Nonlinear model predictive control with logic constraints for COVID-19 management December 2, 2020
Handling new epidemics that appear from time to time is a very challenging task until the proper medication or vaccine becomes available, due to the fact that usually decision-makers must balance not only conflicting but also rapidly changing goals to save as many lives as possible while ensuring the continued operation of society and economy. There are several mathematical models available that describe and predict the spread of epidemics in varying levels of detail, whose structure is well-known in the literature and the parameters of which can be determined from current statistics.
In the article published in Nonlinear Dynamics, the researchers of PPCU-FIT collaborated with Prof. Gergely Röst from the Mathematical Epidemic Modeling and Epidemiology Analysis research groups to construct a decision support system based on the theory of nonlinear dynamic systems. The goals to be achieved (e.g. constraints on the maximum allowed number of hospitalized patients, even changing in time; compliance with predefined levels of action) are expressed as concise logical statements, from which the complex computational problem is automatically generated by utilizing the system model.
In this way, by integrating the most up-to-date statistics concerning the epidemics into the system, the optimal strategy (that tries to simultaneously ensure the continued operation of the economy and the health care system) can be efficiently supported and updated.