跳到主要內容

簡易檢索 / 詳目顯示

回結果列表
研究生： 林宏軒
Hung-Hsuan Lin
論文名稱： 利用馬可夫決策過程製定狀態檢修策略
Purposing a Condition Based Maintenance Policy Using Markov Decision Process
指導教授： 曾富祥
Fu-Shiang Tseng
口試委員:
學位類別： 碩士
Master
系所名稱： 管理學院 - 工業管理研究所
Graduate Institute of Industrial Management
論文出版年： 2023
畢業學年度： 111
語文別： 英文
論文頁數： 83
中文關鍵詞： 狀態檢修 、Cox 比例風險模型 、馬可夫決策過程
外文關鍵詞： Condition Based Maintenance, Cox Proportional Hazard Model, Markov Decision Process
相關次數： 點閱：22  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

    • 在現今自動化、大量生產的製造業中，機台維修的議題持續受到人們的關注。維修策略的制定對於管理者來說相當困難，關於機台的信息充滿著不確定性，如運作的時間長短、工作條件的設定、機台效能的衰退，管理者需要在有限資訊的情況下進行決策。在機台維修的研究中，經常利用馬可夫決策過程為模型探討維修時間的規劃，在固定的時間周期進行維修動作的決策，以動態規劃求解得到長期最佳的維修時間計畫，並稱之為最佳決策。我們希望在馬可夫決策過程中考量機台因為工作條件、環境因素不同而有不同的衰退速率或轉移機率。我們希望管理者可以根據機台的資訊即時的估計衰退速率、轉移矩陣，有效的衡量機台所需承擔的風險，可以制定更好的維修計畫使的總成本降低。

    在此研究中我們假設機台的狀態有兩種，而狀態的衰退速率會依據當前的工作條件有所差異。我們希望利用 Cox 比例風險模型透過機台的工作條件、環境資訊估計各個狀態的失效率，並利用其結果估計馬可夫決策過程中的轉移機率。藉由機台資料即時的更新調整維修計畫，降低維修、營運的成本。

    In today’s automated and mass-produced manufacturing industry, the issue of machine maintenance continues to attract people’s attention. It is very difficult for managers to formulate maintenance strategies. The information about machines is diversity. In the research
    of machine maintenance, the Markov Decision Process is often used as a model to discuss the planning of maintenance time, the decision-making of maintenance actions is made in a
    fixed time period, and the long-term optimal maintenance solution is obtained by dynamic programming. We want that managers can estimate the deterioration rate and transition
    probability in real time based on the information of the machine.

    We want to use the Cox proportional hazard model to estimate the failure rate of the machine through the working conditions and environmental information. Using the results
    to estimate the transition probability in the Markov Decision Process, and determine the maintenance policy. By updating information to adjust the maintenance plan in real time, the cost of maintenance and operation can be reduced.

    Contents 中文摘要 i Abstract ii Contents iv List of Tables vi List of Figures vii Chapter 1 Introduction 1 1.1 Background and Motivation 1 1.2 Research Objectives 3 1.3 Research Framework 3 Chapter 2 Literature Review 5 2.1 Condition Based Maintenance 5 2.2 Remaining Useful Life 6 2.2.1 Machine Learning Methods for Remaining Useful Life 7 2.2.2 Statistical Methods for Remaining Useful Life 8 2.3 Maintenance Policy 11 Chapter 3 Methodology 14 3.1 Cox Proportional Hazard Model and Markov Decision Process 14 3.1.1 Introduction of the Cox Proportional Hazard Model 14 3.1.2 Introduction of the Markov Decision Processes 17 3.1.3 Markov Decision Process in Maintenance Problem 20 3.2 Condition Based Maintenance Using Markov Decision Process 25 3.2.1 The Framework of Condition Based Maintenance Policy 27 3.2.2 Condition Based Transition Probability 32 Chapter 4 Numerical Study 36 4.1 Introduction of the Datasets 36 4.2 Condition Based Maintenance Policy of the Datasets 39 4.3 Condition Effects on Maintenance Policy 52 4.4 Cost Effects on Maintenance Policy 57 Chapter 5 Conclusion and Summary 60 Appendix 65 Reference 70

