Optimization Of Fleet Size And Parking Capacity Of Carsharing System Considering Adaptive Relocation

With the increasing number of private cars and the increasingly serious environmental pollution problem,the carsharing mode,as a new environmentally friendly transportation mode,has become an important construction direction of the intelligent transportation system,but it also faces conflicts of interests.How to more accurately simulate realistic carsharing system and provide more realistic decision-making for carsharing service operators is still worthy of research.Based on the mixed fluid queuing model of electric carsharing system,this paper studies the system operation strategy.From the perspective of operators,users,and the public,a mixed-integer nonlinear programming(MINLP)model is built to optimize the operation strategy for alleviating system imbalance problems.We used Nomad Algorithm(NA)and genetic algorithm to solve the model,and the optimization results of the dynamic pricing strategy based on the nonlinear elastic demand model and the user-based adaptive relocation strategy are analyzed and compared respectively,proving the effectiveness of the adaptive relocation strategy.Based on the comparison between the dynamic pricing strategy and the adaptive relocation strategy,the user-based adaptive relocation strategy is introduced into the mixed fluid queuing network,also the fleet size and parking capacity optimization model of the electric carsharing system.With and without considering the adaptive scheduling strategy,the collaborative optimization of the fleet size and parking capacity of the electric carsharing system is achieved respectively,and the electric carsharing system configuration scheme and differential pricing under adaptive relocation are given to verify the effectiveness of the model.Mesh Adaptive Direct Search(MADS)and NA algorithm are designed to analyze and compare the optimization results in the two models according to the optimization model scale.In addition,we set some numerical experiments in different scenarios,studying the effects of demand,road congestion,pricing and station standard congestion of adaptive relocation on the system optimization results.