Bus Entering Lane-changing Behavior And Its Impacts

Urban public transport is an important way for urban residents to travel,and plays an important role in urban road traffic.In recent years,the study on the lane changing behavior of buses has become more in-depth.The behavior of bus entering lane change(BELC)is an important part of it,which attracts more and more attention.BELC tends to cause chaos and frequent traffic congestion near the bus station.In order to explain in more depth the traffic conditions within a certain range of the upper area of the bus station and to reveal the effect of the BELC behavior on road traffic,this paper mainly focuses on internal rules of the BELC and its impact on the external environment.In order to describe BELC behavior,firstly,its process and the four main links in the process were analyzed,including waiting for lane change,conducting lane change,lane change interfering,lane change done,explaining the characteristics of each link.Secondly,a survey plan was established to investigate the BELC behavior.The lane distribution,departure distance distribution,and lane change time distribution of BELC were analyzed by using the collected and processed survey data.Then,based on the BP neural network model,the prediction model of BELC point was established,and the traffic volume,the number of buses and the distance from the station were taken as the main factors affecting the BELC.Taking these three factors as input variables,a BP neural network model with the number of BELC points in each upstream segment as output was established.The sensitivity analysis of the input variables in the model was performed by using the Weighted Product method.The results showed that the traffic volume,departure distance and the number of BELC points were negatively correlated,while the number of buses was positively correlated with it,among which the sensitivity coefficient of departure distance was the largest.By using the principal component analysis method,a multivariate linear regression model was established for five factors,including headway distance and lane change time,number of lane changes,departure distance,number of lanes to cross,traffic volume,etc.The model test results showed that the model was well fitted and passed the test.At the same time,the model was compared with the empirical regression model.The comparison results further proved that the vehicle headway time effect model based on principal component regression analysis had good validity.Based on the analysis of the principal components,the main influencing factors for the headway time at the upstream of the bus station were the lane change time,traffic volume,number of crossing lanes,and distance from the station,respectively.Meanwhile,the degree of change in the number of lane changes was not obvious.At the same time,the lane change time,the number of lane changes,the number of lanes to cross,the traffic volume and the headway time were all proportional to each other.On the contrary,the distance from the station was inversely proportional to the headway.Finally,for the problems of BELC leading to lower traffic capacity at the upper reaches of the bus station and an increase in headway time during peak hours,three management measures were proposed,including setting the bus lanes for BELC,the presignal for BELC,and the prohibited BELC line.The definition and setting methods of these measures were described.The improvement effect on the surrounding traffic after the installation was analyzed,which could be used to improve the traffic capacity of the bus station.It provided theoretical basis and technical support for the quality of public transport services,the enhancement of urban transport image,and the promotion of urban development.