Investigating the safety influence path of right-turn configurations on vehicle–pedestrian conflict risk at signalized intersections

Abstract

Right-turning vehicles and pedestrians share the right-of-way during the permitted signal phase at intersections in countries with right-handed traffic. Although right-turning vehicles are required to stop or yield to pedestrians according to the traffic rules, there still remains circumstances where the two will compete, posing significant safety risks to pedestrians. To investigate the impact mechanism of right-turn configurations, driver characteristics, and traffic operational features on vehicle–pedestrian conflict risk, a driving simulator experiment was conducted. The driving trajectory data of 51 drivers across 28 different scenarios encompassing customized intersection configurations and various traffic conditions were collected. Evaluation indicators, including average crossing speed, maximum deceleration, and post encroachment time (PET) were extracted, of which the first two represented the driving performance of right-turning vehicles, and the last was used to assess vehicle–pedestrian conflict risk. Using a categorical boosting (CatBoost)–Shapley additive explanations (SHAP) approach, pedestrian volume was identified as the most significant influencing factor, with its two levels having the most differential impact on each of the three evaluation indicators. Consequently, a multigroup path analysis was conducted to explore the varying safety influence paths of diverse factors on vehicle–pedestrian conflict risk under different pedestrian volumes. The mediating effects of the average crossing speed and maximum deceleration were found to be significant only under low pedestrian volumes, indicating that intersection configurations not only affect right-turn safety directly but also produce significant indirect effects by influencing driving performance. However, in scenarios with high pedestrian volumes, intersection configurations influenced right-turn safety directly but with no significant indirect effects. The corresponding quantitative insights can help urban road designers construct safer intersections.