Evaluating driving behavior patterns during wildfire evacuations in wildland-urban interface zones using connected vehicles data

Wildfire risk is increasing all over the world, particularly in the western United States and the communities in wildland-urban interface (WUI) areas are at the greatest risk of fire. Understanding the driving behavior of individuals to evacuate fire-affected WUI areas is important as the evacuees may encounter low visibility and difficult driving conditions due to burning material and steep topography. This study investigates the driving behavior patterns of individuals during historical wildfire events in rural and urban areas with mandatory evacuation orders using a connected vehicle dataset. This dataset provides the geolocation and timestamp of vehicles’ hard-braking (HB) and hard-acceleration (HA) events. The comparative analysis of the data demonstrated that the reported HA & HB event patterns were consistent in representing the varying driving behavior in response to the changing driving conditions and the depiction of the temporal and spatial impact of the fire in all studied areas. Moreover, the HB event dataset located critical traffic congestion points and the HA event dataset revealed the hurried response of evacuees on the exiting routes as a result of short-notice evacuation. In addition, significant differences in driving behavior patterns were noticed between rural and urban areas.