Research On Traveling Salesman Problem In Logistics Lending Platform

With the constant collision of "Internet Plus" and the traditional library industry,the model and level of library service have undergone tremendous changes.The traditional borrowing model which wastes too much time and unnecessary work has not been able to meet people's needs satisfactorily.Therefore,this article proposes the concept of "logistics borrowing" and designs and implements the platform.Logistics distribution is the last and most essential part of the logistics borrowing platform.The design and implementation of it is closely related to user experience and the distribution cost of the library.Therefore,the selection of the most proper path of logistics distribution is in the central position of the current thesis.To design the most reasonable distribution route reasonable with minimum total cost and roughly the same workload of the staff,the Multi-Objective Multi-Traveling Salesman Problem(MOMTSP)is put forward in the study based on situations in real life.The MO-MTSP combines Multi-Objective Optimization and Multi-Traveling Salesman Problem.Because of its complicated constrains and wide range of feasibility,there is not so many researches on MO-MTSP as classical TSP problem.Ant colony algorithm as a classic algorithm for solving TSP problem.It works out the optimal solution through pheromone evaporation mechanism and heuristic information induction.However,it is not suitable for MO-MTSP problem.The current paper proposes an improved Ant Colony Algorithm to solve the problem mentioned above.The base of the improved one is based on the classic Ant Colony System(ACS).It improves the ant setting so that the ants can carry more information to complete the work and the stochastic initialization of the pheromone strategy enables a wider range of research field to enhance the possibility of working out the optimal solution.Then,by introducing the Selective Travelling Salesman Strategy and introducing Local Search Operator,the algorithm can better balance the total cost and the "amplitude" target and effectively improve the quality of the solution.Finally,the Pareto optimal front is filtered through the Pareto filter,and the TOPSIS method is used to find the optimal solution for the logistics borrowing platform.Through several sets of comparison experiments,it is proved that the improved Ant Colony Algorithm has a strong search ability and can find the optimal solution that meets the actual needs of the current thesis.This article details the platform's architecture,overall design,and functional implementation.The platform employs Baidu map API technology to generate distance matrix between distribution locations to improved Ant Colony Algorithm to optimize the book distribution path.The platform includes a background management system and a foreground borrowing system.The background management system implements functions of book management,order management,personnel management,authorization management,operation management,statistical reporting,and settings.The foreground borrowing system is realized by a mini program in Wechat,including the functions of homepage,books,shopping carts,order management,and personal center.