Integrating Machine Learning and Optimization with Spatiotemporal Techniques to Develop a Methodology for Assessing Rural Resilience

Research into urban resilience has dwarfed the very limited disaster resilience research in rural settings. Because of their different characteristics, a resilience solution in an urban city may not work in a rural environment. This gap between urban and rural readiness became more apparent after catastrophic events such as Hurricane Michael. Adding complexity are the populations at risk, such as the aging population, the most rapidly growing population segment in the State of Florida, and disproportionally the most adversely affected people from storms. As such, there is a clear need to develop novel methodologies along with improvements to the resilience of existing and future infrastructure that can better fit the distinct needs of these rural communities. Moreover, the performance of physical infrastructure systems – whether intact or damaged – is a function of their interaction with social systems. Therefore, there is a need to identify this interaction to fully comprehend the impacts, coping strategies, and barriers to recovery of hurricane victims, particularly differential effects on vulnerable groups such as low-income households, minorities, outdoor workers, the elderly, and the chronically ill. With a focus on Florida’s Panhandle as a test bed, this project will develop a methodology to assess the resilience of these communities using machine learning, optimization, and spatiotemporal techniques based on historical and real-life infrastructure status with data on the environment, socio-economic, demographic, and health-related characteristics of the population