Research On Electric Vehicle Routing Problem

Research on electric vehicle routing problem in the traditional logistics distribution is a new perspective, it will have a great effect on solvingcurrent energy shortage, air pollution and traffic congestion.Electric vehicle routing problem with limited battery capacity and necessitate visits to recharging stations during delivery tours is very different from traditional fuel vehicle's, thus the theories and methodsfortraditional vehicle routing problemare no longer suitable for electric vehicles.On one hand, it's still in the early stage of large-scale application for electric vehicles, the existing research on routing optimization models and methodsconsidering the futures of electric vehicle israther weak. On the other hand, theinfrastructure facilities construction of electric vehicle's popularization and application is still in the primary stage, which hinders the application of electric vehicles in a certain degree. Thus it's necessary to research on electric vehicle routing problem, it can provideguidance for the construction and operation of charging facilities, promoting the large-scale popularization and application of electric vehicles.This paper is on the research of the traditional vehicle routing problem basis to study electric vehicle routing problem with recharging stations. Firstly, a mixed integer programming mathematical model of EVRPRS problem is proposed, the load limit and service time windows of traditional limitation are considered, also with battery capacity, recharging station and recharging time of specific limitations in vehicle routing problem are considered in this model. A modified particle swarm optimization algorithm is proposed to solve this problem. Specifically, we use a three-dimensional particle coding method to combine the particle and the problem solutions. Tests performed on instances demonstrate the positive effect and the high performance of the algorithm we proposed, and can provide guidance for suitable routing planning for logistics companies.As there are many parameters involved in the basic model of electric vehicle routing problem, some parameters in instances testing are set as per relevant literatures and personal experiences, which is lack of strict scientific basis. This paper conducts multiple tests to make sensitivity analysis for time windows, vehicle cost, load capacity and battery capacity, related test results can provide reference information for logistics companies.As for the logistics company choose to establish multiple depots in the customers concentrated areas to meet the delivery requirements for the wide distribution and large-scare customers, this paper extends the basic EVRPRS model and proposes the model of multi-depots electric vehicle routing problem. Particle swarm optimization algorithm is used to solve this model with two solution strategies of multiple depots optimization simultaneously and the optimization of multiple depotstransfer to single depot. Specifically, for the optimization of multiple depots transfer to single depot, the hierarchical clustering method of average linkage is presented to classify the customers into the corresponding depots. Simulations on instances for the two strategies demonstrate that the optimization of multiple depots transfer to single depot has higher effect on solving the medium to large scare multi-depots electric vehicle routing problem.