多類型故障型態下最適預防保養策略研究

THE OPTIMAL PREVENTIVE MAINTENANCE UNDER MULTI-FAULT PATTERNS

本文使用數值模擬方法，規劃一套最適預防保養策略，以提升設備維護績效。因若等到設備發生故障才進行更換，則會使正在運行中的作業停止，更會造成各相關成本的增加。本研究在已知機器設備之壽命分配下，根據隨機過程中之更新理論，於使用期間採取適當的預防保養策略。本文建構壽命期間內總期望成本模式，其中考量預防保養成本、故障維修成本與更新成本等三項因素，配合各種可能發生之情境與給定之成本參數，利用電腦軟體設計數值積分與運算程式，找尋最佳預防保養週期，以期達到總期望成本最小化之目標。此外，透過敏感度分析以了解各參數對總期望成本與最適預防保養週期之影響。最後，提出十點結論以供後續研究與實務應用之參考。

In this paper, we planned an optimal preventive maintenance strategyto improve maintenance performance under the lifetime distribution wasgiven. If we wait to replace the device until it has failed, then the shutdownand related cost will be enormous. Therefore, based on stochasticprocess and renewal theory, we constructed the total expected cost model.Furthermore, we designed a numerical integration program to find out theoptimal preventive maintenance interval that minimized the total expectedcost. The total expected cost includes the preventive cost, maintenancecost, and renewal cost. The sensitivity analysis was also taken to realizethe influence of parameters on total expected cost and optimal preventivemaintenance interval, respectively. Nine conclusions were drawn for usefuture studies and practical applications.