Exploring the impact of mobility restrictions on the COVID-19 spreading through an agent-based approach

Mobility restriction is considered one of the main policies to contain COVID-10 spreading. However, there are multiple ways to reduce mobility via differentiated restrictions, and it is not easy to predict the actual impact on virus spreading. This is a limitation for policy-makers who need to implement effective and timely measures. Notwithstanding the big role of data analysis to understand this phenomenon, it is also important to have more general models capable of predicting the impact of different scenarios. Besides, they should be able to simulate scenarios in a disaggregated way, so to understand the possible impact of targeted strategies, e.g. on a geographical scale or in relation to other variables associated with the potential risk of infection. This paper presents an agent-based model (ABM) able to dynamically simulate the COVID-19 spreading under different mobility restriction scenarios. The model uses the Italian case study with its 20 administrative regions and considers parameters that can be attributed to the diffusion and lethality of the virus (based on a virus spread risk model) and population mobility patterns. The model is calibrated with real data and reproduces the impact that different mobility restrictions can have on the pandemic diffusion based on a combination of static and dynamic parameters. Results suggest that virus spreading would have been similar if differentiated mobility restriction strategies based on a-priori risk parameters instead of a national lockdown would have been put in place in Italy during the first wave of the pandemic. The proposed model could give useful suggestions for decision-makers to tackle pandemics and virus spreading at a strategic level.