本論文旨在應用貝氏分析法和總測試時間圖解法於系統維護策略上，探討最佳之維護策略，在此條件下，本論文包括以下三項研究主題：
1. 使用貝氏分析法，推導出具有小修理年齡置換策略之最佳解．首先藉由小修理、有計劃性與非計劃性置換等策略，推導出長期與置換週期兩種情況之每單位時間期望成本數學公式，並證明在某些合理的條件下，兩種情況 皆具有最佳置換時間唯一且有限解。當失效密度函數具有不確定參數之韋伯分配時，建立貝氏分析表示法及修 正前述之不確定參數，即可導出最佳年齡置換策略。
2. 應用貝氏分析法，推導出調適性預防維護策略之最佳解。結合小修理，維護及置換等策略，推導出每單位時間 期望成本公式，並在某些合理的條件下，証明最佳維護時間與最佳置換時間有唯一及有限解。當失效密度函數 具有不確定參數之韋伯分配時，建立貝氏分析公式，並修正前述之不確定參數，即可導出最佳之維護時間及置 換時間．
3. 另外一種方法稱為總測試時間(TTT)圖解法，它的觀念已被証實為在許多可靠度之應用上是非常有用的一種工具 。因為此種方法，系統的失效分配函數可以不需要特別的指定，僅需要系統（或產品）在測試階段所獲得的完 整樣本資料即可，且求出的最佳置換策略之結果接近於理論上之最佳值。本文係利用TTT圖解法,並藉由小修理 ，有計劃性與非計劃性置換等策略，推導出有衝擊型式的k-out-of-n 系統置換策略之最佳解。

The aim of this thesis is to apply the Bayesian and Total Time on Test (TTT) graphic approach to system maintenance policy, which minimize the expected cost per unit time. The three topics of the thesis may be summarized as follows:
1. We present a Bayesian decision theoretic approach for determining optimal age replacement policy with minimal repair. By introducing costs due to minimal repairs, planned replacement and unplanned replacement we derive the expected cost per unit time for the infinite-horizon case and the one replacement cycle case, respectively. We show that, under certain conditions, there exists a finite and unique optimum policy. A Bayesian decision approach enables us to formally incorporate, express and update our uncertainty when determining optimal age replacement policy with minimal repair.
2. A Bayesian approach is developed for determining an optimal adaptive preventive maintenance policy with minimal repair. By incorporating minimal repairs, preventive maintenance, and replacement, the mathematical formulas of the expected cost per unit time are obtained. We show that there exists a unique and finite optimal maintenance interval and the replacement age under some reasonable conditions. When the failure density follows a Weibull with uncertain (scale, shape) parameters, a Bayesian approach is established to formally express and update the uncertain parameters for determining an optimal adaptive preventive maintenance policy.
3. An alternative approach is the total time on test (TTT) concept which has proved to be a very useful tool in many reliability applications. Since this method can depend on only the complete sample data obtained in testing phase for system or product, the distribution function need not be specified subjectively and the resulting estimator of optimal replacement policy will be optimal asymptotically. In this thesis we deal with the problem of incorporating minimal repair, planned replacement, and unplanned replacement into a k-out-of-n system subject to shocks for determining an optimal age replacement.

封面
中文摘要
英文摘要
誌謝
目錄
表索引
圖索引
第一章 導論
1.1 問題背景與研究動機
1.2 相關文獻探討
1.3 研究範圍與限制
1.4 研究步驟與論文內容
1.5 預備知識
第二章 具有小修理年齡置換策略之貝氏分析
2.1 模式之建立與分析
2.2 貝氏方法
2.3 特殊情形
2.4 數值分析
第三章 調適性防維護策略之貝氏分析
3.1 模式之建立與分析
3.2 貝氏方法
3.3 特殊情形
第四章 有衝擊型式的 k-out-of-n 系統置換策略之圖解法
4.1 模式之建立與分析
4.2 總測試時間觀念
4.3 圖解法與統計估計法
第五章 結論與未來研究方向
5.1 結論
5.2 未來研究方向
參考文獻
附錄
作者簡介

[1] Abdel-Hameed, M.S. and Proschan, F.,"Non-stationary shock models", Stochastic Processes and Their Application, Vol.1, pp.383-404 (1973).
