傳統產業處於貿易全球化、自由化的競爭壓力大之現代環境中,若不能提升產業競爭能力,將會面臨被淘汰的危機。而現今經濟快速發展與消費型態的改變,消費者對商品的選擇自主性提高,使得以往由供應商主導的銷售方式改變成為以消費者需求為依歸,因此貨運業者除了為要滿足顧客少量多樣以及在一定時間點內送達的服務需求外,且需要在有限的工作人員及運送設備條件下,降低運送成本以提高企業的利潤,故路線貨運業如何提升競爭力、降低運輸成本,遂成為重要目標之一。 個案公司歷史悠久,目前仍以人工作業方式進行多場站車輛途程之排程問題,而個案公司之車輛途程有最早最晚抵達時間的限制,因此本研究企圖建立一排程決策支援系統,在考慮其時窗限制下,規劃多場站之貨物路徑及配送數量,希望能藉此求得最少之運送車輛數。目前多數的文獻都以單一演算法來解決車輛途程的問題,然而在這些演算法中,何者較適用卻無法比較,因此,本研究建立之系統將包括最小鄰近法、基因演算法和禁忌搜尋演算法,進行運送車輛數、運送成本與運算時間之比較,以提供業者在不同的情境下最佳的選擇,並做為業者全面運籌管理電子化之參考。
To strengthen the competitiveness and to reduce the transportation cost are very important in transportation business environment. Most former researchers used only one algorithm to solve Vehicle Routing Problems. Not only have the business owners wanted to solve the transportation allocation problems but also the most efficient way. In our research we set up the system of Multi-Depot Vehicle Routing Problem with Time Windows. The performance of the Smallest Neighboring Law, Genetic Algorithm and Tabu Search Algorithm are compared by operating time and transportation cost in this decision support system. The conclusion is proposed as a recommendation for a logistic management industry to an e-industry.