Research On Multi-stage And Multi-objective Vehicle Routing Problem Under B2C Environment

With the rapid development of information network technology and the advancement of globalization,people’s consumption patterns and consumption habits are also changing.Contemporary way of commodity trading makes the logistics distribution becoming a commodity trading’s last step.In the current fierce competition,the traditional logistics distribution cannot satisfy the current logistics demand,electricity enterprises should provide a variety of logistics services to attract customers.Faced customers’ diversified requirements,enterprises need to improve the quality of distribution equipment and the quality of service,enhance the competitiveness of the enterprise in the industry.In this paper,we put forward two kinds of vehicle routing problem with the time window according to the characteristics of logistics demand in the B2 C environment.The first kind,delivery time window of the customers had strong flexibility in the contemporary e-commerce mode.It was different from the traditional mode of distribution,during the distribution of the customers,customer demands can postpone within a certain time,and we divided the customer’s delivery time window into multiple-period and determine which customers’ needs can delay according to the time constraint conditions,which can delay to the next period.In each period customers were divided into can postponable customers and cannot-postponable customer based on penalty function.We proposed an evaluation index for postponable customers to select qualified customers,and insert them into the current delivery path.We established a multi-period dynamic model of vehicle routing problem,designed the adaptive genetic algorithm to solve the model,and finally verified its effectiveness by simulation experiments on the model and algorithm.The second kind,we studied the multi-objective vehicle routing problem.As the order’s timely distribution directly affected the customers’ satisfaction,different from the traditional to the goal of minimizing cost single objective vehicle routing problem,this paper put forward multi-objective vehicle routing model with distribution cost minimization(including vehicle fixed cost,travel cost,punishment cost),the maximization of customer satisfaction and the least number of the vehicles.We designed corresponding punishment function and the linear weighted method to deal with multi-objective function.We also designed the adaptive genetic algorithm to solve the model;finally the experimental simulation verified the validity of the model and algorithm.