[2] Archibald, T.W, and Dekker, R.," Modified block- replacement for multiple -component systems", IEEE Transactions on Reliability, Vol. 45, pp.75-83 (1996).
[3] Bai, D.S., Yun, W.Y., and Chung, S.W., " Redundancy optimization of k-out-of-n systems with common-cause failures", IEEE Transactions on Reliability, Vol. 40, No. 1, pp. 56-59 (1991).
[4] Bain, L.J. and Engelhardt, M., " Statistical Analysis of Reliability and Life-Testing Models: Theory and Methods", 2nd Ed., Marcel Dekker, New York, 1991.
[5] Barlow, R.E. and Campo, R., " Total time on test processes and applications to failure data analysis". Reliability and Fault Tree Analysis, Vo1.SIAM, pp.451-481 (Philadelphia 1975).
[6] Barlow, R.E. and Hunter, L.C., " Optimum preventive maintenance policies", Operations Research, Vol.8, pp.90-100 (1960).
[7] Barlow, R.E. and Proschan, F., Mathematical Theory of Reliability, Wiley, New York, 1965.
[8] Barlow, R.E., Bartholomew, D.J., Bremner, J.M. and Brunk, H.D., Statistical Inference Under Order Restrictions, John Wiley & Sons, New York, 1972.
[9] Bassin, W.M., " A Bayesian optimal overhaul interval model for the Weibull restoration process".American Statistical Association Journal, Vol.68, pp.575-578 (1973).
[10] Beichelt, F., "A general preventive maintenance policy", Mathematics of Operations Research , Vol.7, No. 6, pp.927-932 (1976).
[11] Berg, M. and Cl roux, R., " The block replacement problem with minimal repair and random repair costs", Journal of Statistical Computation & Simulation, Vol. 15, pp.1-7 (1982).
[12] Bergman, B., " Some graphical methods for maintenance planning". Proceedings Annual Reliability and Maintainability Symposium, pp.468-471 (1977).
[13] Bergman, B., " On age replacement and the total time on test concept". Scandinavian Journal of Statistics, Vo1.6, pp.161-168 (1979).
[14] Bergman, B., and Klefsj ,B., "A graphical method applicable to age replacement problems", IEEE Transactions on Reliability, Vo1.31, pp.478-481 (1982).
[15] Bergman, B., and Klefsj ,B., "The total time on test concept and its use in reliability theory", Operations Research, Vo1.32, pp.596-606 (1984).
[16] Bergman, B., and Klefsj ,B., "the TTT-concept and replacements to extend system life", European Journal of Operational Research, Vo1.28, pp.302-307 (1987).
[17] Block, H.W., Borges, W.S., and Savits, T.H., "Age-dependent minimal repair", Journal of Applied Probability, Vol.22, pp.370-385 (1985).
[18] Boland, P.J., " Periodic replacement when minimal repair costs vary with time", Naval Research Logistics Quarterly, Vol.29, pp.541-546 (1982).
[19] Boland, P.J. and Proschan, F., " Periodic replacement with increasing minimal repair costs at failure", Operations Research,
Vol.30, pp.1183-1189 (1982).
[20] Chan, P.K.W. and Downs, T.,"Two criteria for preventive maintenance",IEEE Transactions on Reliability, Vol.R-27, pp.272-273 (1978).
[21] Chikte, S.D. and Deshmukh, S.D., " Preventive maintenance and replacement under additive damage", Naval Research Logistics, Vol. 28, pp.33-46 (1981).
