Emerging infectious dis eases are responsible for high morbidity and mort ality\, economic damages to affected countries\, a nd are a major vulnerability for global stability . Technological advances have made it possible to collect\, curate\, and access large amounts of da ta on the progression of an infectious disease. W e derive a framework for using this data in real-t ime to inform disease management. We formalize a treatment allocation strategy as a sequence of fu nctions\, one per treatment period\, that map up-t o-date information on the spread of an infectious disease to a subset of locations for treatment. A n optimal allocation strategy optimizes some cumu lative outcome\, e.g.\, the number of uninfected locations\, the geographic footprint of the diseas e\, or the cost of the epidemic. Estimation of an optimal allocation strategy for an emerging infe ctious disease is challenging because spatial prox imity induces interference among locations\, the number of possible allocations is exponential in t he number of locations\, and because disease dyna mics and intervention effectiveness are unknown a t outbreak. We derive a Bayesian online estimator of the optimal allocation strategy that combines simulation-optimization with Thompson sampling. T he proposed estimator performs favorably in simula tion experiments. This work is motivated by and i llustrated using data on the spread of white-nose syndrome a highly fatal infectious disease devast ating bat populations in North America.

Related Links

- http://www.laber-labs.com - Our home p age \;