本研究探討多機開放型生產型態之動態排程問題,以最小化總完工時間為衡量指標,此問題已被證實為NP-Hard,因此發展啟發式演算法以縮短求解時間,使適用於實務上。 本研究將文獻所提幾個用在多機開放工廠靜態排程的派工法則擴展,使適用於本研究,並且提出考量不同優先條件的啟發式演算法,最後擷取這些方法之優點發展啟發式演算法MDLTRP。此法搜尋工作或機台最先可以開始的時間,接著排入此時間點可以排的作業,此作業必須是後續加工時間總和最大的。最後測試機台數與工作數相同的大規模問題。而依據實驗結果顯示,本研究所提方法之績效遠大於標準測試問題,顯示本研究方法有實際的應用價值。
This study considers the dynamic open shop scheduling problem with the objective of minimizing the makespan. It has been established that the problem is NP-hard. An efficient heuristic, MDLTRP( modified longest total remaining processing on other machines first), is therefore proposed. A point of time is picked among the earliest available job release time and machine ready time. A job with the largest remaining processing time is selected to process at that time. Computational results show that the heuristic outperforms the existing algorithm.