Research Of Logistics Distribution Vehicle Routing Optimization Based On The Primary And Secondary Population ACO

The rapid development of global economy has been driving the progress of the logistics industry. In the logistics distribution, distribution cost is an important part which people cannot ignore. In order to maximize cost savings, the problem of vehicle routing optimization in logistics distribution is an especially important.Firstly, This thesis introduces the definition and classification of logistics distribution, and also the basic method to solve the logistics distribution path optimization. Secondly, This thesis introduces the basic ant colony algorithm and the improved ant colony algorithm. finally, with the improved ant colony algorithm, we can solve the vehicle routing optimization problem in logistics distribution, achieved the purpose of saving cost.This thesis works mainly in the following several aspects:(1)This paper study the logistics distribution routing optimization problem and the current research situation on ant colony algorithm at home and abroad. this thesis introduces some existing problems in the logistics distribution, explain the importance of the vehicle routing optimization problem research.(2)And it introduce the related theory knowledge and the commonly used algorithm in logistics distribution vehicle routing optimization problem briefly. Research and analyze on the algorithms of basic ant colony, optimal-worst ant colony, maximum-minimum ant colony, primary-secondary population ant colony, to solve problem of slow preliminary pheromone accumulation in primary-secondary population ant colony, the thesis improved the algorithm with path crossover strategy(2-opt) and saving strategy respectively. On the matlab platform, it apply the maximum-minimum ant colony algorithm and primary-secondary population ant colony algorithm to simulate the Oliver30 TSP problem. Verification that the primary-secondary population ant colony algorithm is better than the maximum-minimum ant colony algorithm in the perfect route, And on the Matlab platform, the advantages of primary-secondary population ant colony with Path crossover strategy and primary-secondary population ant colony with saving strategy were verified separately. Draw a conclusion that path crossover strategy method is better than saving strategy method.(3) Take the typical instance eil22 in VRP problems library as an example. analysis and establish the mathematical model for vehicle routing optimization problem with capacity limits(Capacity Vehicle Routing Problem, CVRP). Apply the improved primary-secondary population ant colony algorithm to study the VRP problems, and complete the vehicle routing simulation of VRP on Matlab software platform.(4) Study and analysis the specific situation of the logistics distribution of an express company in shaanxi xi ’an, use the improved primary-secondary population and VRP model in this thesis to solve the express company’s logistics path optimization problems.