Investigation of e-Bike Red-Light Running Heterogeneity with Intelligent Intersection Sensing Data

Le résumé fourni par la source

Red-light running (RLR) behavior by e-bikes significantly increases the risk of traffic accidents, endangering the safety of e-bike riders, vehicle drivers, and pedestrians. This study explores the heterogeneity in RLR behavior among different e-bike rider groups at urban intersections using data collected from intelligent sensing systems in Beijing. The research integrates continuous trajectory data with signal phase, vehicle presence, and non-motorized vehicle/pedestrian information to assess the factors influencing RLR behavior. A Multinomial Logit (MNL) model was employed to analyze and compare the RLR tendencies of regular riders, delivery riders, and couriers. The findings highlight that delivery riders are more likely to engage in RLR due to time constraints, whereas higher densities of non-motorized vehicles and pedestrians reduce the likelihood of such behavior across all rider types, likely due to increased caution. The presence and direction of vehicles within intersections were shown to play a critical role in shaping RLR decisions. These results provide valuable insights into violation patterns and inform targeted traffic management strategies aimed at improving road safety and reducing accidents involving e-bike riders.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.