應用混合蟻群演算法於整合消費性電子供應鏈中具順序相依性設置時間之生產與空運配送之排程問題

Abstract

消費性電子產業由於需求不確定性高、產品壽命較短，以存貨作為各階段間的緩衝，已不是一個有效的策略，因而逐漸傾向零存貨與訂單式生產。且此類產品對時間較為敏感，價值高，重量相對也輕，加上顧客遍佈全球，企業多選擇空運，來縮短產品送達顧客的前置時間。然而在生產與空運配送各階段之目標間，彼此存在互償(trade-off)關係。若以系統整體最佳化作為首要目標，就有必要同時考慮此兩階段的作業。另外，隨著及時生產的興起，作為服務水準的績效指標，準確地完工或完成配送越來越重要。在生產與配送階段中，太早或太晚完工或完成配送，將可能導致額外的浪費。因此應同時協調生產與配送的排程，在服務水準與各階段成本間取得帄衡。本研究探討整合消費性電子產品供應鏈中，生產與空運配送之排程問題，期望找出一套同時協調兩階段的排程，讓生產與配送階段藉由互償的方式，使所有的訂單盡可能準時完工並送達顧客手中，且花費最低的運輸成本。其中每項訂單都有其特定的交期，代表必頇在此期限上交貨到該顧客手上，否則就會產生懲罰。本研究將整合兩階段之排程問題分成空運規劃與具順序相依性設置時間之生產排程問題，藉由求解空運規劃問題，找到一組滿足工廠產能限制，且最小化提前或延遲將成品送達顧客的總時間成本加上總運輸成本之訂單配送排程後，將運送該訂單的貨機之起飛時間當作工廠的交期，再進行具順序相依性設置時間之生產排程問題的求解。而在生產排程階段，則以找到一組工作生產排程，使所有工作皆可趕上配置的貨機，且最小化商品提前完工之總時間成本為目標。本研究發展一套具有動態規劃、以蟻群演算法為架構，並混合模擬退火法之混合式巨集式演算法之演算流程，來求解本問題。此外，為了避免在空運規劃問題階段錯估工廠產能，本研究設計一個回饋機制，當生產排程階段發現產能無法滿足訂單配送排程時，空運規劃問題將重新考慮工廠產能限制，產生新的訂單配送排程，並重新安排加工順序，直到空運規劃問題所考慮的產能與生產排程階段所需要的真實產能相符為止。本研究亦針對整合考量與順序式考量進行比較，由測詴結果顯示，帄均相對差距約在50%，顯示整合兩階段進行排程之價值與重要性，並可做為相關產業實務上的參考依據，讓業者在考量成本與顧客服務水準間做出合理的決策，進而達大最大化整體利潤的目的。

More and more enterprises have chosen to use make-to-order or direct-order business models in order to be competitive in demanding market. In such business models, enterprises are forced to reduce their inventory but still have to respond quickly. The reduction of inventory results in closer interaction between production and distribution and thus increases the usefulness of an integrated model. This paper studies the problem of synchronized scheduling of manufacture and air transportation to achieve accurate delivery with minimized cost in consumer electronics supply chain. The overall problem is decomposed into two sub-problems The air transportation allocation problem in which an order is delivered as whole using only one flight is formulated as an integer linear programming problem with the objective of minimizing transportation cost and delivery earliness tardiness penalties. The production scheduling problem which considers sequence-dependent setup time on unrelated parallel machines seeks to determine a schedule ensuring that the order are completed on time and catch the flight such that the waiting penalties between production and transportation is minimized. A dynamic programming algorithm is designed to solve the air transportation allocation problem with un-split delivery, and a hybrid meta-heuristic algorithm which combines several features from ant colony optimization and simulated annealing is presented to solve the sequence-dependent production scheduling problem. The computational results show that the hybrid algorithm produces good results. Furthermore, there is a mechanism to feed back the information of un-enough production capacity to air transportation stage by adjusting the restriction on production rate. This study also investigates the benefits of using the integrated model relative to a sequential model. The test results show the value of integration and the importance of integrated model for an integrated production and distribution scheduling problem.