摘要
本研究以WinQSB軟體中的CDS (Campbell、Dudek and Smith)、Palmer與RA (Rapid Access)建構式算法及Excel規劃求解之演化式算法求解流程式生產排列排程問題，並以Carlier及Rec測試問題為研究案例。研究結果顯示，CDS、Palmer與RA皆無法求得問題之最佳解，僅能求得較佳解。Excel規劃求解演化式算法能夠求得全部Carlier測試問題之最佳解，但無法求得全部Rec測試問題之最佳解。不過，演化式算法求得之解與真正最佳解的距離最小，仍優於CDS、Palmer與RA建構式算法。
關鍵字：流程式生產排列排程、Excel規劃求解、建構式、演化式。
ABSTRACT
This paper aims to compare the difference among constructive algorithms and Excel Solver evolutionary algorithm in solving the permutation flow shop scheduling problems. The scheduling problems for testing the algorithms are Carlier and Rec problems. First of all, constructive heuristics such as CDS (Campbell, Dudek and Smith), Palmer, and RA (Rapid Access) provided by WinQSB software are applied to solve the testing problems. Then, the evolutionary algorithm provided in Excel Solver is applied to solve the testing problems. According to the results, it can be found that CDS, Palmer, and RA cannot find optimal solutions to the testing problems. The evolutionary algorithm is able to find the optimal solutions of all Carlier testing problems but is unable to find the optimal solutions of all Rec testing problems. However, the evolutionary algorithm is still better than CDS, Palmer, and RA algorithms.
Keywords: Permutation Flow Shop Scheduling; Excel Solver; Constructive; Evolutionary
|