Traffic Safety at Signalized Intersection Under Heterogeneous Flow Conditions with Red Light Violation and Lane-Changing Effects

Abstract

Traffic accidents are now a major road safety issue. A significant proportion of these accidents occur at intersections, mainly in urban areas, due to failure to obey traffic lights. Modeling and simulation appear to be effective tools for identifying the causes of accidents and proposing appropriate prevention strategies. This study introduces a physics-based and computational modeling framework to characterize how accident probability $P_{a c}$ varies as a function of traffic conditions at a signalized intersection. Numerical simulation results show that the relationship between $P_{a c}$ and the number of vehicles is nonlinear. The accident probability reaches a maximum in the freeflow traffic regime and subsequently decreases, while in congested conditions $P_{a c}$ stabilizes at an approximately constant value. Furthermore, the risk of collision increases with the increasing probability of lane changes $P_{c h g}$, particularly when approaching intersections. In addition, high traffic speeds contribute to a significant increase in the possibility of accidents and the severity of their consequences. Ultimately, when it comes to reducing the likelihood of accidents, the use of extended traffic light cycles is not an optimal solution. This model highlights that, depending on the traffic conditions observed, it is possible to estimate the level of accident risk and guide traffic management strategies accordingly.