A Hybrid Multiobjective Evolutionary Algorithm For The Vehicle Routing Problem With Time Windows Application

Vehicle routing problem with time Windows (VRPTW) is a complicated multi-objective optimization problem with significant economic benefits. For this problem, the traditional multi-objective evolutionary algorithm makes a poor performance on the convergence of the solution sets, how to improve the degree of convergence of the solution set is a research hot spot and difficult point in the field.This paper presents a hybrid multi-objective evolutionary algorithm(HMOEA) mixed with global search, the research content focuses on how to improve the crossover operator and how to mix the algorithm, to increase the degree of convergence of the solution set. And having intense research for the effect of the improved algorithm for solving the multi-objective Vehicle routing problem with time Windows. Test data set experimental results verify the effectiveness and superiority of the algorithm, the superiority of the algorithm is more obvious in wenzhou city tobacco distribution route optimization problem.This paper main research work and achievements are as follows:1. In algorithm hybrid process, draw on the idea of path construction and path improvement in the classic heuristic algorithm. Insert the Push Forward Heuristic Algorithm (PFIH) in the initial population structure, increasing the proportion of feasible solutions to improve the efficiency of the algorithm. Probabilistic global search of different grades non-dominated solutions after fast non-dominated sorting, search for improved solutions in the solution fields to avoid the algorithm falling into a local optimum, setting the probabilistic search, can reduce the algorithm operation time.2. In algorithm improved process, based on the deficiencies of route exchange cross, introduce eliminate route operation, reducing the number of routes increases, increasing the total distance decreased intensity. For the algorithm overall performance, improved cross operator and global search performance comparison, the simulation results show the effectiveness of the algorithm improvements and mixed.3. In the application of algorithm, the primary algorithm and improved hybrid multi-objective evolutionary algorithms are applied to the wenzhou city tobacco distribution route optimization problem, the result obtained by improved algorithm is better than that of outsourcing software company and the primary algorithm, effectively reduce the total distance and improve the vehicle loading rates, the customer point of distribution roadmap also appears orderly.