Research On Vehicle Routing Problem With Soft Time Windows Simultaneous Delivery And Pickup

With the rapid development of the economy,especially the rapid rise of e-commerce,the logistics and distribution industry as an important part of e-commerce has become more and more competitive,and the vehicle path planning problem as one of the key issues has received widespread attention.In the delivery service,there is usually a delivery time limit,and the previous research on the problem only considers the pickup or delivery,which will cause the waste of delivery vehicle transportation capacity.The routing planning problem of simultaneous delivery and pickup vehicles gives full play to the transportation capacity of vehicles,improves the distribution efficiency of logistics enterprises and reduces distribution costs,which is currently emerging.Based on this,this paper carries out the following research:1.Based on the summary and analysis of the current situation of the vehicle routing planning problem at home and abroad,we study the relevant solution techniques in the problem,master its principles and related codes,and compare and analyze the solution methods through cases to lay the foundation for the final solution of the vehicle routing planning problem with soft time window simultaneous delivery and pickup.2.A hybrid genetic algorithm based on K-means clustering is proposed for the problem of low efficiency of traditional genetic algorithm to solve the vehicle routine planning problem with soft time windows.The algorithm firstly performs K-means clustering analysis of customer groups based on their geographical locations and time window characteristics attributes to generate initial feasible solutions to improve the quality of the initial solutions and facilitate the convergence speed of the algorithm;Secondly,the global search algorithm in the global search phase is improved by adopting the selection strategy of retaining good individuals,the partial matching crossover strategy,and the inversion variation strategy for global search to obtain better individuals;Meanwhile,a two-stage variable neighborhood search strategy based on three neighborhood operations is designed to improve the local search ability of the algorithm;Finally,simulation experiments are conducted by Solomon’s algorithm to verify the performance of the algorithm in this paper.3.An improved brainstorming optimization algorithm is proposed for the routing planning problem of simultaneous delivery and pickup vehicles with time windows constraints,with the optimization objective of minimizing the sum of fixed cost,travel cost and penalty cost of violating constraints for distribution vehicles.The algorithm introduces a large-scale neighborhood search algorithm into the local search strategy of the brainstorming optimization algorithm,and performs a two-stage optimization search.In the global search phase,on the basis of the description of the encoding,decoding and construction of the initial solution of the algorithm,the best individuals are selected according to the fitness value,and the individuals are clustered and analyzed with the best individuals as the clustering center,and then the variation and crossover strategies are introduced for global search;In the local search phase,the population objective function values are arranged in ascending order,and the top 60% of individuals are selected as the local search objects,the local search phase introduces the idea of destruction and repair in the large-scale domain search algorithm,the destruction operator is designed to delete some customers from the current solution,and then the repair operator is designed to reinsert the deleted customers into the destroyed solution;Finally,the model and algorithm are simulated with the international public data set to verify the effectiveness of the algorithm in solving the vehicle routing problem with soft time windows simultaneous delivery and pickup.4.Based on the above work,the Flexsim simulation software is used to simulate and solve the actual case,and the results obtained are compared with the results of the algorithm proposed in this paper,and the results show the effectiveness of the algorithm in this paper.