Overtaking decision modeling in heterogeneous, weak lane discipline traffic

Abstract

Developing countries have heterogeneous traffic conditions with weak or no lane-discipline, which allows vehicles to occupy entire width of the road. When a vehicle approaches slow moving vehicles in front of it, the driver will decide whether to overtake or follow it, based upon certain traffic conditions. In this paper, the authors have estimated this overtaking decision in mixed traffic conditions, based upon lateral and longitudinal positions of leading and following vehicles, their speeds and their vehicle types, using logistic regression. It is observed that overtaking chance increases significantly with increase of lateral staggering or absence of any hindrance to overtake. It decreases with increase of longitudinal distance, or decrease of relative speed. Developed models are validated for different vehicle pair combinations. The results from this paper would be helpful to predict the overtaking decision-making of drivers in mixed traffic conditions, and can be input in a traffic simulation model.