Optimized allocation of Ebola intervention resources using a geospatial epidemic modelAdd to Calendar
12:00 pm – 1:00 pm
The geospatial allocation of intervention resources plays a vital role in controlling geographically diverse and rapidly evolving infectious disease epidemics like the 2014 West Africa Ebola outbreak. We develop a two-stage model to determine when and where to optimally assign Ebola treatment unit (ETU) beds across geographic regions. The first stage includes a dynamic transmission model that projects new cases on a regional level, capturing connectivity among regions. We introduce a coefficient for behavioral adaptation in response to changing epidemic conditions. The second stage optimizes the allocation of intervention resources (ETU beds) across affected regions. The optimized allocation could prevent up to 584 infections over a time horizon of 18 weeks, a 40% reduction compared to the planned allocation, and a substantial improvement over other heuristics. Our methodology could be readily applied to other diseases, such as Zika virus, or other types of interventions, such as vaccination or community education programs.