In this talk, I present the work the SIMID group (www.simid.be) has done so far in informing the Belgian government on the evolution of the COVID-19 pandemic. Specific focus goes to the use of both statistical and mathematical models of infectious disease transmission. Whereas initially a meta-population model previously used to describe the spatio-temporal pattern of seasonal influenza was adapted for COVID-19; it became clear that an individual-based model outperformed this meta-population model upon the implementation of a lockdown and after when advising the government on deconfinement strategies.