Modeling Factors Influencing Motorcycle Riders’ Decision When Facing Pedestrians at Unsignalized Mid-Block Crosswalks: A Case Study from Bangkok

Abstract Motorcyclist–pedestrian conflicts at unsignalized mid-block crosswalks represent a critical safety concern, particularly in urban areas of developing countries where motorcycles account for a substantial proportion of the vehicle fleet. This study aims to investigate the key factors influencing motorcyclists’ gap acceptance behaviour when encountering pedestrians at such crosswalks. Field data was collected through video recordings at eight unsignalized mid-block crosswalks in Bangkok. A total of 1,957 motorcyclist observations were analysed using binary logistic regression to identify key predictors of gap acceptance. The results indicated that situational and geometric factors significantly affect stopping behaviour. Specifically, shorter headway time, complex pedestrian movement directions, the proximity of other vehicles nearby, and multiple pedestrians crossing significantly increase the probability of motorcyclists stopping. The study also found that red-white crosswalk markings significantly improve stopping behaviour; however, this effect was significantly reduced when combined with raised medians. The model demonstrated good predictive performance (Nagelkerke R 2 = 0.621), indicating a strong explanatory power of the identified variables. The findings highlight critical implications for design guidelines that emphasize geometric features and promote safety policies related to crosswalk installation in locations with high pedestrian volumes for maximum safety impact.