A Bilevel Programming Model For Congestion Pricing Of Multi-user Considering Enission

With the rapid development of economy,the number of vehicles is increasing rapidly.As a result,automobile emission has become the main source of air pollution,but also brought the problem of urban congestion.Because of the restriction of land resources,the government can not only build road infrastructure,but also need to guide travel behavior through policies.Traffic demand management is particularly critical.The experience of Stockholm,London,Singapore and other cities shows that road congestion pricing is the most effective means of traffic demand management.Road congestion pricing can not only promote travelers to transfer to low-carbon and green public transport,but also regulate the distribution of traffic volume in time and space,so as to alleviate congestion and reduce emissions.In this paper,the charging strategy of congested road network is taken as the object.Firstly,the users in the road charging system are classified into two dimensions: vehicle type and time value.According to the modeling needs and data analysis,cars are divided into three categories according to the emission standards.SP survey was used to collect the preference of urban car owners for toll roads and free roads,and the time value distribution of car travelers was calculated by multiple logit models.According to the distribution,time value can be divided into three categories: high time value(45 yuan / h),medium time value(30 yuan / h)and low time value(15 yuan / h).Secondly,the emission factors of CO,HC and NOx of cars with different emission standards at different speed levels are calculated by COPERT.Using the method of regression analysis,the emission factor models with average speed as independent variable were obtained.The results show that the CO emission factor model is a third-order function,the HC emission factor model is a power function and the NOx emission factor model is a second-order function.Then,based on the results of user classification and emission factor calculation,a multi-user double-layer congestion pricing model is established.The lower level model is the user equilibrium traffic assignment under a certain charging level.In the process of network distribution,factors such as travel demand reduction and mode transfer caused by charging are considered.The difference of user’s time value is considered in travel cost calculation.The goal of the upper model is to maximize the social net benefit.The travel time cost and emission cost are considered in the benefit calculation.Finally,genetic simulated annealing algorithm is designed for the model.A simple example is analyzed.The results show that the total emission of the system is reduced by 18.21% and the total travel time of the system is reduced by 11.35% compared with that of the system without toll.