Abstract Manager

The rise in adoption of electric vehicles (EVs) presents a unique challenge for disaster planning. Their reliance upon the grid for fuel requires capable resilient electricity infrastructure to withstand the surge in demand from evacuees. For those living in areas highly vulnerable to wildfires, safe and resilient evacuations are especially important. Simultaneously, EVs present a novel opportunity to act as power sources that fulfill the electricity needs of communities that would otherwise be cut off from power. Their mobility makes them invaluable for transporting power to crucial facilities in emergencies. As such, the actions of EV-owners during disasters is critical in helping us understand how to prepare for outage events for the rapidly growing ubiquity of these vehicles. This research utilizes discrete choice models in an effort to understand electric vehicle-enabled actions and choices in wildfire-prone regions.

The models used in this research draw from the results of an evacuation study distributed among populations in the Canadian provinces of Alberta and British Columbia who are at high risk of exposure to wildfires. The survey encompasses factors such as risk perception, willingness to share, and potential behavior in evacuation scenarios with access to electric vehicles. These factors play a crucial role in shaping individual decisions during a disaster. The methodology involves utilizing Biogeme software to create discrete choice models that accurately represent the relationship between survey-taker responses and their EV-enabled actions. The procedure includes the development of various binary and multinomial logit models with the ultimate goal of finding a link between electric vehicle use and other vital user behaviors and during crises.

While the models are currently under development, it is anticipated that they will be fully completed within the next six months. We hope the results of these models will entice a comprehensive discussion of the results, implications, and potential applications. The research findings could hold significant value for informing policymakers, urban planners, and stakeholders involved in disaster evacuation and planning in wildfire-prone regions. The conclusions from this research will further aid the construction of an agent-based model (ABM) to fully understand the interactions of autonomous decision-makers within the context of EV resilience during disasters. The results from this research can inform targeted interventions to facilitate the safe and effective use of electric vehicles during outage events and evacuations.