頁籤選單縮合
題名 | 應用類神經網路於物流中心區位及車輛排程問題之探討=The Study of the Distribution Center Location and Vehicle Routing Problem by Applying the Hopfield-Tank Neural Network |
---|---|
作者 | 池文海; 傅家啟; 劉興鴻; Chih, Wenhai; Fu, Jachih; Liu, Shinhong; |
期刊 | 工業工程學刊 |
出版日期 | 20000100 |
卷期 | 17:1 2000.01[民89.01] |
頁次 | 頁25-39 |
分類號 | 494.578 |
語文 | chi |
關鍵詞 | 物流中心; 區位-車隊規模-排程整合模式; 霍普菲爾-坦克類神經網路; Distribution center; Location fleet-size routing model; Hopfield-tank neural network; |
中文摘要 | 隨著工商業的競爭越形激烈,業界漸漸體會到,物流是企業體本身保持競爭優勢並且永續生存的重要因素之一,本研究針對物流中心整體規劃中的區位問題,加入車輛及排程因素的考量,形成一個區位一車隊規模一排程整合模式,並應用類神經網路中的霍普菲— 坦克類神經網路(Hopfield-Tank Neural Network)加以求解,其中必須進行能量函數、運動方程式等的修正,使之符合此區位整合模式的特性;然後以實驗設計的方法,分析網路中主要參數之間的範圍及最適參數組合,並對其顯著性項目進行敏感度分析。在實例驗證中,霍普菲爾—坦克類神經網路,對於最佳區位之選擇,展現了高達100%的正判率,同時也節省了大量的運算時間。 |
英文摘要 | Choosing the right location for distribution centers is crucial to the long-term competitiveness of enterprises ranging from distribution services to large-scale manufacturers. This paper presents a neural network based method for the integrated planning of distribution centers which considers the location, fleet size, and routing factors. The network is based on the Hopfield-Tank model, with the energy function and equation of motion tailored to the location integrated model. The practical use of the network is demonstrated by giving it various combinations of input values in several example problems. The output is used to analyze the range of problem solutions and perform sensitivity analysis. In all of example problems, the Hopfield-Tank Neural network executed rapidly and yielded good practical solutions. |
本系統之摘要資訊系依該期刊論文摘要之資訊為主。