Assessing the Applicability of the Multiagent Modeling Approach to the Epidemic Surveillance of COVID-19 in Russian Cities

This study presents a qualitative and quantitative evaluation of multiagent modeling approach as a tool for COVID-19 epidemic surveillance. For that purpose, we regard a multiagent model based on authors' earlier research and an open source COVASIM framework, comparing their differences in structural complexity, accuracy and computational performance. The mentioned models are calibrated using incidence data from COVID-19 outbreak in Saint Petersburg, Russia, in January-April 2022. The simulation results demonstrate that both models can be applied for retrospective analysis and prediction of COVID-19 incidence. At the same time, there exist important aspects of their usage related to employed demographic data, which should be taken into account when applying them for the tasks of epldemiological surveillance.