Coordination of Covid-19 Vaccation: An Optimization Problem and Related Tools Derived from Telecommunications Systems

AbstractThe discovery of vaccines against Covid-19 is coupled with an initial shortage due to the time required to produce them and vaccinate people. This raises the crucial issue of vaccine allocation to optimize people protection, especially the most vulnerable, as they are more likely to contract a severe form of the disease. First, this chapter proposes a review of the existing literature on prioritization criteria to consider as well as methods to optimize the allocation of vaccines and avoid deaths or long-term impacts of the disease. Then, the authors propose a novel decentralized approach to optimize vaccine allocation. A novel way is presented for combining traditional convex optimization problems with Artificial Intelligence (AI) tools. The latter tools are used to characterize the boundary condition for the convex optimization problem, which is then used applying Lagrangian optimization methodologies. AI is thus applied to suitably frame the problem instead of directly deriving the final solution—in this approach, the advantages of AI are exploited while maintaining a high level of explainability for the overall optimization approach. Finally, a discussion presents the outcomes expected from this cross-domain approach.KeywordsData scienceData analyticsArtificial intelligenceCovid-19PandemicVaccination