Research On Vehicle Routing Problems In Intelligent Transportation System

Transportation as the main carrier of the national socio-economic activities, plays asignificant role, it is very important in various industries circulation, it is the main artery ofthe country and the development of the national economy. However, due to the increasingnumber of vehicles, road load is growing heavier, roads blocked, traffic congestion,environmental pollution, frequent traffic accidents and other issues more and more serious.With these social issues become increasingly prominent, intelligent transportation systems(ITS) came into being, it is widely recognized as a comprehensive and effective solutiondescribed by transportation the best way to bring a range of issues.Vehicle Routing Problem (VRP) is the core point of the intelligent transportation system,plays a very important role in the logistics industry. The development of intelligent vehiclerouting optimization system not only can solve the backward logistics management in China’snational conditions, and also help solve energy shortages, traffic congestion, air pollution,which has been plagued by social problems of the people, to achieve the unity of theresources, the environment, the efficiency, to promote economic progress and socialdevelopment of the logistics industry. In this thesis, through optimized the path of thedistribution of logistics vehicles, to improve logistics and distribution efficiency as the goal,use Ant Colony Algorithm (ACA) to study the distribution vehicle routing optimization.Firstly, use the transportation logistics as background, elaborated the researchbackground, purpose and Status quo of VRP problem in this paper, point out the importanceof the VRP problem in modern logistics transport industry. Then, described the VRP problem,including the raised of VRP, the elements of the study, as well as detailed classification ofVRP.Secondly, point out the main solving method of VRP problem is heuristic optimizationalgorithm, Then introduced the four typical swarm intelligence optimization algorithm, andanalyzed the characteristics of four algorithms, disadvantages and applicability are compared.ACA is better in solving NP-hard problem by comparison of results, it is one of strongestperformance algorithm in the modern heuristic algorithm. At last, considering this thesis usesant colony algorithm as the optimization method, introduces the basic idea of the ant colonyalgorithm, the basic model and algorithm implementation steps, laid the theoreticalfoundation for the improvement of the ant colony algorithm for the later chapters.Thirdly, point out that the basic ant colony algorithm is easy to fall into local optimum,the search efficiency is low, then aiming at the disadvantage of ACA has, we improve it fromthe algorithm of path selection, pheromone update strategy, pheromone diffusion, the initial solution heuristic. And given the basic model of improved ant colony algorithm, the basicsteps of the improved ACA is g provided.Finally, in order to verify the effectiveness of improved ACA in solving the VRPproblem, this thesis selects two typical problems of standard CVRP library VRPLIB asexperimental object, program design and simulation using Matlab software. By comparing theresults with the basic ACA and analysis the results in all aspects, Verified that the improvedACA is effective, feasible in solving the VRP problem. And the main parameters of antcolony algorithm are optimized, given a set of ideal parameters.