Electric Vehicle Routing For End-of-line Scheduled Delivery

With the rapid development of the new retail model,the demand for terminal reservation and distribution is also increasing.Different from traditional time window delivery,end-of-line scheduled delivery mostly provides merchants or delivery companies with fixed delivery time periods for customers to choose.On the other hand,electric vehicles are widely used in terminal distribution due to their green environment and low driving difficulty.At the present stage,the main problems in the route planning of electric vehicles for terminal reservation and distribution are the lack of scientific basis for the distribution plan,resulting in the waste of distribution costs and distribution resources;In addition,electric vehicles have limited energy and long charging time,and unscientific distribution plans are easy to make the distribution unable to proceed smoothly,affecting customer satisfaction.This thesis first establishes the problem model.Since the electric vehicle energy can significantly affect the vehicle path planning,the electric vehicle energy consumption model is determined first.The energy consumption model established in this thesis is different from the traditional energy consumption model that only considers the relationship between driving distance and energy consumption in the past,but determines the energy consumption model of electric vehicles considering the driving speed and vehicle load according to the conversion relationship between "battery energy-electric energy-mechanical energy" of electric vehicles and the impact of load and speed on the energy consumption of electric vehicles,and establishes a time window constraint based on the energy consumption model combined with the terminal reservation distribution path planning problem,Mixed integer mathematical model of vehicle load constraint,vehicle energy consumption constraint and path constraint.Secondly,according to the characteristics of NP-hard,an improved brainstorming algorithm is designed and proposed,which combines large-scale domain search and simulated annealing algorithm with brainstorming algorithm.In order to better enhance the search ability of the algorithm,an elite strategy is proposed,that is,local search operation is carried out for the top 50% of the excellent individuals.This operation uses the idea of "destruction" and "repair" in the large-scale neighborhood algorithm and the idea of simulated annealing algorithm to improve the global search ability of the operation.Taking the dataset provided by Cwioro et al.as an example,taking the dataset of 50 and 100 scales as an example,the algorithm is compared with the traditional brainstorming algorithm,ant colony algorithm and simulated annealing algorithm.The result shows that the improved brainstorming algorithm can get better distribution scheme in a shorter time than other control algorithms.Finally,take the booking and distribution service at the end of Shijiazhuang B Supermarket as a case for analysis.Select 50 customers in a certain day for path planning.In the data preparation stage,use the superior and inferior solution distance method to select the algorithm parameters in this thesis.Run the improved brainstorming algorithm in MATLAB to solve the problem of the energy consumption model considering load and speed,and the traditional energy consumption model,and use LTMap to generate the actual distribution route map.By comparing and analyzing the results of the two models,it is concluded that the distribution scheme obtained by the energy consumption model considering load and speed is more in line with the actual operation situation than that obtained by the traditional energy consumption model.For small scale,short distance and low load cases,the traditional energy consumption model can be used to solve.This conclusion provides a scientific basis for the path planning of terminal electric vehicles.Finally,the sensitivity analysis of the time window shows that the size of the time window can affect the distribution plan,and the time window will narrow,the distribution distance,the distribution vehicles and the distribution costs will increase.This conclusion provides a theoretical basis for setting the time window.This article has 27 figures,15 tables,and 99 references.