Exploring factors of E-bike rider’s hit-and-run behavior of E-bike crashes on urban and rural roads using Light Gradient Boosting machine and SHapley additive exPlanations

Electric bicycles (e-bikes) are a primary mode of short-distance transportation in China but pose significant traffic safety challenges. While prior research has largely regarded e-bike riders as vulnerable road users, this study highlights their role as potential perpetrators in crashes involving other vulnerable road users, particularly in hit-and-run events. We analyze 6022 e-bike crashes using the Light Gradient Boosting Machine and SHapley Additive exPlanations, with the binary dependent variable defined as whether a crash was a hit-and-run, to identify risk factors on urban and rural roads. Key differences were identified: on urban roads, crashes during morning rush hours, on nonstandard lanes, or involving fewer injuries were more likely to result in hit-and-runs. On rural roads, crashes at dawn or dusk, involving two e-bikes, caused by violations, or with fewer injuries showed higher hit-and-run tendencies. These findings offer valuable insights for managing and preventing e-bike-related hit-and-runs.