Evacuating individuals and allocating space in shelters are complex tasks that depend on disaster characteristics, roadway network characteristics, the demographics and socioeconomic status of people in the affected region, as well as the effectiveness of the emergency plans. This problem becomes even more challenging when special needs (access and functional needs) populations and those with pets are considered since they may require more time in the event of an evacuation along with more space in shelters. As such, this paper focuses on the evacuation and sheltering needs of such a population group, those 85 and older with special needs (access and functional needs) or pets, in Southeast Florida. Data from Miami-Dade and Broward counties are analyzed using a GIS-based spatial capacitated p-median optimization model. First, an extensive analysis is conducted to assess the accessibility of the American Red Cross-approved shelters with respect to different population groups through which static congested travel times are obtained between each shelter location and U.S. census units. Additionally, a variety of scenarios are constructed to simulate evacuating 85+ populations living in the evacuation zone, providing dynamic congested travel times. Finally, a capacitated p-median optimization model is implemented to maximize the accessibility and capacity of the existing shelters for 85+ populations with special needs (access and functional needs) or pets. Results indicate that the location and allocations of people to shelters are sensitive to demographics and roadway conditions, and the accessibility to shelters can be improved for this group of people based on optimization model results.
Read full abstract