The use of crowdsourcing data for analyzing pedestrian safety in urban areas

Abstract

Pedestrians are the most affected vulnerable road users by traffic collisions. Due to incomplete and inconsistent collision statistics, assessing pedestrian safety remains a complex issue in developing countries. This study investigates the potential of using crowdsourced data to identify hotspot locations by observing pedestrian-vehicle interactions. Safety analysis was carried out using traffic incident data in Eastern Cairo, Egypt. Incident data included collisions, near misses, and infrastructure issues. Spatial autocorrelation analysis was undergone to determine whether incidents are clustered, dispersed, or randomly distributed. The results showed that incidents in the study area are generally dispersed. Nevertheless, local spatial autocorrelation showed that some locations on four major corridors were identified as hotspots with a 99% confidence level. The approach proposed in this study shall help transportation authorities in developing countries to identify and prioritize sites that require more safety attention.

Keywords

• Streetguards
• Crowdsourcing
• Spatial analysis
• Pedestrian-vehicle collisions
• Hotspot locations