Red Light Running Behavior of Electric Bicycles at Signalized Intersections in China

A comparative analysis was conducted on the red light running behavior of the riders of scooter-style electric bicycles (e-scooters), bicycle-style electric bicycles (e-bikes), and bicycles as they crossed signalized intersections. The crossing behaviors of 5,646 riders (1,568 e-bikes, 2,612 e-scooters, and 1,466 bicycles) were observed. The average red light running rates for e-bikes, e-scooters, and bicycles at the selected sites were 24.9%, 25.0%, and 18.3%, respectively. The rates for e-bikes and e-scooters were significantly higher than those for bicycles. The difference in the rates between e-bikes and e-scooters was not statistically significant. A binary logit model was developed to evaluate how various factors affected the red light running rates of two-wheeled vehicles at signalized intersections. Results of a comparison of traffic conflicts caused by various red light running two-wheeled vehicles indicated that bicycles were slightly more likely to be involved in traffic conflicts than e-bikes or e-scooters as they crossed intersections during a red pedestrian phase. Also, e-bikes were less likely than e-scooters and bicycles to be involved in traffic conflicts in the middle stage of a red pedestrian phase. Both e-scooters and e-bikes had higher crossing speeds than bicycles. Although e-scooters traveled at speeds significantly higher than e-bikes did when crossing intersections during green phases, the difference in crossing speeds between red light running e-bikes and red light running e-scooters was not statistically significant.