Session: ASME Undergraduate Student Design Expo

Paper Number: 173228

Modeling Vehicle Abandonment in Wildfire Evacuations 

The recent January 2025 Palisades and Eaton wildfires had a life-changing effect on many people in the Los Angeles area. The unanticipated disaster left people scrambling to evacuate, with many evacuating in their personal vehicles. Because these wildfires spread to more densely populated areas, traffic congestion made evacuating safely especially difficult. This caused many evacuees to abandon their vehicles and escape on foot, in some cases even being prompted to by emergency responders. This resulted in many vehicles being abandoned in roadways, making evacuation increasingly difficult for others and potentially complicating response efforts by fire fighters and other emergency responders. Because of the lack of data, not much research has been done on this topic, and it is currently difficult to predict and plan for when and where vehicles would be abandoned in an emergency situation. This research will help emergency managers be more prepared with better evacuation plans, inform the public on when they should abandon a vehicle and how to do it without adversely affecting emergency responders, and possibly even assist city planners in creating roadways and communities that are less susceptible to high levels of abandoned vehicles in an evacuation. As fast-moving wildfires such as those that occurred in Los Angeles county are happening at an increasing rate, this research needs to be done to help communities, residents and emergency planners alike, better prepare for possible emergencies.

 

This project intends to investigate several questions, such as where and why vehicles are abandoned in an evacuation, to what degree abandoned vehicles impede further evacuation and emergency response efforts, and how agencies coordinate to move these abandoned vehicles during emergency situations. This research will be carried out using multiple sources of data, such as local traffic data, wildfire locations, emergency notifications by evacuation zone, and interviews with evacuees and agency personnel, such as firefighters, law enforcement, and emergency managers. This research will also implement several statistical models to reach its objectives. These models will investigate the factors that cause a vehicle to be abandoned in a specific location and time,  how everyday traffic patterns differ and are similar to traffic conditions during ongoing evacuations, how well emergency traffic conditions can be predicted by previous traffic data in the same location, and together, predict where vehicles would be most likely to be abandoned in an evacuation situation, both in Los Angeles county and other areas with a high risk of wildfires and vehicle abandonment. This research aims to discover insights that will help emergency managers, city planners, and everyday citizens to be more mindful of vehicle abandonment and its effects and more prepared for possible evacuation situations.

Presenting Author: Dawson Tree Utah State University

Presenting Author Biography: Dawson Tree is a junior in the Department of Math and Statistics at Utah State University pursuing a B.S. in Data Science with minors in Political Science and Statistics. He is interested in applying data science in several diverse fields such as transportation, sports, and political science. He is currently working as an undergraduate research assistant for the Civil and Environmental Engineering Department at USU with a research focus on mobility in natural disasters.

Authors:

Dawson Tree Utah State University
Muhammad Talha Quddoos Utah State University
Matt Palm University of North Carolina at Chapel Hill
Madeline Brozen University of California Los Angeles
Ryan Miller California Polytechnic State University San Luis Obispo
Sarah Grajdura Utah State University