Research On Path Optimization Of Fresh E-commerce Community Distribution Under Time-dependent Road Network

In recent years,with the rapid development of e-commerce,using e-commerce platform to purchase fresh products has become one of the important ways to buy fresh products.Fresh products have the characteristics of perishability and deterioration,and are sensitive to delivery time and transportation environment temperature.Due to the high frequency and small single purchase of fresh products,the demand for fresh products with different temperature ranges is always high,so the logistics and distribution cost of fresh products remains high.On the other hand,the change of road network state often causes the change of travel time of distribution vehicles,which affects the timeliness of fresh distribution.In this paper,fresh e-commerce community distribution as the research object,the specific research work is as follows:(1)the current research situation at home and abroad is reviewed.Most of the existing research on fresh food logistics distribution assumes that the vehicle speed is constant,and there are few studies considering the time-varying characteristics of the road network,then,the model and algorithm of vehicle routing problem(VRP)in TDVRP are introduced.Taking fresh e-commerce enterprise as the breakthrough point,this paper analyzes several different operation modes of fresh e-commerce,and analyzes the demand of fresh products in different temperature layers.(2)because of the time-varying characteristics of the road network,the travel time is influenced by the travel time.According to the collected historical speed data,the future speed is predicted based on BP neural network,and the travel time is calculated under the cross-time interval.The factors of fresh community distribution network are analyzed,and the vehicle load,distribution time limit and the maximum number of cold storage and insulation boxes are taken as constraints,an optimization model of fresh e-commerce community distribution based on the objective function of minimizing the total cost is constructed.(3)solving the model based on two heuristic.A better initial population is constructed by inserting algorithm,and an improved proportional selection method is used for selection operation.At the later stage of population evolution,the elite retention strategy is adopted to avoid destroying the good individual structure,and the crossover and mutation operators are improved to change with the fitness value of the individual and increase the diversity of the population.(4)based on the example of a fresh food distribution center serving many communities,the distribution goods contain three kinds of warm-layer fresh food.The parameters are solved by MATLAB software.The results of case study show that the improved AGA algorithm has better performance,faster convergence and can effectively jump out of local optimum Changing the vehicle travel time can effectively reduce the vehicle waiting time,on-the-way time and the total cost of distribution,enterprises can adjust the distribution center departure time according to the actual situation.There are 31 figures,9 tables and 69 references in the dissertation.