【摘要】：物流配送合理利用社會資源,有效降低能源消耗,在優(yōu)化產業(yè)結構、完善產業(yè)鏈、提升經濟運行質量等方面發(fā)揮著強大的促進作用。其中,車輛路徑優(yōu)化是物流配送問題的一項重要研究內容。選擇合適的車輛路徑,不僅能降低貨物運輸和配送的成本,提高企業(yè)和社會效益,同時能節(jié)省能源,保護環(huán)境,緩解交通壓力。 GIS是對空間信息進行收集、存儲、分析以及可視化表達的信息管理系統(tǒng),主要應用于對一定區(qū)域范圍內分布的現(xiàn)象進行分析處理,同時進行規(guī)劃管理以及決策。因此,本文設計了一個基于GIS技術的物流配送系統(tǒng)。利用ArcGIS平臺技術中的網絡分析模塊,計算物流中心與客戶以及客戶與客戶之間的距離矩陣,使用蟻群算法得到能夠滿足用戶所需貨物需求、時間窗口需求、總體路徑最短、運輸成本最小的配送策略,最后使用ArcGIS技術為客戶提供可視化的界面。 此外,本文對車輛路徑問題求解所用的蟻群算法的路徑選擇機制、信息激素更新機制進行了改進,一方面,通過隨機性選擇和確定性選擇相結合的方式,在信息素濃度最大的路徑上,按照一定的規(guī)則進行轉移,對于其它的可選路徑,則根據輪盤賭的方式來選擇。確定性可以保證螞蟻總是選擇轉移概率最大的路徑,而隨機性則可以避免算法過快的收斂于局部最優(yōu)解；另一方面,充分利用目前已經找到的最優(yōu)解,在每次循環(huán)之后,依據當前循環(huán)中最優(yōu)解的螞蟻的信息素對所有路徑上的信息素濃度進行更新；除此之外,采用信息素平滑方法通過增強信息素軌跡量較低的解元素的信息素軌跡量,從而提高選擇其選擇概率,來擴大算法的搜索范圍。 最后,以重慶市煙草公司為例,設計基于GIS技術的物流配送系統(tǒng)得以實現(xiàn)。而通過該系統(tǒng),煙草公司車輛的裝載率提高了三分之一,送貨的次數降低了三分之二,有效的避免了車輛出行不滿載的情況,同時節(jié)約了將近二分之一的物流配送成本,取得了明顯的經濟效益。
[Abstract]:Logistics distribution makes rational use of social resources, effectively reduces energy consumption, and plays a strong role in optimizing industrial structure, perfecting industrial chain and improving the quality of economic operation. Among them, vehicle routing optimization is an important research content of logistics distribution problem. Choosing the appropriate vehicle path can not only reduce the cost of freight transportation and distribution, improve the social and social benefits, but also save energy, protect the environment and alleviate the traffic pressure. GIS is an information management system for collecting, storing, analyzing and visually expressing spatial information. It is mainly used to analyze and process the phenomena distributed in a certain area, and to carry out planning management and decision-making at the same time. Therefore, this paper designs a logistics distribution system based on GIS technology. Using the network analysis module of ArcGIS platform technology, the distance matrix between logistics center and customer, as well as between customer and customer is calculated, and the ant colony algorithm is used to obtain the distribution strategy which can meet the needs of goods, time window, the shortest overall path and the least transportation cost. Finally, ArcGIS technology is used to provide a visual interface for customers. In addition, this paper improves the path selection mechanism and information hormone updating mechanism of ant colony algorithm used to solve vehicle routing problem. On the one hand, through the combination of random selection and deterministic selection, the path with the maximum pheromone concentration is transferred according to certain rules, and for other optional paths, it is selected according to roulette. Certainty can ensure that ants always choose the path with the largest transition probability, while randomness can avoid the algorithm converging too quickly to the local optimal solution. On the other hand, make full use of the optimal solution that has been found at present, after each cycle, update the pheromone concentration on all paths according to the pheromone of the ant with the optimal solution in the current cycle. In addition, the pheromone smoothing method is used to expand the search range of the algorithm by enhancing the information element trajectory of the solution elements with low pheromone trajectory, so as to improve the selection probability of the solution elements. Finally, taking Chongqing Tobacco Company as an example, the logistics distribution system based on GIS technology is designed. Through this system, the loading rate of tobacco company vehicles is increased by 1/3, the number of delivery is reduced by 2/3, the vehicle travel is not full of load, and nearly 1/2 of the logistics distribution cost is saved, and obvious economic benefits are obtained.
【學位授予單位】：浙江工業(yè)大學
【學位級別】：碩士
【學位授予年份】：2013
【分類號】：P208;TP311.52

【參考文獻】

相關期刊論文 前10條

1 石欣,程代杰;區(qū)域配送中心物流調度模型[J];重慶大學學報(自然科學版);2004年01期

2 夏國成;趙佳寶;;智能螞蟻算法求解多目標TSP問題的改進研究[J];計算機工程與應用;2006年09期

3 郭平;鄢文晉;;基于TSP問題的蟻群算法綜述[J];計算機科學;2007年10期

4 萬旭,林健良,楊曉偉;改進的最大-最小螞蟻算法在有時間窗車輛路徑問題中的應用[J];計算機集成制造系統(tǒng);2005年04期

5 胡小兵,黃席樾;蟻群優(yōu)化算法及其應用[J];計算機仿真;2004年05期

6 馬良,姚儉,范炳全;螞蟻算法在交通配流中的應用[J];科技通報;2003年05期

7 王穎,謝劍英;一種自適應蟻群算法及其仿真研究[J];系統(tǒng)仿真學報;2002年01期

8 崔雪麗,馬良,范炳全;車輛路徑問題(VRP)的螞蟻搜索算法[J];系統(tǒng)工程學報;2004年04期

9 姜大立,楊西龍,杜文,周賢偉;車輛路徑問題的遺傳算法研究[J];系統(tǒng)工程理論與實踐;1999年06期

10 覃剛力,楊家本;自適應調整信息素的蟻群算法[J];信息與控制;2002年03期

，

本文編號：2511208