Characterizing and Predicting Wildfire Evacuation Behavior: A Dual-Stage ML Approach

TL;DR

This study uses machine learning on survey data to identify behavioral groups and predict evacuation outcomes in wildfires, aiding targeted emergency planning and resource distribution.

Contribution

It introduces a dual-stage ML approach combining unsupervised and supervised methods to analyze wildfire evacuation behavior and predict key outcomes.

Findings

Identified distinct behavioral subgroups based on household resources and preparedness.

Transportation mode can be reliably predicted from household characteristics.

Evacuation timing remains difficult to classify due to dynamic fire conditions.

Abstract

Wildfire evacuation behavior is highly variable and influenced by complex interactions among household resources, preparedness, and situational cues. Using a large-scale MTurk survey of residents in California, Colorado, and Oregon, this study integrates unsupervised and supervised machine learning methods to uncover latent behavioral typologies and predict key evacuation outcomes. Multiple Correspondence Analysis, K-Modes clustering, and Latent Class Analysis reveal consistent subgroups differentiated by vehicle access, disaster planning, technological resources, pet ownership, and residential stability. Complementary supervised models show that transportation mode can be predicted with high reliability from household characteristics, whereas evacuation timing remains difficult to classify due to its dependence on dynamic, real-time fire conditions. These findings advance data-driven understanding of wildfire evacuation behavior and demonstrate how machine learning can support targeted preparedness strategies, resource allocation, and equitable emergency planning.