Understanding and Managing Wildfire Risks to Residential Communities and Supply Chain Networks

Fangjiao Ma

doi:10.7273/000006311

Wildfire has become an increasing threat to humans, the built environment, and ecosystems in the United States. Several factors contribute to such an increase in wildfire risk, including climate change, rapid population growth and infrastructure development at the wildland-urban interface, and accumulated fuels from past wildfire management practices. Increases in wildfire activity have resulted in substantial human and economic losses in the past decade. For example, the 2023 Hawaii wildfires razed more than 2,200 homes and businesses while tragically claiming the lives of at least 115 individuals. A series of California wildfires in 2015, 2017, and 2018 resulted in direct economic losses of $4.8 billion, $18 billion, and $26.3 billion, respectively. In addition to the direct economic losses, the 2018 California wildfires disrupted many supply chain systems, which propagated across the regional economy and induced $88.6 billion in direct losses. These recent wildfires have underscored the urgent need for understanding, assessing, and managing wildfire risks to residential communities and supply chains. To this end, this dissertation aims at understanding and managing wildfire risks to humans, properties, and the regional economy, with a particular focus on residential communities and supply chain networks. To advance our understanding of various proactive and emergency activities, this dissertation begins by examining homeowners’ decisions on wildfire-related proactive actions, such as home hardening, vegetation treatment, and homeowners insurance, through an online survey and subsequently assesses the effect of these actions on the process of housing recovery. Next, this dissertation shifts its focus towards individual behaviors during wildfire events, encompassing their preferences and decisions made during wildfire evacuations. This entails the study of factors like evacuation triggers and timing, as well as a series of en-route decisions made by residents in wildfire-prone areas, all gathered through an online survey. Based on the survey results, data-driven models are developed for predicting evacuees’ behaviors during wildfires. Furthermore, this dissertation integrates these data-driven predictive models with wildfire simulations, vulnerability assessment, and traffic simulation to construct a comprehensive agent-based modeling (ABM) framework for wildfire evacuations under damaged transportation settings. The framework is designed to simulate traffic conditions during a wildfire evacuation and identifies the critical parts of the transportation network for pre-fire risk mitigation actions aimed at improving mobility during a wildfire evacuation. To assess wildfire risk to a supply chain network, this dissertation also proposes a probabilistic wildfire risk assessment framework. It provides rigorous probabilistic descriptions of wildfire ignition likelihood and growth, interaction between supply chain components and wildfire, consequent component damage, and network-level performance reduction. Then, a hypothetical forest-residuals-to-sustainable-aviation-fuel supply chain network is utilized as an illustrative example to demonstrate the capability and applicability of the proposed framework. The proposed framework can be used as a planning tool to evaluate network performance subject to a set of what-if scenarios and the effect of pre- and post-wildfire risk mitigation measures. Overall, this dissertation provides valuable insights for understanding the inherent drivers of individual’s preference on both wildfire proactive actions and evacuation decisions. This information can serve as a foundation for increasing community resilience by helping policymakers and stakeholders to increase participation rates in proactive actions and the responsiveness to evacuation orders. Moreover, the simulation tools and quantitative frameworks developed in this dissertation provide valuable support for stakeholders and policymakers in forecasting post-wildfire performance and implementing more effective pre-event mitigation strategies. These adaptable tools and frameworks show potential for broader applications across various domains, including water distribution networks, transportation systems, and electric power grids, making them valuable assets in addressing the complex challenges posed by dynamic and interconnected systems.

Understanding and Managing Wildfire Risks to Residential Communities and Supply Chain Networks

Files and links (1)

Abstract

Metrics

Details