Optimal Vaccination Campaigns Using Stochastic SIR Model and Multiobjective Impulsive Control

ABSTRACT

A multiobjective impulsive control scheme is proposed to answer how optimal vaccination campaigns should be implemented, regarding two conflicting targets: making the total number of infecteds small and the vaccination campaign as handy as possible. In this paper, a stochastic SIR model is used to better depict the characteristics of a disease in practical terms, where little influences may lead to sudden and unpredictable changes in the behavior of transmissions. This model is extended to analyze the effects of impulsive vaccinations in two phases: the transient regime control, taking into account the necessity to reduce the number of infected individuals to an acceptable level in a finite time, and the permanent regime control, that will act with fixed vaccinations to avoid another outbreak. A parallel version of NSGA-II is used as the multiobjective optimization machinery, considering both the probability of eradication and the vaccination campaign costs. The final result using the proposed framework shows nondominated policies that can guide public managers to decide which is the best procedure to be taken depending on the present situation.

Keywords:
planning of vaccination campaigns; multiobjective optimization; impulsive control; stochastic SIR