Auto tuning SIR model parameters using genetic algorithm

Liu, F.; Lee, K.

Earlier studies comparing Covid-19 simulations using extended SIR model with observed new cases in New Jersey and United States showed good agreement between simulated results and observational data. The parameters of the SIR model controlling the behavior of the model have to be manually adjusted until the modeled results and observations reach good agreement. The parameter tuning process is tedious and time consuming. In this work, we have developed an approach using genetic algorithm to automatically select the most optimal set of parameters to minimize the residual between simulated result and observational data. The parameter tuning process applying SIR model can now be automated without tedious and time consuming manual intervention.