    1. Abeygunawardane, S. K., Jirutitijaroen, P., and Xu, H. Adaptive maintenance policies for aging devices using a markov decision process. IEEE Transactions on Power Systems 28, 3 (2013), 3194–3203.
    2. Ahmad, R., and Kamaruddin, S. An overview of time-based and condition-based maintenance in industrial application. Computers & industrial engineering 63, 1 (2012), 135–149.
    3. Carroll, J., Koukoura, S., McDonald, A., Charalambous, A., Weiss, S., and McArthur, S.Wind turbine gearbox failure and remaining useful life prediction using machine learning techniques. Wind Energy 22, 3 (2019), 360–375.
    4. Crowley, J., and Hu, M. Covariance analysis of heart transplant survival data. Journal of the American Statistical Association 72, 357 (1977), 27–36.
    5. Fengfei, W., Shengjin, T., Xiaoyan, S., Liang, L., Chuanqiang, Y., and Xiaosheng, S. Remaining useful life prediction based on nonlinear random coefficient regression model with fusing failure time data. Journal of Systems Engineering and Electronics (2023).
    6. Gebraeel, N., and Pan, J. Prognostic degradation models for computing and updating residual life distributions in a time-varying environment. IEEE Transactions on Reliability 57, 4 (2008), 539–550.
    7. Howard, R. A. Dynamic programming and markov processes.
    8. Jorgenson, D. W., McCall, J. J., and Radner, R. Optimal maintenance of stochastically failing equipment. Tech. rep., RAND CORP SANTA MONICA CA, 1966.
    9. Kharoufeh, J. P., and Cox, S. M. Stochastic models for degradation-based reliability. IIE Transactions 37, 6 (2005), 533–542.
    10. Li, W., and Liu, T. Time varying and condition adaptive hidden markov model for tool wear state estimation and remaining useful life prediction in micro-milling. Mechanical Systems and Signal Processing 131 (2019), 689–702.
    11 Ling, M. H., Tsui, K. L., and Balakrishnan, N. Accelerated degradation analysis for the quality of a system based on the gamma process. IEEE Transactions on Reliability 64, 1 (2014), 463–472.
    12. Nourelfath, M., Nahas, N., and Ben-Daya, M. Integrated preventive maintenance and production decisions for imperfect processes. Reliability engineering & system safety 148 (2016), 21–31.
    13. Rodrigues, L. R. Remaining useful life prediction for multiple-component systems based on a system-level performance indicator. IEEE/ASME Transactions on Mechatronics 23, 1 (2017), 141–150.
    14. Si, X.-S., Wang, W., Hu, C.-H., and Zhou, D.-H. Remaining useful life estimation–a review on the statistical data driven approaches. European journal of operational research 213, 1 (2011), 1–14.
    15. Thijssens, O., and Verhagen, W. J. Application of extended cox regression model to time-on-wing data of aircraft repairables. Reliability Engineering & System Safety 204 (2020), 107136.
    16. Wang, S., Jin, S., Bai, D., Fan, Y., Shi, H., and Fernandez, C. A critical review of improved deep learning methods for the remaining useful life prediction of lithium-ion batteries. Energy Reports 7 (2021), 5562–5574.
    17. Yan, M., Wang, X., Wang, B., Chang, M., and Muhammad, I. Bearing remaining useful life prediction using support vector machine and hybrid degradation tracking model. ISA transactions 98 (2020), 471–482.
    18. Yang, L., Ma, X., and Zhao, Y. A condition-based maintenance model for a three-state system subject to degradation and environmental shocks. Computers & Industrial Engineering 105 (2017), 210–222.
    19. Zhang, H., Mo, Z., Wang, J., and Miao, Q. Nonlinear-drifted fractional brownian motion with multiple hidden state variables for remaining useful life prediction of lithium-ion batteries. IEEE Transactions on Reliability 69, 2 (2019), 768–780.
    20. Zhang, H., Zhou, D., Chen, M., and Xi, X. Predicting remaining useful life based on a generalized degradation with fractional brownian motion. Mechanical Systems and Signal Processing 115 (2019), 736–752.
    21. Zhang, M., Gaudoin, O., and Xie, M. Degradation-based maintenance decision using stochastic filtering for systems under imperfect maintenance. European Journal of Operational Research 245, 2 (2015), 531–541.
    22. Zhang, X., and Gao, H. Road maintenance optimization through a discrete-time semimarkov decision process. Reliability Engineering & System Safety 103 (2012), 110–119.
    23. Zhou, P., and Yin, P. An opportunistic condition-based maintenance strategy for offshore wind farm based on predictive analytics. Renewable and Sustainable Energy Reviews 109 (2019), 1–9. 71
    24. Zhu, Q., Peng, H., and van Houtum, G.-J. A condition-based maintenance policy for multi-component systems with a high maintenance setup cost. OR Spectrum 37 (2015), 1007–1035

    QR CODE
    :::