Pedestrian Gap Acceptance Behavioural Modelling at Midblock and Uncontrolled Intersections

Abstract

Pedestrian safety is one of the major concerns and pedestrian gap acceptance (PGA) behaviour depends on various characteristics of pedestrians, vehicles, geometry and environment. This study is intended to develop PGA models at various pedestrian crossing locations (mid-blocks and uncontrolled intersections). Video graphic data was collected from four different pedestrian crossing locations and extracted the required data from videos manually. The multiple linear regression (MLR) technique was used to model the pedestrian gap acceptance behaviour by the vehicular gap size accepted by the pedestrians. MLR models reveal that the frequency of attempts significantly affects the PGA behaviour only in the case of mid-blocks whereas the frequency of disturbances significantly affects only in the case of intersections. The statistical results showed that the pedestrian road crossing behaviour depends on the type of crossing location also. The results can be used to analyse the pedestrian safety levels at various crossing locations.