The Prediction Model Of Traffic Breakdown Probability Based On Lane-Changing Behavior

This paper analyzes the operational characteristics of traffic flow at the bottleneck of urban expressway and explores the impact of lane-changing behavior on traffic flow operation at the bottleneck,and finally,establishing a breakdown prediction model based on lanechanging behavior.Based on the moving bottleneck model,the influence of the lane-changing vehicle in the target lane is analyzed.It is considered that when the speed of the lane-changing vehicle is less than the speed of the target lane,this vehicle would become a slow one after changing,which would form a moving bottleneck and affects the operation of other vehicles in the target lane.The simplified lane-changing behavior was simulated by the Full Velocity Difference model in this paper.The study found that when there is a lane-changing vehicle in the simulation,the speed difference between vehicles is larger,the disturbance changes more drastically,and the traffic flow is more unstable.The simulation shows that lane-changing behavior is an important factor that causes the traffic operation at the bottleneck to change from steady state to unsteady state.It is proved by the catastrophe theory that the operation of the traffic flow through the bottleneck can be described by the cusp catastrophe model,that is,“jump” of speed that occurs when the traffic flow passes through the bottleneck is not an accidental phenomenon.It is an inevitable result of the change of state variable with the control variables when the traffic flow in an unsteady state,in other words,the change of the traffic velocity at the bottleneck is abrupt.If the traffic flow of the bottleneck is considered as a platoon composed of vehicles,the change of the speed of the vehicle in this platoon can be regarded as a discontinuous process.Based on the above analysis of the lane-changing behavior and traffic flow characteristics at the bottleneck,a model for estimating breakdown probability considering lane-changing behavior is established.This model consists of a platoon formation model and a speed transfer model which based on Markov chain.The model introduces lane-changing behavior by using different speed distributions on different lanes,and it also proves that the speed transfer of lane-changing vehicles in the platoon has Markov property.Therefore,the applicable range of the probability prediction model is extended from one lane to two lanes.The validity of the model is verified by simulation data based on VISSIM.It is proved that the model can deal with the randomness of traffic flow and the heterogeneity between vehicles.It is considered that the model can be applied to urban bottlenecks.The model proves the abrupt change of breakdown at the bottleneck by theoretical calculation: when the flow rate exceeds the critical threshold,the probability of breakdown increases significantly.The result also show that the lane-changing behavior will increase the heterogeneity of vehicle speed in the lane,which will increase the formation probability of platoons,and then increase the probability of the traffic breakdown.The establishment of the model can help traffic decision-makers to better understand the traffic breakdown and provide a theoretical basis for more reasonable improvement of urban expressway capacity.