Research On Optimization Of Multi-Objective Urban Logistics Distribution Route Based On SA

With the acceleration of economic globalization and urbanization, urban logistics plays an important role in promoting the urbanization process. Scientific and rational planning urban distribution has great significance to improve the urban environment and economic efficiency, promote the city's economic develop healthily and sustainably. Vehicle distribution route optimization is the key to planning urban distribution. And many domestic and foreign scholars pay attention to this problem. Studying the vehicle routing problem of urban distribution is the oretically and practically significance to allocation of urban resource rationally, to improve enterprise management level, to relieve urban congestion and other areas.This paper studies the multi-objective vehicle routing problem in urban distribution, constructs a mathematical model to minimize the vehicle mileage vehicle routing problem. And this issue being research and analyzed from the model-building and solving.Firstly, study the article research background and the significance, summarized the research status of vehicle routing problem and the method to solve the VRP, analyze the existing research shortcomings. And then this study presents the innovations. Then this paper discusses the concepts of urban logistics, the main flow of city distribution, the concept of the vehicle routing problem(VRP). And classify the VRP according to the research focus, summarizing the method of solving the VRP problem commonly used.Secondly, based on the general model of vehicle routing problem, this study elaborated service time window concept and classification, and build the vehicle routing problem with time window sincrease service time constraints in VRP. And then described uncertain road conditions in urban distribution, then it introduced in VRPTW, build uncertainty traffic conditions with VRPTW. The traditional simulated annealing algorithm(SA) has been able to solve these model problems. And this study discussed the basic idea, components and features about SA. But the traditional SA will affect the efficiency when confront with larger data size. Thus a series of improved methods in vehicle routing problem for large scale data will proposed, mainly includes geographic information systems, SPSS, parallel search, increase memory function, repeat the search methods and so on. And the implementation process of improved simulated annealing algorithm will be described, it made good bedding for research and analysis of subsequent instances.Finally, the chapter to solve vehicle routing problem with time windows(VRPTW) and uncertainty traffic conditions with VRPTW using the improved simulated annealing algorithm. And analysis the optimization process of specific issues and selection process of key parameters in algorithms. Through computer programming to solve the specific issues, and the results from both quality solution and solving time were analyzed. The results showed that the constructed new model is closer to the actual situation, while improved SA contributes to vehicle routing problem large-scale application in actual logistics.