Modeling And Optimization Of Mobility On Demand System Under Technological Integration

With the rapid development of vehicle automation,connection,electrification,and sharing,the integration of these technologies,i.e.,the Automated Connected Electric Vehicle sharing system is a potential way to resolve problems such as traffic congestion,air pollution,traffic accidents,and energy shortage.Also,it can greatly improve the safety,mobility,accessibility and environmental protection of transportation systems.However,there are still many problems in the practical application of the current level of ACEV sharing system,which needs further research.In view of this,this paper firstly summarizes and analyzes the key issues of the application of ACEV sharing systems.After that,the modeling methods for the determination of the operational design domain of service vehicles and the operation of ACEV sharing system,and the optimization methods for the planning and operation of ACEV sharing system are given.Finally,the comprehensive experimental analysis of the ACEV sharing system is carried out.(1)Modeling and complexity quantification of vehicle operation design domainIn order to accurately describe the operating environment of the automated driving system of the service vehicles,a specification-based operation design domain modeling method is proposed with the consideration of the uncertainty,complexity and dynamic characteristics of the actual road traffic environment.Also,a traffic scenario-based complexity analysis method of operation design domain is deduced.Eventually,the proposed method is verified by the application of Changchun Jingyue National High-tech Industrial Development Zone as a research area,which is divided into five complexity levels.(2)The operation modeling of ACEV sharing systemAfter the introduction of automated driving technology,V2 X technology,and electric vehicle technology,the attribute characteristics of the main components of the sharing system,as well as the overall operation process will change.Moreover,in the case of considering the operation design domain of the service vehicles,it is necessary to add the operational constraints of the service vehicles in the sharing system.Therefore,in order to accurately describe the SAE L4 ACEV sharing system considered in this paper,it is necessary to model and analyze the key characteristics and operational processes of the sharing system.(3)Two-stage multi-objective planning model of the ACEV sharing systemOn the basis of the operation modeling of ACEV sharing system,a two-stage planning model of the sharing system is constructed.To cope with the discrete and unbalance vehicle distribution within the service area of the free-floating carsharing system,a service area optimization model is proposed in stage 1,which can balance the system operational complexity/cost and the user benefits.Faced with the long recharge time and short vehicle range problems of the current electric vehicle technology,a charging infrastructure deployment optimization model is proposed in stage 2,which can achieve a balance between the vehicle charging demand and the infrastructure cost in the sharing system.Then,based on the characteristics of the two-stage planning model of the sharing system,a solution algorithm based on NSGA-II is designed.Finally,the model is verified through application in a real-life case.The system can meet most of the user trip requests,but the service mileage of the service vehicles in the system and the waiting time of the users are rather long.(4)Operation optimization model of the ACEV sharing systemAfter the planning of the ACEV sharing system is completed,considering the dynamic and uncertainty characteristics of the user trip demand and vehicle charging demand in the sharing system and the close relationship between the two kinds of demands,a IP-based jointly optimization model for the service vehicle allocation problem of the user trip demand and the charging station allocation problem of the vehicle charging demand is proposed to improve the operational efficiency of the sharing system.Then,based on the characteristics of the system operation optimization model,a tabu search-based solution algorithm is designed.Finally,through the simulation experiments with the system operation optimization model,it is found that the user trip service performance and vehicle charging performance have been significantly improved after adopting the demands assignment optimization model.(5)A comprehensive experimental analysisAfter the operation design domain,the model representation,the system service area and the charging infrastructure deployment planning,and the operation optimization scheme are all determined,a comprehensive simulation experiment analysis is carried out.The simulation results show that the service vehicle cruising range,charging facility charging speed,ACEV technology,service vehicle fleet size,and user trip demand scale has a significant impact on the user trip service performance and vehicle charging operation performance of the system.In particular,the increase of the charging speed can effectively improve the efficiency of the vehicle charging operation.In terms of environmental impact,the adoption of ACEVs in the system can reduce carbon dioxide emissions by about 42.03% and energy consumption by about 31.34% compared with conventional fuel vehicles.Finally,when the operation design domain complexity level of the sharing system is Level 5,the user usage cost per kilometer is about twice as high as that is Level 1.