Abstract
In practice, a system or equipment has a finite useful life. When an aged system is replaced by a new system, the new system seldom has exactly the same characteristics, functions, investment cost, maintenance expenses, etc., as those of the aged system. However, the literature has shown that most researchers are devoted to studying maintenance problems in an infinite time span, which assumes that the replaced system in each replacement cycle (or period) has the same conditions and costs. Apparently, the assumption might not be practical, since the useful operating life of most systems is finite in the real world. Therefore, the purpose of this research is to develop an optimal preventive maintenance (PM) policy having failure rate reduction within a finite time span under a warranty consideration by minimizing the expected total maintenance cost. The PM cost is assumed to be a linear function of each PM effect. Two cases are studied and compared; one is with no warranty provided and the other is with a given warranty period. For the second case, we examine two policies: (1) no PM within the warranty period is assumed and (2) the PM will be performed within the warranty period. In this article, the algorithm for searching the optimal solution is presented. Examples with Weibull failure cases are given and the sensitivity analysis of the optimal solution is also provided.
在實務上 , 一個系統或設備的可用壽命是有限的。 一個老舊設備被汰舊換新時 , 新設備很少會和汰舊的老設備有完全相同的特性、 功能、 投資成本、 以及維護成本等。 然而 , 維護問題相關文獻所呈現的大部份屬於無限時間域的研究 , 其假設在每個置換週期的汰舊設備會和新設備具有完全相同的條件和成本。 很顯然地 , 此假設可能不能完全符合實務狀況。 因此 , 本研究針對有限時間域(finite time span) , 發展一個具有失效率縮減(failure rate reduction)特性的最佳預防維護政策(optimal preventive maintenance policy)。 本研究探討無保固期與給定保固期兩種情況的最佳維護政策 ; 針對給定保固期情況 , 再比較在保固期內 (1)不執行預防維護以及(2)執行預防維護兩種政策。 本研究另提出搜尋最佳解的演算法 , 並以韋伯(Weibull)失效分配為例提供範例 , 列出各情況的最佳維護政策與敏感度分析。
(*聯絡人: [email protected])
Keywords:
Acknowledgements
This research was supported by the National Science Council of Taiwan for Project NSC 96-2221-E-324-010.
Notes
(*聯絡人: [email protected])