以限制理論為基礎之工廠績效評估及派工法則

Abstract

在整個生產規劃與控制的決策過程中，交期和週期是兩個重要的評量準則，透過排程派工法則可以有效的管控訂單週期的時間長短和交期的達成。一般而言常以「平均流程時間」和「最大延遲交貨時間」指標來代表週期和交期的評量。從過去文獻上得知有很多派工法則被發展出如何有效達成交期和週期的績效，如最短處理時間(SPT)和最早交期到達(EDD)派工法則，分別可以達成訂單平均流程時間及最大延遲交貨時間最小化的效益。但是根據限制理論(TOC)認為，平均流程時間和最大延遲交貨時間，此兩種指標只有考慮時間性而且是針對生產線局部化的評量基準，此兩項績效指標對工廠整體性而言不夠好，例如就顧客訂單而言，當工廠產能有限而導致延遲交貨發生，訂單金額大小所影響的損失程度就會不同，所以訂單價值應該區分出他的重要性。另方面工廠為了預防訂單遲交往往採取提前下料的政策，但這種策略提高了在製品存貨水準，也拉長了整個訂單的生產週期時間，所以不同物料價值其加工時間長短也應該加以重視，針對上述兩點TOC提出了兩種新的評量指標即「有效產出天元(TDD)」與「存貨持有天元(IDD)」，前者代表每項延遲訂單金額乘上其遲交時間之加總，後者代表每項訂單物料成本乘上其流程時間之加總。整體而言這兩項指標可以取代傳統的評量基準，來引導生產線去做對整個工廠系統有益的事。所以如果TDD和IDD是不錯的評估指標，那傳統上宣稱在平均流程時間和最大延遲時間指標下表現不錯的派工法則，對於 TDD與IDD指標表現又產生何種效果? 是否存在好的派工法則能產生較佳的TDD/IDD績效?是否能發展出一套以TDD/IDD績效最適化之啟發式派工法則? 而TDD和IDD兩項績效指標是否能整合運用? 經由本文模擬驗證工廠在限制理論準則的運作下，其績效評估及派工法則都能得到更令人滿意表現。

Due date and cycle time are two major criterion in the aspect of production planning and decision-making control for various factories. Two time-based indexes, the mean flow time and maximizing tardiness, have been used for a long time as performance indicators in cycle time and due date respectively. The SPT and EDD dispatching rules can perform well in terms of mean flow time and maximizing tardiness individually. However, the TOC (theory of constraint) thinks the two indexes be time-based and local measurement merit. They are not good indicators for factory as a whole. For example, the impact of delaying to delivery depends on the value of order. Moreover, for preventing the order from being tardiness, the policy of early releasing raw material in advance will cause higher WIP level and increase flow time of throughput. Therefore, TOC addresses the concepts of TDD (Throughput dollar-day) and IDD (Inventory dollar-day). The TDD is the summation of the sales value multiplied by the days of delay and IDD is the summation of the material value multiplied by the time since the WIP is processed. However no one has studied dispatching rules based on TDD and IDD performance indicators. Therefore the paper develops the heuristic algorithm (TDD/IDD dispatching rule) to maximize TDD and IDD performance. Some traditional dispatching rules are compared with the proposed algorithm in terms of TDD and IDD. The results demonstrate that the TDD/IDD dispatching rule is feasible and outperforms than other dispatching rules.