- 學位論文
運用群組基因演算法設計預防維護系統
Using Grouping Genetic Algorithms to Design Preventive Maintenance System
摘要
隨著產品日益複雜與顧客期望的提升，維護保養與增進產品的品質受到越來越多的注意。現今的高科技產業為了避免突發故障所造成的巨額損失，設備都必須具有高的可靠度與穩定性。企業投資越來越多的資源來改善機台的使用率以維持設備良好的運轉。因此，發展一個良好的方法來設計預防維護系統是非常重要的。 本研究的主要目的為建構一個預防維護的系統。在本文中,它是一個可交換零件最優化指派 (Optimal Assignment of Interchangeable Component, OAIC) 的問題，多重配置問題 (Redundancy Allocation Problem, RAP) 中的一種。從最近迅速增加的文獻中，許多啟發式的方法與維護的策略會被討論與歸納。之後，本研究將針對可交換零件最優化指派問題提出群組基因演算法。Falkenauer於1992年首先提出群組基因演算法(Grouping genetic algorithm, GGA)，它是針對群組化問題所發展出來的特殊結構的基因演算法，並且改良基因演算法的缺點。群組基因演算法可以有效的處理複雜結構的問題，例如本文的可交換零件最優化指派問題。在本文中，績效指標證明群組基因演算法比基因演算法優良。計算結果顯示群組基因演算法在求解可交換零件最優化指派問題時，更具有效率。
並列摘要
As manufactured goods becoming more complex and customer’s expectations growing, increased attention is being paid to maintenance and improving product quality. To avoid the huge losses caused by sudden failures, the precision manufacturing of the Hi-Tech industry requires highly reliable and stable equipment. Enterprises have invested more and more resources to improve the rate of utilization of the machine in order to maintain the good operation of the process flow. Hence, it is important to develop a good method for preventive maintenance system design. The main purpose of this study is to design a preventive maintenance system, which is an optimal assignment of interchangeable component (OAIC) problem, a kind of redundancy allocation problem. Various heuristic methods and maintenance policies are discussed and summarized from the rapidly growing literature. Afterward, the grouping genetic algorithm (GGA) is presented for OAIC problem. GGA was first developed by Falkenauer in 1992 as a type of GA which exploits the special structure of grouping problem, and overcomes the drawbacks of GA. GGA can be effectively adopted for complex combinatorial problems, such as OAIC problem. In this study, the performance has been verified that the GGA is better than GA. The computational results show that GGA has more effectiveness in solving of OAIC problem.