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

Area of Science:

• Transportation Engineering
• Human Factors Psychology
• Traffic Safety

Background:

• Understanding driver-pedestrian interactions is crucial for urban safety.
• Mixed-street environments present complex challenges for road users.
• Existing research often lacks detailed analysis of specific scenario variables.

Purpose of the Study:

• To quantify the impact of various scenario variables on driver behavior towards pedestrians.
• To develop and validate a discrete choice model for driver yielding behavior.
• To assess the influence of policy interventions on driver yielding probabilities.

Main Methods:

• A driving simulator experiment involving 96 university students.
• Analysis of driver behavior using Kruskal-Wallis tests.
• Development of a discrete choice model incorporating predictor variables like approach velocity and pedestrian presence.

Main Results:

• Lower approach velocity, absence of curb-side parking, presence of crosswalks, and higher pedestrian numbers significantly reduce aggressive driver behavior.
• Approach velocity, curb-side parking, and pedestrian count were significant predictors of driver yielding behavior.
• The developed model demonstrated the impact of policy variables on yielding probabilities.

Conclusions:

• Driving simulators are effective tools for studying driver-pedestrian interactions.
• Findings provide valuable insights for urban planners to develop safety measures and traffic calming strategies.
• Specific scenario variables significantly influence driver behavior, offering targets for safety interventions.