Group Member: V. Belik (CEE, MIT)
Motivated by significant economical losses due to miscalculated vaccination campaigns against 2009 H1N1 epidemic, we address the problem of economically optimal geographical vaccine distribution. We take into account “spatial coupling” between different regions due to human mobility. We use census data on human mobility in the USA. We apply an approximate method developed by Prof. D. Simchi-Levi et al. allowing to reduce the dimensionality of the optimization problem. For an influenza-like illness given the source state of epidemic we could provide optimal mass vaccination levels for different US states. The overall savings as compared with conventional proportional vaccination strategy based on the mean filed prediction for a single region could be of the order of 10%.
Impact: the research project contributes to clarification of how to incorporate the available data on human mobility into theoretical models of optimal geographical vaccine distribution. As a the realistic scenario we study the problem of optimal vaccine distribution using a metapopulation defined by the US aviation network.
Collaborators: D. Simchi-Levi (CEE, MIT)