Optimal COVID-19 epidemic control until vaccine deployment

Ramses Djidjou-Demasse; Yannis Michalakis; Marc Choisy; Micea T. Sofonea; Samuel Alizon

By varying coefficient B, we can weight the relative importance of the death term and the control cost term in the objective function (3.6) . Small values of B correspond to situations where the control can be seen as "inexpensive", e.g. the host population can support strict public health measures both socially and economically. As a consequence, corresponding optimal controls can be very intense (Figure 3a ) and further decrease severe cases and deaths ( Figure 3b,c) . On the other hand, large values of B correspond to situations where control measures are "expensive" for the population, making strict public health measures unsustainable at the population level. In this case, control intensities remain intermediate (Figure 3a) , with more infections and deaths compared to cases with smaller values of B (Figure 3b,c) . However, except for very little costs where constant maximum control is the optimum, we find that the shape of the control function remains unchanged with an intense control by week 20, followed by a decrease in control intensity.