The past four decades have seen a significant increase in wildfire frequency, magnitude, and resulting human and economic losses as driven by climate change and rapid population growth into the wildland-urban interface (WUI, where homes and infrastructure meet the wildland). Wildfire risk can be significantly reduced by three types of pre-event household hazard adjustments—mitigation, preparedness to stay, defend, and survive, and readiness to evacuate. However, the factors that influence the adoption of these hazard adjustments remain poorly understood. Existing wildfire evacuation models rarely consider fire spread dynamics; lack trilateral integration of people, fire hazard, and traffic components; and are based on limited social-behavioral data. To address these knowledge gaps, this project will integrate behavioral data with active learning and goal-setting techniques for increasing WUI residents’ adoption of pre-event hazard adjustments. In addition, social-behavioral data will be infused into transportation engineering models to create more accurate and actionable agent-based models (ABMs) for evacuation. To achieve these objectives, the researchers will collaborate with four WUI communities in three states to (1) identify factors influencing households’ pre-event hazard adjustment adoption and evacuation decision-making for wildfire hazard, and (2) integrate social-behavioral data into wildfire evacuation scenarios using ABMs to evaluate alternative evacuation strateg