Pedestrian safety analysis at urban midblock section under mixed traffic conditions using time to collision as surrogate safety measure

Abstract

Pedestrians are the most vulnerable road users, and pe-destrian safety has become a major concern among re-searchers in recent years due to the increasing number of road fatalities. Conflict analysis using surrogate safe-ty measures (SSMs) helps study pedestrian safety, as there are several limitations with collision data. More-over, it is a cost-effective technique compared to historical crash data analysis. The present study analyses pedes-trian safety at urban midblock crosswalks using time-to-collision (TTC) as SSM. The data for the present study were collected from four different midblock pedestrian crossing locations in different cities in the western part of India using the videographic technique. The trajectory of pedestrians and vehicles was extracted for micro-level analysis of pedestrian–vehicle interactions. The trajec-tory data were further used to calculate TTC at regular time intervals during the interaction of pedestrians and vehicles. Two different types of pedestrian road crossing behaviour, viz. vehicle pass first and pedestrian pass first were identified, and TTC analysis was carried out differently for each scenario. The variation of TTC based on gender and vehicle category was analysed to evaluate the influence of such parameters on pedestrian safety. The generalized linear mixed model approach was used to develop linear regression models for TTC based on empirical data. The threshold values for TTC were used to define various safety levels of pedestrians using a clustering approach.