Research On New Energy Vehicle Order Allocation Based On Ant Colony System

Low-carbon economy is becoming the trend of economic and social development in the world today.In the construction of low-carbon urban transportation systems,new energy vehicle(NEV)is gradually being promoted as a promising alternative to fuel vehicles due to low energy consumption and environmental protection.With the vigorous development of the Internet of Things technology,online ride-hailing service for NEVs has become a new trend.The receiving and processing of online orders are usually completed by a centralized service platform to facilitate the unified management of vehicle resources.This centralized scheduling mode provides customers with high security and efficiency but also brings a heavy computational burden to the scheduling platform.Therefore,an efficient global scheduling method is required to support the allocation of daily service requests.Online order allocation studies how to allocate suitable NEVs to a series of order requests within a specific time window in order to maximize the overall benefit.It is essentially a discrete combinatorial optimization problem(COP).As a new type of transportation,NEVs have the characteristics of fuel-driven vehicles,but the dispatch is different because of their main reliance on battery power,limited battery capacity,and long charging time.In practical transportation planning,it is necessary to consider factors such as charging facilities and battery characteristics,which brings challenges to traditional planning and design methods.In recent years,swarm intelligence algorithms,especially ant colony optimization(ACO)algorithms,have shown good global search capabilities in discrete COPs.In this paper,the application of ACO in the order allocation of NEVs in static and dynamic environments is studied.This paper first analyzes the application background and proposes an NEV dispatch framework.This framework elaborates the application mode of NEV dispatch from the underlying infrastructure to the upper-level business goals.Based on the dispatch framework,this paper comprehensively considers the remaining power and charging facilities in the NEV service process,and establishes the NEV dispatch model.In order to solve this model,a variant of ant colony optimization,namely ant colony system(ACS)algorithm is adopted as the basic optimizer for NEV dispatch.In the static dispatch environment,considering the high response efficiency requirements of order allocation tasks,a pre-selection strategy and a local pruning strategy are proposed.They can reduce the number of NEVs involved in the allocation before and during the process of the standard ACS algorithm,respectively,to ensure high-quality solutions and improve efficiency.In the dynamic dispatch environment,this paper proposes a population-based ACS approach(P-ACS).This approach saves the searched high-quality solutions to a population list,and repairs the saved solutions when the environment changes.These solutions can help quickly transfer the knowledge accumulated from the old environment to the new environment to guide the searching behavior of ants.The population list mechanism enhances the adaptability of the algorithm in dynamic environment.A series of experiments are carried out to verify the performance of the algorithm.The experimental results show that the proposed algorithm can effectively and efficiently solve the NEV dispatch problem in static and dynamic environments.It can effectively improve customer satisfaction and reduce customer waiting time at the global level.