Preventing epidemics in networks using integer programming

The main goal of this project is to develop mathematical programming-based models for preventing disease spread taking into account network effects and proposing efficient exact solution approaches to solve them. So far, the studies concerning the mathematical modeling of COVID-19 and its dynamics have been focusing on the parameter estimation for certain disease spread models and the analysis of the impact of exercising certain measures such as isolation. Although some of these studies involve an optimization aspect with the objective of minimizing the spread, the solutions are usually not targeted since the network structure of populations is ignored. However, the network topology may effect the spread significantly. For example, it is known that disease spreads faster in small-world networks than in many other network structures. Thus, our goal is to study optimization problems which allow to close this gap in research. For all the studied problems, we plan to make the developed solution codes available online to other researchers. Naturally, our work does not only include the modeling of problems and design of solution algorithms, but also their evaluation on real-world, as well as on synthetic instances. By focusing on exact solution approaches, which provide solutions together with performance guarantees our research can help to improve the acceptance of and the trust on such software systems by decision makers in the sense of explainable AI.