Pedestrian crash risk near bus stops in Delhi: development of safety performance functions

The present study develops safety performance functions (SPFs) to identify key factors contributing to pedestrian crashes near high-risk bus stops in Delhi, India. Bus stops associated with fatal and non-fatal incidents were identified across multiple administrative districts. To capture temporal variation, the analysis was conducted separately for daytime and nighttime crashes based on reported crash timings. Video-graphic surveys and built-environment audits were undertaken to identify potential factors influencing pedestrian safety. Negative Binomial and Hurdle Negative Binomial models were estimated to analyse crash occurrence. The results indicate that various built-environment attributes-such as sidewalk condition, on-street parking, crossing type, and lighting conditions-significantly affect crash risk near bus stops. Additionally, higher bus commuter volumes were associated with increased crash frequency. The findings further suggest that Hurdle models outperform conventional count models for such data. Overall, the study highlights critical design and operational factors requiring targeted interventions to enhance pedestrian safety near bus stops.