本研究考慮雙機流程型工廠下,所有工作在機台上有其加工時間、設置時間與移除時間;每個工作在兩個機台加工之前,皆需要有一個設置的時間,此設置動作須由單一操作人員執行,即在兩台機台上的設置動作不能同時進行;每個工作在各機台完工之後,需要有一個移除的時間,此移除動作須由同一操作人員執行之,兩台機台上的移除動作亦不能同時進行,本研究目標為使延遲工作個數(Number of tardy jobs)最小化。過去文獻中提到類似的問題已證明為NP-hard,而本研究將發展啟發式演算法、下限值與分支界限演算法;經實驗後證明,本研究提出的H1啟發式演算法與最佳解的平均誤差在(α,β)值為(0.1,0.3)與(0.2,0.4)的情況下皆不超過10%。
This work studies the scheduling problem where a set of jobs are available for processing in a separate setup and remove two-machine flow shop system with a single server. The setup time and removal time are incurred and needs to be attended by a single server which can perform at most one setup or remove at a time. The performance measure considered is the number of tardy jobs. The restricted version of the problem under study is strongly NP-hard, which implies the the general problem is also strongly NP-hard. We establish two heuristics, one lower bound and the branch and bound algorithms. Computational results showed that the deviation of proposed heuristic H1 and optimal solution does not yield 10% when (α,β) is equal to (0.1,0.3) or (0.2,0.4).