隨著科技的進步，車輛發展日益複雜，性能及技術不斷精進，除購買成本高昂外，更增加操作、維護的複雜度，當車輛運作異常時，會造成任務延宕或人員傷亡，其花費之人力、物力、費用相對提高，且會耽誤任務執行是否順遂。因此，運用資料探勘技術，規劃一套最適預防保養維修策略，以提升車輛維護績效。
維修策略因應不同裝備型式或維修成本考量的因素不同，相對採取的預防保養作為也不相同。因此，運用資料探勘技術，將歷年維修資料分析出精進維修策略之具體做法，並使維修金額有更客觀的分級制度，在各系統預防保養作業下，找出最重要之預防保養項目及重點，以期達到總成本最小化之目標，並且降低事故發生機率。
本研究所提出的維修金額分級模式應用於預算編列上，提升匡列預算額度的準確度，降低維修總成本，並預防事故發生，進而確保戰力的發揮。

With the technological progression, the development of vehicles is getting more elaborate, and the performance and technology continue to be improved. Besides the extremely high cost, it is harder to operate and maintain a vehicle. When there is malfunction happening, it may delay missions or cause deaths or injuries. Then, much more labor, material resources and expenses will lift up the cost; moreover, it can affect the efficiency and effectiveness to do tasks. Thus, we should have data mining techniques applied to set up a series of strategies for prevention, maintenance and restoration in order that the performance of vehicle maintenance will surge.
Based on diverse models or cost consideration for repairing, there would be different preventive maintenances applied correspondingly. Putting data mining techniques into use is able to analyze historical data and then specify what are the better maintenance strategies. Also, to have a more objective rating system of maintenance charges leads to find the most critical preventive maintenance items and focuses under varied systems of preventive maintenance so as to meet the goal of minimizing the total cost and to downsize the occurring rate of incidents. Applying the model of grading maintenance charges proposed by our institute to budget planning can increase the accuracy when designing budget, decrease total maintenance costs yet prevent incidents in order to assure the military performance.

1. 鄭達才（2000）。設備維護管理-現在與未來。臺北：財團法人中國生產力中心。
2. 陳沅銘（2002）。一般週期預防維護模型系統之發展。碩士論文。朝陽科技大學。
3. 李汶娟、吳忠武、洪清文、何侃錚和林義欽（2013）。利用上記錄值樣本對具有廣義指數分配的產品之壽命性能指標做績效評估。品質學報，第20 卷，第3期，第275-304 頁。
4. 張家華（2012）。壽命具廣義伽瑪分配之型II 設限階段應力加速壽命試驗的可靠度分析。碩士論文。國立中央大學。
5. 蘇國瑋（1997）。壽命資料在對數常態分配與韋伯分配檢定上的探討。碩士論文。國立中央大學。
6. 曾鵬仁（2018）。運用資料探勘技術於高價值顧客之探索研究。碩士論文。國立屏東科技大學。
7. 陳樹衡、郭子文、棗厥庸（2007），以決策樹之迴歸樹建構住宅價格模型—台灣地區之實證分析。住宅學報第十六卷第一期，p.1-20。
8. 蔡崇煌、翁紹仁、周駿安、吳信宏、洪偉展，運用C4.5決策樹分析失眠症狀。台灣衛誌 2017，Vol.36(5)，p.449-460。
9. 資料探勘：維基百科，自由的百科全書https://zh.wikipedia.org/zh-tw/数据挖掘
10. 林盈源 （2003）。決策樹在資料庫行銷決策之應用。碩士論文。國立成功大學。
11. 曾瑞智（2013）。應用資料探勘技術建構整合型目標客戶選擇模式。碩士論文。大同大學。
12. 尹相志（2008），SQL Server 2008 Data Mining 資料採礦，悅知文化。
13. 袁梅宇（2015）。王者歸來：WEKA機器學習與大數據聖經[第3版]。佳魁資訊。
14. 葉智瑤（2011）。從效益後勤觀點探討野戰保修維保模式及核心能量之研究。聯合後勤季刊。第25期，pp.74-76。
15. 林德偉（2011）。啟發式演算法於預防維修成本最佳化之研究。博士論文。國防大學理工學院國防科學研究所。
16. 鄭滄祥（2008） 。資料探勘。高麗圖書有限公司。
17. 孫惠民（2007） 。資料採掘理論與實務規劃手冊。松崗電腦圖書有限公司。
18. Jabar, H. B. (2003).Plant Maintenance Strategy: Key for Enhancing Profitability. Maintenance Resources. http://www.maintenanceresources.com/referencelibrary/ezine/chemcl ean.htm
19. Moghaddam, K. S. and J. S. Usher.(2011) . Preventive Maintenance and Replacement Scheduling for Repairable and Maintainable Systems Using Dynamic Programming.Computers and Industrial Engineering ,60 (4),pp.654-665.
20. Nguyen, D. G., and D. N. P. Murthy(1981).OptimalPreventive Maintenance Policies for Repairable Systems. Operations Research ,29 (6),pp1181-1194.
21. Doostparast, M., F. Kolahan, and M. Doostparast(2014).A Reliability-Based Approach to Optimize Preventive Maintenance Scheduling for Coherent Systems. Reliability Engineering and System Safety 126,pp.98-106.
22. W. Frawley and G. Piatetsky-Shapiro and C. Matheus (1992). "Knowledge Discovery in Databases: An Overview". AI Magazine: pp. 213-228. ISSN 0738-4602.
23. Berry, M. J. A., & Linoff, G. (1997). Data Mining Techniques for marketing, sales and customer support. 2000. John Wiley & Sons. Cap, 1(2), pp.3.
24. Fayyad, U. M., Piatetsky-Shapiro, G., & Smyth, P. (1996). Knowledge Discovery and Data Mining: Towards a Unifying Framework. In KDD,96, pp.82-88.
25. McSherry, D.(1999). Strategic induction of decision trees. Knowledge-Based System,12(5/6), pp.269-275.
26. Yada, K. (2011). String analysis technique for shopping path in a supermarket. Journalof Intelligent Information System, 36(3), pp.385-402.
27. Cortes, C., & Vapnik, V.(1995). Support-vector networks. Machine learning, 20(3), pp.273-297.
28. Vapnik, V. (2014). Invited Speaker. IPMU Information Processing and Management of Uncertainty in Knowledge-Based System.
29. Chang, C. C., & Lin, C. J. (2011). LIBSVM: a library for support vector machines. ACM transactions on intelligent System and technology (TIST), 2(3),pp.27.
30. Quinlan, J. R. (2014). C4. 5: programs for machine learning. Elsevier.
31. Viera, A. J., &Garrett, J. M. (2005). Understanding interobserver agreement: the kappa statistic. Fam Med, 37(5),pp.360-363.
32. MacQueen, J. (1967, June). Some methods for classification and analysis of multivariate observations. In Proceedings of the fifth Berkeley symposium on mathematical statistics and probability,1(14), pp.281-297.
33. Wang, H.(2002). A Survey of Maintenance Policies of Deteriorating System. European Journal of Operational Research,139, pp.469-489.
34. Phan, H. and Wang, H. Z. (1996), Imperfect Maintenance. European Journal of Operational Research,94, pp.425-438.