An agent-based modelling framework for assessing SARS-CoV-2 indoor airborne transmission risk

We develop a framework for modelling the risk of infection from airborne Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) in well-mixed environments in the presence of interventions designed to reduce infection risk. Our framework allows development of models that are highly tailored to the specifics of complex indoor environments, including layout, people movements, and ventilation. We explore its utility through case studies, two of which are based on actual sites. Our results reflect previously quantified benefits of masks and vaccinations. We also produce quantitative estimates of the effects of air filters, and reduced indoor occupancy for which we cannot find quantitative estimates but for which positive benefits have been postulated. We find that increased airflow reduces risk due to dilution, even if that airflow is via recirculation in a large space. Our case studies have identified interventions which seem to generalise, and others which seem to be dependent on site-specific factors, such as occupant density.