摘要
國防軍備生產體系下各軍工廠之武器系統生產研製，係依各軍種客戶之訂單（Make to Order, MTO）來執行武器生產製造作業。武器系統生產型態具有：製程複雜（包含零工Job Shop與組裝Flow Shop）、高品質及高精密度要求、少量多樣、加工時間長、資源限制多、訂單製令變動頻繁及訂單交期遞延須嚴懲等特性。上述特性導致生產規劃之困難度相對地提高，以傳統生產規劃及控制系統，如物料需求規劃（Material Requirement Planning, MRP）及粗略產能規劃（Rough-Cut Capacity Planning, RCCP）已不能滿足武器生產規劃之需求。且因政府採購法的規定，物料的取得有許多的限制，由於供應商品質與交期無法有效的掌握，導致物料到達時間多有遲延，所以必須正確且快速的得知生產排程需求物料的時點及數量，以及早規劃採購交貨方案。因此，良好的生產排程規劃將可以提高武器生產系統的物料供應與產出效率及生產排程彈性，並提升整體生產營運績效。 本研究針對武器生產系統發展一以先進規劃與排程技術（Advanced Planning and Scheduling; APS）為基礎之啟發法則（Heuristics），同時考量產能與物料限制，發展以工作站（Workstation）為製程焦點及以低階加工作業指派優先之排程方法，以滿足武器產品多階組合、結構複雜特性之生產排程作業模式。本研究使用實際資料評估所發展之演算法，透過實驗設計分析，結果顯示本演算法可有效提高訂單達交率及機台使用率並達到較佳之機台間負荷平衡。
並列摘要
Weapon production system manufactures weapons for various military divisions on a make-to-order basis. The characteristics of this system include complicated manufacturing process, high quality and precision requirement, high order variety, low order volume, long lead time, multiple resource constraint, frequent change of orders, and high penalty of late delivery. These characteristics make it difficult to use traditional production planning and control system such as material requirement planning and rough-cut capacity planning. This research proposes an advanced planning and scheduling heuristic for weapon production system. This heuristic is workstation-focused and assigns higher priority to lower-level operations. Both capacity and material constraint are taken into account. Real data are used to evaluate the performance of the proposed heuristic. Preliminary simulation results based on experimental design show that the heuristic leads to higher on-time delivery percentage, higher equipment utilization, and better equipment loading balance.