本研究之目的在以競標方式爭取訂單的供應鏈環境中，提供賣方(供應商)決定其競標價格的決策機制。為了提升供應商之利潤與得標機會，本研究整合競標決策、訂單允諾機制與生產計劃，以交期允諾及生產成本做為競標價格決策之基礎。其中，訂單允諾機制乃以進階可允諾量(Advanced Available-to-Promise, AATP)之觀念為基礎，亦即考慮未來可應用產能之最佳配置。本研究以混合整數規劃(Mixed Integer Programming, MIP)模型來描述競標決策問題，模型中並包含模糊限制式以表達決策者在制定標價時對顧客價格容忍度的認知。在本研究之規劃環境中，訂單的允諾乃採批次處理與滾動規劃週期的設計，亦即每隔一固定期間，供應商會重新審視所收到而未允諾或已允諾但尚未完成生產之訂單，並重新執行上述之決策模式以更新生產計畫。模型之求解過程包含以模糊方法搜尋最大利潤與最大得標機會之妥協解，並以遺傳演算法(Genetic Algorithm, GA)實作求解程序。本研究以電腦模擬的方式進行實驗以驗證本研究方法之績效。 This study integrates the bidding decision and production planning based on the concept of advanced available-to-promise (AATP) inventory with a rolling planning horizon. Customer requests arrive in a random fashion, and bidding decisions are made for a batch of requests collected over a batching interval. This decision process repeated for every specified batching interval, and the current decision-making must take into account the previously committed orders in earlier phases. The problem is formulated as a mixed integer programming model with fuzzy constraints, which express the decision-maker’s subjective judgment regarding customer’s price tolerance. The proposed model embeds the AATP concept to support accurate computation of profit and customer order promising. A genetic algorithm is developed to solve the problem. Performance of the proposed approach is evaluated through experiments conducted by computer simulation.
