arXiv:2601.14276v1 Announce Type: new
Abstract: Easy bikes have emerged as a popular and affordable mode of last-mile transport in Bangladesh, yet their widespread use has been accompanied by growing concerns about road safety. This study investigates the underlying factors influencing both the occurrence and severity of easy bike crashes by analyzing nationwide crash data spanning from 2016 to 2024. The findings reveal that crashes predominantly occur during daytime, on paved (pucca) roads, and in low-density peri-urban areas. Intersections and curved road segments are also identified as high-risk zones for crash occurrence. Crash severity analysis, supported by binary logit and probit models, emphasizes that the type of collision plays a crucial role in determining the likelihood of fatal outcomes. Pedestrian-involved crashes and rear-end collisions are more frequently associated with fatalities, whereas crashes involving overturns tend to result in less severe consequences. In contrast, environmental factors such as temperature, rainfall, and time of day exhibit limited impact on crash severity. The distribution of crash types also varies across vehicle categories, with motorcycles, buses, and trucks commonly involved in more dangerous collision scenarios. These findings focus on the urgent need for targeted road safety interventions focusing on specific crash types, high-risk locations such as intersections and curves, and vulnerable road users. Moreover, the study underscores the necessity for improving crash data quality, especially in underreported cases, to support informed decision-making. Based on these insights, the study also proposes a set of evidence-based and context-specific interventions aimed at reducing both the frequency and severity of easy bike crashes in Bangladesh.