Modelling lane changing behaviors for bus exiting at bus bay stops considering driving styles: A game theoretical approach

Abstract

Bus is one of the most important travel modes for citizens, but lane changing behaviors for bus exiting at bus bay stops have negative impacts on traffic flow operations. It is necessary to investigate the characteristics of drivers’ behavioral decision-makings in lane changing for bus exiting for improving public transportation operations. The interactions between passenger car drivers in target lanes and bus drivers with different driving styles (e.g., aggressive or conservative style) are analyzed based on the game theory. A two-player, non-zero-sum, non-cooperative game model with incomplete information is formulated to examine the impacts of driving styles on the lane changing behavioral decision makings. The bi-level programming approach is applied to calibrate the parameters in the formulated model with solving the Bayesian equilibria and minimizing the prediction errors. The vehicle trajectory data extracted from Lvboqiao bus bay stop, Dalian, China are used for model calibration and validation. The results reveal that the formulated model has 73.01%, 86.60% and 83.54% accuracy rates on average for predicting the passenger car, aggressive and conservative bus drivers’ behaviors, respectively. At a 95% confidence level, the time error of a lane changing behavior is statistically significant by the paired t-test. Compared with the previous research, it can be proven that the consideration of driving styles is helpful to deeply explore the interactions between passenger car and bus drivers’ behaviors. The proposed model provides decent theoretical support for the drivers to make better decision-makings, and improves the traffic safety and efficiency around bus bay stops.