Optimization Of Logistics Vehicle Path Based On Ant Colony-Genetic Fusion Algorithm

The route planning of logistics distribution vehicles is an important link in the process of logistics distribution,which is reasonably planned to achieve the purpose of reducing the transportation cost of enterprises,improving the benefits and promoting the development of enterprises.Since there are some differences in the solution of this problem in modern heuristic intelligent algorithm,this paper,on the basis of the improvement of the traditional ant colony algorithm and genetic algorithm,mixes the two and proposes a combinatorial optimization algorithm based on ant colony algorithm,so as to solve CVRP and VRPTW problems in VRP problem.VRP problem is a NP problem,so the choice of algorithm is very important.At present,the algorithm for solving VRP can be divided into precise formula and heuristic method.Based on the selection of ant colony algorithm and genetic algorithm,this paper analyzes and compares the advantages and disadvantages of the two algorithms respectively.It can be seen that ant colony algorithm has advantages of strong robustness and positive feedback.Genetic algorithm(GA)has the advantages of genetic algorithm(GA),such as the operation of parameter coding and non-deterministic optimization rules.Both algorithms have good random search and global search ability when solving VRP problem.This paper also puts forward an improvement strategy for a single algorithm,which combines the two improved algorithms to form a new algorithm.Secondly,the initial solution is taken as the initial pheromone needed for the operation of ant colony algorithm.Finally,the pheromone concentration on the path is changed dynamically in the later stage of the algorithm,so as to give play to the advantages of the algorithm's positive feedback and fast convergence speed,and improve the operation efficiency and solution quality of the algorithm.This paper first explains the relevant theories of the optimization of logistics distribution path,not only from the theoretical and practica l significance of the proposed research content of this paper,but also the current research status of the algorithm and VRP in the way of reference.Secondly,in this paper,the related definition of logistics vehicle routing problem is described on the basis of further analysis of the similarities and differences of the TSP and VRP,and expounds the constituent elements of VRP and classification,also analyzed the three types of algorithms to solve the problem of VRP and the advantages and disadvantages o f various algorithms,which focuses on analyzing the advantages and disadvantages of ant colony algorithm and genetic algorithm in TSP,on the basis of the VRP problem with capacity constraints and with hard time window of VRP and with soft time Windows for as constraints to clarify,and based on the constraint conditions conform to the objective function of the mathematical model is established.In chapter 3 and chapter 4,this paper describes the principles and basic concepts of GA and ACO respectively,analyzes the advantages and disadvantages of the two algorithms,the specific operation steps of the algorithm and the deficiencies of the current two algorithms,and gives the basic algorithm model of ant colony algorithm in solving the VRP problem.In the fifth chapter,this paper focuses on the improvement of the above two algorithms,and puts forward the idea of mixing the two improved algorithms,and analyzes the operation steps and flow chart of the combination algorithm.The experimental results show that the proposed hybrid algorithm is superior to the results obtained in the literature and proves that the proposed algorithm has some feasibility and effectiveness.Secondly,the R101 25 customer points in the Solomon are verified by external verification.The results show that the proposed algorithm meets the goal of low vehicle number and low cost.Finally,25 customer points of other problem types in the question bank are also tested.the simulation results show that the results obtained by the combined optimization algorithm in solving RC type problems are close to the existing optimal results.Furthermore,this paper summarizes and studies the work of combinatorial algorithms in solving VRP problems,and points out the future researc h direction of combinatorial optimization algorithms.