Evaluation of Pedestrian Mid-Block Road Crossing Behavior Using an Artificial Neural Network (ANN)

Abstract

Pedestrians usually cross the road at mid-block locations in India because of the ease and convenience of reaching their destination as compared to intersection locations. It is important to evaluate the pedestrian gap acceptance behavior at mid-block locations because of inadequate vehicular gaps under mixed traffic conditions, which translates into the pedestrian road crossing behavior. The present study examines the pedestrian gap acceptance behavior by employing an artificial neural network (ANN) model for understanding the decision-making process of pedestrians (i.e., acceptance or rejection of vehicular gaps at a mid-block location). From the results, it has been found that the pedestrian rolling gap, frequency of attempt, vehicular gap size, pedestrian speed change condition and vehicle speed have major roles in pedestrian gap acceptance. These results can lead to a better design of pedestrian crossing facilities where adequate gaps are not available in vehicular flow at mid-block crosswalk locations.