Study On Multi-Objective And Multi Depot Path Optimization Of Fresh Agricultural Products

In recent years,due to the upgrading of consumption concept,consumption patterns,urban and rural residents is an increasing demand for high quality fresh agricultural products;At the same time,the popularization of the Internet of Things,Beidou navigation and other technologies enables logistics enterprises to obtain more information about the road network in the process of distribution,so as to improve the whole distribution process.In the distribution process of fresh agricultural products,logistics enterprises need to establish a number of cold chain logistics centers and cold storage for transportation and storage of fresh goods.However,the current domestic fresh agricultural products in the circulation process there are still large energy consumption,high distribution cost,low freshness of products.Therefore,this paper studies the distribution of fresh agricultural products under time-varying conditions.The main research contents of this paper include:(1)Firstly,analyze the development status of fresh cold chain logistics industry,discuss the necessity of topic selection,summarize the research results of cold chain logistics,multiobjective VRP,multi-depot VRP,VRP under time-varying conditions at home and abroad,propose the research content of this paper,and build the basic framework of the paper.(2)Subsequently,the paper expounds the relevant concepts of vehicle routing problem for fresh agricultural products distribution,combs the commonly used heuristic algorithm,traditional heuristic algorithm and accurate algorithm,and analyzes the advantages and disadvantages of these algorithms.(3)Analyze the common demands of fresh logistics enterprises and customers,and divide the departure speed of distribution vehicles in different time periods considering the complexity of urban road network structure.The T-MO-MDVRPTW model is constructed on the basis of the time-efficiency,perishable and carbon emission indexes of cold chain logistics distribution,which minimizes the sum of six modules of distribution cost and maximizes customer satisfaction,including vehicle use cost,transportation cost,soft time window penalty cost,refrigeration cost during transportation and unloading,carbon emission cost and cargo damage cost.In order to solve the logistics enterprises how to reasonably arrange the urban cold chain logistics scheduling,to meet the needs of customers in different geographical locations in the city.(4)For the T-MO-MDVRPTW model constructed in this paper,a two-stage algorithm is constructed to solve it.In the first stage,k-means algorithm is used to divide customer points and logistics centers.The second stage: NSGA-Ⅱ algorithm was selected as the solution algorithm.The C-W saving algorithm was introduced into NSGA-Ⅱ algorithm to construct the initial solution of the problem.The NSGA-Ⅱ algorithm was improved by improving the mutation operator and congestion distance calculation method,and the feasibility of the improved NSGA-Ⅱ algorithm was verified by Solomon.Finally,the effectiveness of the twostage algorithm for solving T-MO-MDVRPTW model is verified by a multi-yard example.(5)Through the case study of cold chain logistics distribution,the rationality of the model constructed in this paper is effectively tested.In cold chain distribution,when considering timevarying factors,logistics enterprises can reasonably arrange the departure time of vehicles,which can not only improve customer satisfaction,but also reduce the distribution cost.