The Study On Travel Behavior And Congestion Tolling Strategies Of Multi-bottleneck Model

The vehicle ownerships grow fast with the rapid development of urban economy in city,which make incisive contradiction between traffic supply and demand.Under the circumstance,many cities have suffered traffic congestion problem,especially the morning rush hour commuter problem,in which massive commuters choose the same departure times,has attracted considerable attention.It is generally accepted that congestion pricing,as an economic strategy of traffic demand management,plays an active role in alleviating traffic congestion.Thus,explaining and forecasting commuters' travel behavior is one of the basic researches.Particularly,the bottleneck model is widely viewed by researchers as the most efficient models because it formulates the commuters' travel behavior,with all its advantages of clarity and efficiency.However,most of the literatures focus on the single bottleneck model and its extension,and lack of congestion pricing in multi bottleneck network.In contrast to the large body of theoretical work on roadway bottlenecks,there is relatively little empirical work addressing similar questions about the bottleneck model and its extensions using real data.In view of this,this paper firstly analyzes the causes and effects of traffic congestion,then explains the concept of road congestion charges,and briefly summarizes the classical single bottleneck model and different congestion tolling strategies.Secondly,based on the perspective of multi bottleneck network,time-varying toll model and single step toll with tactical waiting model are constructed for Y traffic network.The commuters' departure time choice is assumed to follow the user equilibrium(UE)principle.The equilibrium departure rate and individual trip cost under different merging rules are derived.According to commuters' departure time choice behavior,the optimal charging period and toll level are derived.Our results show that the time-varying charging scheme can completely eliminate the queue congestion at each bottleneck,but may make the commuter's individual travel cost larger;At the same time,the proposed single step toll scheme can indeed improve the efficiency of the traffic network in terms of decreasing the total system travel cost,but may lead to increase of individual's trip cost.In addition,the effectiveness of the proposed pricing scheme not only depends on merging rules,but also depends on the relative difference of the capacities of the upstream and downstream bottlenecks.Finally,Braess paradox isobserved in the Y-shaped traffic network: the total system travel cost can be increased when the capacity of the upstream bottleneck increases.Thirdly,this paper study evolution of departure decisions in a multi-player online experimental network game that emulates the single bottleneck model with different congestion tolls.Experimental results show that: there does not appear any convergence to any theoretical equilibrium in our experiment.Furthermore,compared with the theoretical model,the average travel time and travel cost are higher.Two groups of controlled experiments show that congestion toll affect the travel time decision of participants and can effectively reduce queuing waiting time and the total travel cost of the system.And in terms of players' decision making behavior;almost all players shift their departure times at least in half the number of rounds.The shift is observed to be caused by possible factor that a reduction in individual's travel cost in order to improve payoff.