Modeling the Impacts of Driver Aggression during a Metropolitan Evacuation

Abstract

Emergency evacuation plans for metropolitan areas frequently rely on transportation simulation models to predict evacuation durations. Previous work frequently assumes drivers will demonstrate similar aggressiveness during an evacuation as during peak hour commuting. On the contrary, several studies suggest driving behavior is different during evacuations. The objective of this study was to find if changing driver aggressiveness will significantly impact the expected duration of a no-notice evacuation. In this study, the authors used microscopic traffic simulation software to model travel demand for a no-notice evacuation scenario in the St. Louis metropolitan area and represented driver aggressiveness by adjusting headway, following distance, and other parameters. This model included the traffic volumes predicted by the regional planning agency and the route diversions and freeway service patrol operations expected by the local state Department of Transportation. Although the findings suggested that changes in driver aggressiveness will not significantly change the evacuation duration; for key evacuation routes, a significant reduction in travel time and delay was found when driver aggressiveness decreased, compared to the observed normal peak hour driver behavior. These findings suggest that traffic engineers and emergency managers should carefully consider the wording of evacuation information provided to drivers in an attempt to reduce anxiety and prevent aggressive driving during a no-notice evacuation. Further, research should attempt to capture driver behavior during future evacuations so modelers can model travel behavior more accurately.