Study On Routing Optimization Of Logistics Distribution Based On Improved Ant Colony Algorithm

The rapid development of economy has contributed to the high-speed innovation of logistics industry, and we have continually looked for the method of transportation-cost reduction because of the great importance of transportation-cost in logistics distribution, as a result, the way that decrease the cost by optimizing distribution path has naturally became the hot issue researched by the scholars at home and abroad.Ant Colony Algorithm(ACA) is a kind of bionic algorithm, and it is a random search method obtained by inspiration and further study of biological group behavior in ants, moreover, its high robust and distributed feature severely strengthen the reliability and global search capabilities, which makes it a easily-combined method . However ,there are still two prominent defects, one is the searching period it takes ,the other is that it is likely to be tripped in partial optimal solution.The paper have studied the optimization problems of the transport network under two different path by improving the Ant Colony Algorithm. The main work is as follows:(1) Analysis of the optimization issue of domestic and international logistics and distribution path and research status of Ant Colony Algorithm, description of the present logistics distribution problems, reflecting the necessity of researching on the Vehicle Routing Problem .(2) Introduction of the definition, classification, common solving method and distribution vehicle routing optimization model of the optimization of logistics and distribution vehicle routing issue. Illustration of several classical improvement method and the proposition strategies of this paper: improving the pheromone evaporation coefficient and visibility to optimize the searching ability, which can avoid premature convergence and stagnation.(3) Study on CVRP under two types of distribution and transportation network circumstances, one is fully connected, the other is partially connected. As to the partially connected network, the paper has used a 0-1 matrix to indicate the connection relationship between nodes, and used the shortest path algorithm (Dijkstra matrix algorithm) to pre-process the data, then obtained the solution with combining the improved Ant Colony Algorithm, finally proved the effectiveness of the algorithm by programming under the Matlab software platform ,and realized the VRP in the network with existence of non-directly connected edge.