Developing a decision support system for emergency management services in the Fairbanks North Star Borough, Alaska

Abstract

Every year the Fairbanks North Star Borough (FNSB), in interior Alaska, responds to common emergencies, as well as disasters of varying types. This research first tested several geographic information systems (GIS) based network analysis models to demonstrate how these models could serve as 'decision support' tools for emergency planning and response. Six network analysis models based on FNSB data were evaluated to determine the response times of local fire stations. Each model was tested against real call times and simulated summertime fire emergency response call times to verify model accuracy. The simulated times matched the modeled predictions with remarkably high R2 values of 0.946, 0.941, and 0.940 for conditions that represented no adjustment to road data, a time penalty for slopes, and a small time penalty for all turns, respectively. The corresponding results with culled real-time call data had a much lower accuracy of 0.403, 0.429, and 0.415, respectively. The lower accuracy for real-time data was primarily due to discrepancies in response time recording protocols. This study also divided FNSB into evacuation zones and created a map book with critical infrastructure and key resources necessary for improved emergency management.