Exploring the Effects of Pedestrian Red Crossing Behaviors on Traffic Dynamic Based on a Simulation Model

The prime objective of this study is to investigate the effects of pedestrian red crossing behaviors on vehicular traffic dynamic by developing a reliable simulation model that can represent the interactions between the vehicles and pedestrian violators. The simulation model makes an integration of the typical NaSch (Nagel-Schreckenberg) model and SFCA (Social forces- Cellular automaton) model for vehicle and pedestrian movement. An extension for pedestrian red-crossing behaviors is also implemented in this model by dividing pedestrians into two types (i.e., law-obeying ones, opportunistic ones). The proposed model is calibrated with detailed behavioral data obtained from field observations in the city of Nanjing, China. The comparison results about the pedestrian crossing times between the simulation and real-world date validate the reliability of the proposed simulation model. Meanwhile, the results of simulation experiment indicate that the average vehicle delay is 25.30s when there are pedestrians violating the traffic light while it is 22.50s when all the pedestrians obey the traffic rule. The time-space diagrams for vehicular traffic around the signalized crosswalk reveal that a majority of vehicles stopped by the violators experience short delay while a small portion of vehicles are delayed to wait for a next green phase. The proposed simulation model can serve as a practical tool for pedestrian and vehicle traffic dynamic investigation as well as for pedestrian safety assessment and traffic performance evaluation with proper definition of related indicators.