Driver behavior modeling at uncontrolled intersections under Indian traffic conditions

Unsignalized intersections lack explicit traffic management mechanisms, which makes them vulnerable to regular disputes and subsequent vehicle crashes. In India, unsignalized intersections predominantly function as uncontrolled intersections. At these crossings, drivers in developing nations such as India fail to yield to movements with a higher priority, increasing vehicle collisions. The objective of the study was to assess the driving behavior of minor roads, considering their aggressive tendency, at uncontrolled intersections. Videographic data were collected at six intersections in tier 2 cities in India. The binary logit model used for minor road right-turning vehicles found that gap acceptance behavior is influenced by vehicle type light commercial vehicle, lag, temporal gap, acceleration/deceleration of the vehicle, and conflicting vehicle speed. In contrast, multilayer perceptron shows that the temporal gap, acceleration, deceleration, and vehicle in line are important parameters that influence driver decisions to accept or reject the gap at an uncontrolled intersection. Because drivers, vehicles, and traffic flow factors all contribute to total traffic behavior, analyzing such crossings is difficult. The correct prediction by the logit and multilayer perceptron models ranges from 68.33 to 66.6% for minor road right turns at uncontrolled intersections. Since planning and decisions for interventions aimed at enhancing road safety depend on a system’s capacity for prediction, both logit and multilayer perceptron models may generally be useful tools for transportation authorities.