經濟批量排程問題（economic lot scheduling problem；ELSP）的最佳化求解已被證明是一個具NP-hard複雜度的問題。在過去一般的ELSP模式中，大都假設存貨可以無限期的儲存，並未考慮產品具有損耗性。但是在現實生活中，常見產品具有損耗性；本研究在此前提假設下，使用循環週期法找出損耗品ELSP模式下的最佳解。在探討損耗品ELSP模式，本研究推導出在若干現實生活中可能發生的特定條件下，運用循環週期法的封閉解(closed-form solution)即為該模式之最佳解。同時，本研究亦對於所推導之特定條件進行敏感度分析，探討特定條件被寬鬆後，循環週期法所得之解與最佳解間誤差之上限。另外，本研究延伸上述損耗品ELSP模式的推導，以「可運用於整備的機台時間」定義機台產能限制，並允許擁有延遲的訂單；依此，本研究亦推導出在何種特定條件下，使用循環週期法亦可得到最佳解。而且，我們探討循環週期法在於群組技術問題的實務應用。而在捨棄特定條件的假設下，本研究則藉由遺傳演算法(GA)多點平行搜尋的優點，迅速求取損耗性ELSP可能的最佳解。再運用經濟批量排程問題之可行解測試法(Feasibility Testing Procedure for the ELSP；Proc FT)，來判斷GA所求得的解是否可行。 The Economic Lot Scheduling Problem (ELSP) is concerned with the lot sizing and scheduling decision of n items. It has been proved that the ELSP is an NP-hard problem. In the conventional ELSP model, it does not consider the product stock could deteriorate over time. The focus of this study is to employ the rotation cycle (RC) approach to solve the ELSP with deteriorating items. The RC approach secures a feasible production schedule in which each item share the same replenishment cycle. We show that the RC approach’s schedule is optimal for in many realistic situations. Importantly, we give upper bounds for the maximum percentage that the RC approach’s schedule deviates from optimality. On the other hand, we extend our theoretical analysis to the case that includes the machine capacity constraint by considering the long run proportion of time available for setups. We also show how these results can be used when backorders are allowed. An implication for Group Technology is discussed. The proposed algorithm was tested by real data for electronic industry, and this approach can be applied to most constraint-satisfaction problems. If these situations mentioned-aboved does not exist, we derive a hybrid genetic algorithm (HGA) that utilizes the advantage of multi-directional search in GA and employs an efficient heuristic, namely, Proc FT, to test feasibility of the solutions found.
