A Non-Pharmaceutical Intervention Policy for Mitigating COVID-19 Pandemic Using Predictive Control Scheme and SEIR Compartmental Model

Among the frequent effects of COVID-19 disease mitigation are significant income declines, an increase in unemployment, and disruptions in the manufacturing, service, and transportation sectors. COVID-19 mathematical models have been created to comprehend and predict the dynamics of the system. Few models, however, make an effort to offer an ideal response to the fundamental problem of how to include and loosen control measures. A severe lockdown policy has reportedly been shown to have extremely negative cultural and economic effects. The Susceptible, Exposed, Infected, Recovered (SEIR) compartmental model is used in this study to suggest a piecewise predictive control policy to reduce the COVID-19 outbreak. By enforcing community activity restrictions, for example, the method attempts to keep the cost of so-called non-pharmaceutical interventions as low as feasible. Model parameters that had been estimated were used to simulate the suggested control system. The model’s parameters can be adjusted to take into account different regional traits. By minimizing the objective function over the course of the prediction time horizon, the continuous optimal control was calculated. Utilizing a concrete piecewise policy function, the computed continuous control value was discretized. The simulation results show that the proposed control approach accurately using non-pharmaceutical interventations enables the development of an efficient policy for managing the COVID-19 diffusion.