Disaster mitigation actions are considered the most cost-effective alternative to reduce the impact of natural disasters on society. It is generally accepted that predictive models can provide valuable insights for evaluating the benefits of mitigation actions. However, to be effective, such models need to go beyond the common practice of evaluating disaster impact in terms of immediate losses. It is fundamental to investigate how physical, economic, and social infrastructure systems within an urban community interact and affect community recovery. To fulfill this gap, this project will draw knowledge from empirical studies, behavioral and social sciences, planning, and engineering to investigate disaster recovery. A comprehensive agent-based model including water, power, and transportation infrastructure, residential buildings, and households will be developed for the San Francisco Bay Area. Housing recovery and population displacements will be modeled accounting for the limited availability of resources for reconstruction. Socioeconomic aspects of households will dictate their capacity to compete for the scarce supplies they need to recover. The main goal of this project is to understand the factors that make households more vulnerable to the impact of earthquakes, as shown in the figure below. This information will then be distilled down to decisions that if taken by the Bay Area municipalities can reduce the impact of earthquakes and improve the speed and equity of the recovery.