Analyzing driver-pedestrian interaction in a mixed-street environment using a driving simulator

This paper presents the design, analysis and results of a driving simulator experiment conducted to study the interaction between drivers and pedestrians in a mixed-street environment. Ninety-six students of the American University of Beirut (AUB) participated in the experiment that took place in the Transportation and Infrastructure Laboratory of AUB. The study looked at the driver-pedestrian interaction from the driver's perspective, by quantifying the effects of different scenario variables on the driving behavior of the participants. Kruskall-Wallis test shows that drivers' behavior in proximity of pedestrians tends to be statistically significantly less aggressive when their approach velocity is lower, curb-side parking is not allowed, a crosswalk exists, and the number of pedestrians crossing the street is higher. A discrete choice model for the yielding behavior of the drivers was also developed as a function of different predictor variables. Five out of the six predictors considered (except for gender) had a statistically significant effect on the yielding behavior, particularly the effects of curbside parking, number of pedestrians crossing, and approach velocity. The model was then used to evaluate the effect of policy variables on the yielding probabilities of the drivers. The results of this study enrich current knowledge and understanding of drivers' behavior and their interaction with pedestrians, especially with studying the effects of scenario variables that were not addressed before; this would help planners propose and evaluate safety measures and traffic calming techniques to reduce the risks on pedestrians. The study also confirms the effectiveness of driving simulators in studying driver-pedestrian interactions.