[22] Cl roux, R., Dubuc, S., and Tilquin, C., " The age replacement problem with minimal repair and random repair cost", Operations Research, Vol. 27, pp.1158-1167 (1979).
[23] Cox, D.R., Renewal Theory, Chapman-Hall, London 1962.
[24] Dohi, T., Kaio, N. and Osaki, S.," Solution procedure for a repair limit problem using the TTT concept", IMA Journal of Mathematics Applied in Business and Industry, Vo1.6, pp.101-111 (1995).
[25] Dohi, T., Aoki, T., Kaio, N. and Osaki, S., " Performance evaluation of nonparametric age replacement methods -survey and comparison-", Stochastic Modelling in In-novative Manufacturing, Lecture Notes in Economics and Mathematical Systems, Springer- Verlag, Berlin pp. 166- 187 (1996).
[26] Dohi, T., Matsushima, N., Kaio, N. and Osaki, S., " Non-parametric repair limit replacement policies with imperfect repair", European Journal of Operational Research, Vo1.96, pp.260-273 (1996).
[27] Dohi, T., Kaio, N. and Osaki, S., " The optimal control of production time limit",Proceedings of 1996 Pacific Conference on Manufacturing, Vo1.1, pp.409-416 (1996).
[28] Dohi, T., Kaio, N. and Osaki, S., " Non-parametric approach to power saving strategies for a portable personal computer", Electronics and Communications in Japan, Vo1.80, pp.80-90 (1997).
[29] Esary, J.D., Marshall, A.W. and Proschan, F.," Shock models and wear process", The Annals of Probability, Vol.1, pp.627-649 (1973).
[30] Epstein, L., and Sobel, M.,"Life testing ", American Statistical Association Journal, Vol.48, pp.486-502 (1953).
[31] Feldman, R.M., " Optimal replacement with semi-markov shock models",Journal of Applied Probability, Vol. 13, pp. 108-117 (1976).
[32] Gibbons, D.I. and Vance, L.C., " A simulation study of estimators for the 2-parameter Weibull distribution", IEEE Transactions on Reliability, Vol.30, pp.61-66 (1981).
[33] Glasser, G,J., " The age replacement problem", Technometrics, Vol. 9, pp.83-91 (1967).
[34] Klefsj ,B.,"On aging properties and total time on test transforms",Scandinavian Journal of Statistics, Vo1.9, pp.37-41 (1982).
[35] Koshimae, H., Dohi, T., Kaio, N. and Osaki, S., " Graphical/statistical approach to repair limit replacement problem", Journal of the Operations Research Society of Japan, Vo1.39,pp. 230-246 (1996).
[36] Kulasekera, K.B, and White, W.H., "Estimation of the survival function from censored data: a method based on total time on test ", Communications in Statistics-Simulation & Computation, Vo1.25, pp.189-200 (1996).
[37] Lawless, J. F., Statistical Methods and Methods for Lifetime Data, John Wiley, NY, 1982.
[38] Lee, L., and Lee, S. K.," Some results on inference for the weibull process", Techometrics, Vol.20, pp.41-45 (1978).
[39] Lotka, A.J.," A contribution to the theory of self-renewing aggregates with special reference to industrial replacement", Annals of Mathematical Statistics, Vol. 10, pp.1-25 (1939).
[40] Mann, N.R., " Point and interval estimation procedures for the two-parameter Weibull and extreme-value distribution", Technometrics, Vol.10, pp.231- 253 (1968).
[41] Mazzuchi, T.A. and Soyer, R.," A Bayesian perspective on some replacement strategies", Reliability Engineering and System
Safety, Vol.51, pp.295-303 (1996).
[42] Muth, E., " An optimal decision rule for repair vs. replacement", IEEE Transactions on Reliability, Vol. 26, pp.179-181 (1977).
[43] Nakagawa, T. and Osaki, S., " The discrete Weilbull distribution", IEEE Transactions on Reliability, Vol. 24, pp. 300-301 (1975).
