Research On Vehicle Routing Problem Based On Flower Pollination Optimization Algorithm

In recent years,with the rapid development of computer technology,more and more intelligent computing methods are born.As the core of intelligent computing,swarm intelligence algorithm has been the research hotspot of experts and scholars at home and abroad.As a promising algorithm,flower pollination algorithm has been successfully applied in many fields due to its characteristics of few control parameters,easy implementation and simple structure.However,similar to the traditional swarm intelligence algorithm,it still has some shortcomings,such as poor local search,slow convergence to the stable optimal solution and low accuracy of the optimal solution.In order to improve the performance of flower pollination algorithm,the standard flower pollination optimization algorithm is studied and improved from two aspects of algorithm optimization and application.In the improved algorithm of flower pollination,firstly,the variation factor of differential evolution is introduced to increase the diversity of the population,and the local search is guided by the optimal solution to accelerate the convergence speed of the algorithm;the two-stage algorithm fusion of Eagle strategy is adopted,and the standard flower pollination algorithm with Levi flight is used as the first stage of global search algorithm,and the artificial bee colony algorithm with excellent individual optimization ability is used as the second stage of local strengthening search algorithm,which makes the algorithm jump out of local optimization,so as to increase the ability of global optimization.Six test functions are used to verify the accuracy and convergence speed of the algorithm.The experimental data show that the improved algorithm is better than the standard artificial bee colony algorithm and the standard flower pollination algorithm in the accuracy and convergence speed.In the aspect of algorithm application,the improved flower pollination optimization algorithm is applied to solve the vehicle routing problem.Since the development of vehicle routing problem,many experts and scholars at home and abroad have proposed many solutions,among which the most efficient one is to optimize it through swarm intelligence algorithm developed in recent years.According to the characteristics of vehicle routing problem,the improved flower pollination algorithm is discretized,the natural number coding method is adopted,the sequence of initial solution is generated by greedy strategy,the candidate solution of global search is generated by the way of non adjacent cell inversion,and the candidate solution of local search is generated by the way of adjacent neighborhood exchange.Finally,we choose the appropriate objective function and fitness function to test the algorithm.The experimental results show that the improved flower pollination optimization algorithm can find the shortest path for the vehicle routing problem of less than 30 distribution points,and the shortest path for the vehicle routing problem of more than 30 distribution points is not found.However,the shortest path obtained by the improved algorithm is closer to the value of the optimal path than the standard algorithm and artificial bee colony algorithm,which proves the good effect of the algorithm in solving the vehicle routing problem.