive Maintenance Forecast Method for Interval
Triggers” ,The Journal of IEEE Transactions on semiconductor Manufacturing, pp-275-277.
[12] A. Banerjee, S. Jonathan, and J. S. Burton(1990), “Equipment utilization based maintenance task
scheduling in a job shop,” Eur. J. Oper. Res., 45, pp-191–202.
[13] S. O. Duffuaa and A. Raouf(1992), “A simulation model for determining maintenance staffing in
an industrial environment,” Simulation, 59,pp-93–99.
[14] Sarah A. Mosley, Tim Teyner, and Reha M. Uzsoy(1998), “Maintenance Scheduling and Staffing
Policies in a Wafer Fabrication Facility” , IEEE Transactions on Semiconductor Manufacturing,11,
pp-84-97.
[15] C. Marcelo, F. Lapaa(2006),.” A model for preventive maintenance planning by genetic algorithms
based in cost and reliability” ,Reliability Engineering and Safty 91,pp233-240.
[16] Celestine A. Ntuen(1991),” An Economic Preventive Maintenance Scheduling Model with
Truncated Gamma Function”, Reliability Engineering and Safty 31,pp-31-38.
[17] S. Arueti & D. Okrent (1990),” A Knowledge-Based Prototype for Optimization ofPreventive
Maintenance Scheduling”, Reliability Engineering and Safty 30,pp-93-114.
[18]A.Barros, C. Berenguer, A. Grall (2005), “On the hazard rate process for imperfectly monitored
multi-unit systems”, Reliability Engineering and Safty 90, pp-169-176.
[19] L.F. Hsu(1999), ”Simultaneous determination of preventive maintenance and replacement
policies in a queue-like production system with minimal repair”, Reliability Engineering and Safty
63, pp-161-167.
[20] R.I. Zequeira, C. B (2006), “Periodic imperfect preventive maintenance with two categories of
competing failure modes” , Reliability Engineering and Safty 91,pp-460—468.
[21] S. H. Sheu, Y.B. Lin, G.L. Liao (2006),” Optimum policies for a system with general imperfect
maintenance”, Reliability Engineering and Safty 91,pp-362—369.
[22] S. Wu, D. C. Croome (2005),” Preventive maintenance models with randommaintenance quality”,
Reliability Engineering and Safty 90,pp-99-105.
[23] M. Samrouta, F. Yalaouia, E. Chateleta, N. Chebbob (2005),” New methods to minimize the
preventive maintenance cost of series–parallel systems using ant colony optimization” , Reliability
Engineering and Safty 89,pp-346-354.
[24]B.D. Mohamed, M. A. Rahiam (2000) ,” Effect of maintenance on the ecomic design of control
chart ” ,European Journal of operational research, 76,pp-131-143.
[25]Ansi. C, Daoud, A.K (2004), “Analysis of a production/ inventory system with randomly failing
production unit submitted to regular prevent maintenance”, European Journal of operational
research,156,pp-712-718.
[26] J. L. Deng (1982),” Introduction to Grey Systems Theory ” ,The Journal of Grey Systems,
1,pp-1-24.
[27] J. L. Deng (1982), The Application of Grey System Theory. 1st ed. Taiwan: Gao
Li; 1999. p. 55-61.
[28] K. H. Cheng, W. C. Chang (2001), “A Grey Mathematical Model for Earthquake Forecasting”,
pp-131-142.
[29] Y.T. Hsu, C. M. Chen (2001), “Grey-based ART System for classifications”, pp-1-16.
[30]D. Daniel Sheu and Jun Yuan Kuo (2006),. “A Model for Preventive Maintenance Operation and
Forecasting. Journal of Intelligent Manufacturing, 14 pp-441-451.
[31] Y.C. Hu, G. H. Tzeng (2001), ” Using Learning Algorithm to Find the Developing Coefficient and
ControlVariable of GM(1,1) Model”,pp-17-26,.
[32] H. C. C (2001), “A Low Order Polynomial Updating Scheme to Improve the NLMS Algorithm,”
pp.27-34.
[33] B.R. Chang (2001), “Applying Grey Relational Analysis and Regression Analysis to Construct
Multivariate Indexes of Affective Instruction Evaluation”, pp-35-48.
[34] C.H. Wu (2001), “Data Representation on the Grey GM(1,1) Model for Predicting Performance
with theConstruction Learning Curve”, pp-49-58.
[35] J.Shen (2001), “On the Condition of White Exponential Law”, pp.59-64.
[36] Joseph D.Patton (2004), “Preventive Maintenance ”, The Instrumentation Society and Automation
Society.
[37] Nyman, Levitt (2002), “ Maintenance Planning, Schedule and coordination ”, Industrial Press Inc.