[44] Nakagawa, T., " Replacement problem of a parallel system in random environment", Journal of Applied Probability, Vol. 16, No. 1, pp. 203-205 (1979).
[45] Nakagawa, T., " A summary of imperfect preventive maintenance policies with minimal repair". R.A.I.R.O. Operations Research, Vo1.14, pp.249-255 (1980).
[46] Nakagawa, T., " Optimal number of units for a parallel system", Journal of Applied Probability, Vol. 21, pp. 431-436 (1984).
[47] Nakagawa, T., " Optimization problems in k-out-of-n systems", IEEE Transactions on Reliability, Vol. 34, No. 3, pp. 248-250 (1985).
[48] Nakagawa, T., " Periodic and Sequential preventive maintenance policies", Journal of Applied Probability, Vo1.23, pp.536-542 (1986).
[49] Nakagawa, T., and Kijima, M., " Replacement policies for a cumulative damage model with minimal repair at failure", IEEE Transactions on Reliability, Vol. 38, No. 5, pp. 581-584 (1989).
[50] Nakagawa, T. and Kowada, M., " Analysis of system with minimal repair and its application to replacement policy", European Journal of Operational Research, Vol. 12, pp.176-182 (1983).
[51] Nakagawa, T., " Further results of replacement problem of a parallel system in random environment", Journal of Applied Probability, Vol. 16, pp. 923-926 (1979).
[52] Nguyen, D.G. and Murthy, D.N.P.," Optimal preventive maintenance policies for repairable systems", Operations Research, Vo1.29, pp.1181-1194 (1981).
[53] Padgett, W.J. and McNichols, D.T., " Non-parametric density estimation from censored data". Communications in Statistics-
Theory & Methods, Vol.13, pp.1581-1611 (1984).
[54] Park, K.S.," Optimal continuous-wear limit replacement under periodic inspections", IEEE Transactions on Reliability, Vol. 37, No. 3, pp.97-102 (1988).
[55] Park, K.S., " Optimal wear-limit replacement with wear-dependent failure", IEEE Transactions on Reliability, Vol. 37, No. 3, pp.293-294 (1988).
[56] Park, K.S. and Jun, C.H., "A Bayesian approach to a periodic incomplete preventive maintenance model", Proceeding of the fourth International Conference on Reliability and Quality in Design, pp. 184-188 (1998).
[57] Proschan, F., "Theoretical explanation of observed decreasing failure rate". Technometrics, Vol.5 , pp. 375- 384 (1963).
[58] Rami Meddy, C., " Optimization of k-out-of-n systems subject to common-cause failures with repair provision", Microelectron Reliability, Vol. 33, pp.175-183 (1993).
[59] Ross, S.M., Applied Probability Models with Optimization Applications, Holder-Day, San Francisco, (1970).
[60] Sathe, P.T. and Hancock, W.M.," A Bayesian approach to the scheduling of preventive maintenance", AIIE Transactions,
Vol.5, pp.172-179 (1973).
[61] Savits, T.H., " Some multivariate distributions derived from a non-fatal shock model", Journal of Applied Probability, Vol. 25, pp.383-390 (1988).
[62] Sheu, S. H., " A generalized block replacement policy with minimal repair and general random repair costs for a multi-unit system", Journal of the Operational Research Society, Vol. 42, pp.331-341 (1991).
[63] Sheu, S.H., " Optimal block replacement policies with multiple choice at failure", Journal of Applied Probability, Vol. 29, pp.129-141 (1992).
[64] Sheu, S.H., " A generalized model for determining optimal number of minimal repairs before replacement", European Journal of Operational Research, Vol.69, No.1, pp.38-49 (1993).
[65] Sheu, S.H., " Extended block replacement policy with used item and general random minimal repair cost", European Journal of Operational Research, Vol. 79, pp.405-416 (1994).
