A Hybrid Agent-Based Model to Simulate and Re-Think Post-COVID-19 Use Processes in Educational Facilities

The paper presents the development and application of a hybrid multiagent system to simulate people's behavior in educational facilities, to support decisions and strategies related to the post-COVID-19 scenarios. Complex use phenomena as the ones occurring in schools and educational facilities, required mixed, hybrid simulation models where the agent-based component, usually controlling single users/bots, is combined with a process-driven engine that ensures correspondence of the users' behaviors to the general scenario. In our case, hybridization includes also the direct interaction of the intended users through a virtual reality 3D game, to further increase accuracy and adherence to the reality of the simulated phenomena. This paper also presents the application of the simulation model to a real case study, a school in Italy, where use processes have been simulated and currently under assessment during the school opening.