Modelling COVID-19 epidemic and its social consequences

Abstract

The paper studies different approaches to modelling COVID-19 transmission. It is emphasized that the variety of models proposed for forecasting the dynamics of epidemic and its long-term socio-economic consequences deals with the complexity of the object under investigation. So the multiplicity of models makes it possible to describe different aspects of complex reality. It is also highlighted that agent-based simulation is more suitable for modelling social aspects of the processes (human behaviour, social interactions, collective behaviour, and opinion diffusion) in the situation of deep uncertainty.The computer experiments with the parameters of the model are analysed on the basis of a number of agent-based models in NetLogo, namely epiDEM and ASSOCC. It is demonstrated that the dynamics of COVID-19 has different scenarios, and agent-based modelling is a powerful tool in political decisionmaking, taking into account social complexity that often exhibits unpredictable output of intervention policy. The role of agent-based modelling in social learning is also discussed. It is pointed out that social learning can reduce the impact of unsubstantiated statements and rumors that are not always adequate to the situation. It is also stressed that social learning could influence social behaviour that, in turn, facilitates the development of social patterns that reduces the likelihood of disease spreading. Attention is paid to the idea that involving people into the modelling process is a part of effective anti-epidemic policy because of the sensitivity of the output of political intervention to the behavioural reaction. It has been shown that today the ideas of agent-based modelling are widely used by social scientists worldwide. The aim of this endeavour is not only to overcome the current pandemic and its long-term socioeconomic consequences but also to prepare for new challenges in the future. The paper is also aimed at paying attention to the lack of agent-based models in Ukraine that could help policy-makers in developing practical recommendations and avoiding undesirable scenarios.