[66] Sheu, S.H. and Griffith, W.S., "Multivariate age-dependent imperfect repair", Naval Research Logistics, Vol. 38 pp.839-850 (1991).
[67] Sheu, S.H., and Griffith, W.S., and Nakagawa, T., " Extended optimal replacement model with random minimal repair costs", European Journal of Operational Research, Vol.26, pp. 636-649 (1995).
[68] Sheu, S.H. and Griffith, W.S., " Optimal number of minimal repairs before replacement of a system subject to shocks", Naval Research Logistics, Vol. 43, pp.319-333 (1996).
[69] Sheu, S.H., and Liou, C.T., " A generalized sequential preventive maintenance policy for repairable systems with general random minimal repair costs", International Journal of Systems Science, Vol.26,No.3, pp.681-690 (1995).
[70] Sheu, S.H. and Liou, C.T., " Optimal replacement of a k-out-of- n system subject to shocks". Microelectronics and Reliability,
Vo1.32, pp.649-655 (1992).
[71] Sheu, S.H. and Kuo, C. M., "Optimization problems in k-out-of-n system with minimal repair", Reliability Engineering & System Safety, Vol.44, pp.77-82 (1994).
[72] Sheu, S.H. and Kuo, C. M., " Optimal age replacement policy of a k-out-of-n system with age-dependent minimal repair", RAIRO-Operations Research.Vol.28, pp.85-95 (1994).
[73] Sheu, S.H. and Jhang, J.P., " A generalized group maintenance policy", European Journal of Operational Research, Vol. 96, pp.232-247 (1996).
[74] Sinha, S.K. and Sloan, J.A., " Bayes estimation of the parameters and reliability function of the 3-parameter Weibull distribution", IEEE Transactions on Reliability, Vol. 37, pp.364-369 (1988).
[75] Soland, R.M., " Bayesian analysis of the Weibull process with unknown scale and shape parameters", IEEE Transactions on Reliability, Vol. 18, pp.181-184 (1969).
[76] Tahara, A. and Nishida, T.," Optimum replacement policy for minimal repair model", Journal of the Operations Research Society of Japan, Vol.18, pp.113-124 (1975).
[77] Taylor, H.M., " Optimal replacement under additive damage and other failure models", Naval Research Logistics, Vol. 22, pp.1-18 (1975).
[78] Thoman, D.R., Bain, L.J., and Antle, C.E., " Maximum likelihood estimation, exact confidence intervals for reliability, and tolerance limits in the Weibull distribution". Technometrics , Vol.12, pp. 363-371 (1970).
[79] Tilquin, C. and Cl roux, R., " Periodic replacement with minimal repair at failure and general cost function", Journal of Statistical Computation Simulation, Vol. 4, pp.63-67 (1975).
[80] Venugopal, N., Ramachandra Prasad, S., and Rami Reddy, C., "Space-repair cost optimal series-parallel systems", International Journal of Systems Science, Vol. 21, pp. 1325-1334 (1990).
[81] Venugopal, N., Rami Reddy, C., and Meenakshi Bai, M., " Cost optimal-( ) parallel reliability systems", International
Journal of Systems Science, Vol. 19, pp.1623-1629 (1988).
[82] Venugopal, N., Shaffi Ahamed, S., and Rami Reddy, C., " Optimal repair stage for k-out-of-n systems", Microelectronic
Reliability, Vol. 29, pp.17-19 (1989).
[83] Welker E.L., "Relationship between equipment reliability, preventive maintenance policy and operating costs", ARINC Monograph 7, Aeronautical Radio Inc, Washington, D.C. (1959).
[84] Willson, J.G. and Benmerzouga, A., " Bayesian group replacement policies".Operations Research, Vol.43, pp.471- 476 (1995).
[85] Yasui, K., Nakagawa, T. and Osaki, S., " A summary of optimal replacement polices for a parallel redundant system", Microelectronic Reliability, Vol. 28, pp.635-641 (1988).