Electric Vehicle Routing Problem With Dynamic Demand Considering Time Of Use Electric Pricing System

In recent years,since living standers are higher,people’s purchasing power is steadily increasing.Delivery quantity is also increasing rapidly with the development of e-commerce.The traditional distribution tasks are mainly completed by fuel vehicles.However,fuel vehicles have the disadvantage of high fuel costs and fuel vehicle emissions is the important cause of environmental pollution.The state increasingly advocates the development of green logistics,electric vehicles have been vigorously promoted.Compared to fuel vehicles,electric vehicles are favored by more and more enterprises because of lowing operating costs and energy conservation.Electric vehicles have been started to replace fuel vehicle.But the research of Electric routing problem is still in the immature stage.Considering the charging behavior of electric vehicles,the characteristics of dynamic demand,it is difficult to solve the problem with the exact algorithm.And now there are a lot of idle vehicle idle in the society.The dynamic vehicle routing problem of EVs considering crowdsourcing has not been studied.Based on the existing research background,this paper studies the electric dynamic vehicle routing problem based on the adaptive large-scale neighborhood search algorithm.First of all,according to the electric vehicle routing problem,the electric vehicle routing problem considering time of use electric pricing system is presented.A two-stage integer programming model is proposed to minimize the total distribution cost.An improved adaptive large-scale neighborhood search algorithm(IALNS)is developed with the new destroy and repair operators and acceleration strategy in the dynamic stages.Extensive large-scale computational experiments with both static and dynamic demands are conducted to examine the performance of proposed IALNS and compared algorithms which contain variable neighborhood search algorithm(VNS).and large-scale neighborhood search algorithm(LNS).The IALNS performs best.Secondly,on the basis of the above problem,considering the insufficient capacity of distribution enterprises and the existence of idle capacity in society.A two-stage integer programming model of electric dynamic vehicle routing problem based on crowdsourcing and time of use electric pricing systems is proposed.The tabu search algorithm is combined with the improved adaptive large-scale neighborhood search algorithm to solve the crowdsourcing delivery model.Finally,the results of the crowdsourcing distribution model are compared with those of the common distribution model.The experimental results show that the crowdsourcing delivery model has advantages over the common distribution.