Optimizing Influenza Vaccine Composition

Analytics for a better world project

This project connects to SDG 3, Good Health and Well-Being.


  • Dimitris Bertsimas (MIT, Cambridge,  MA, USA)
  • Hari Bandi (MIT, Cambridge,  MA, USA)


Optimizing Influenza Vaccine Composition


(flu) is a highly contagious respiratory viral disease and the seasonal flu epidemics affect about 5-15% of the world’s population, and cause an estimated 3-5 million cases of severe illnesses and up to half a million annual deaths worldwide. The flu shot
(vaccine), which contains two strains of the influenza A virus (H1N1 and H3N2) and one or two strains of the B virus is our first line of defense against seasonal epidemics. Although, most individuals have some level of prior immunity, new strains with mutations
that can escape from host immunity arise frequently and pose a great challenge to producing an effective vaccine. In this project, we propose a holistic framework based on state-of-the-art methods in machine learning and optimization to prescribe influenza
vaccine composition that are specific to a region, or a country. Through numerical experiments, we show that our proposed vaccine compositions could potentially lower morbidity by 8-10% and mortality by 6-9% over vaccine compositions proposed by the U.S. Food
and Drug Administration.