Agent-based epidemic models: International trends and implementation in Russian practice

The article presents an analysis of developments in the modeling of epidemiological processes, primarily conducted using a modern mapping the scientific research landscape. The authors identify the world's leading research centers and groups that have been actively working in this field in recent years, as well as the key approaches used in modeling epidemiological processes that have been tested in practice and proven their adequacy. It is noted that alongside classical compartmental models, which describe epidemiological dynamics as the transition of a population segment from one category (associated with a specific stage of an infectious disease) to another, agent-based models are becoming increasingly widespread. These models simulate the processes of person-to-person infection transmission. The advantages of the agent-based approach include the ability to reproduce population heterogeneity within an artificial society, which significantly influences infection spread processes, and to simulate direct and feedback loops between various processes, including the agents' capacity to adapt their behavior based on the current epidemiological situation. The paper presents the concept of an agent-based epidemiological model for Russia, designed for testing and evaluating the effectiveness of planned anti-epidemic measures. The concept involves simulating the spread of an epidemic through agent interactions: the population of agents replicates the real socio-demographic structure of the population, accounting for all individual characteristics that influence people’s involvement in infection spread, including their social connections, spatial distribution, and regional specifics; the progression of each agent through all stages of the disease is simulated according to the features of the infectious disease causing the epidemic; a realistic picture of population movement across Russia's transport infrastructure is reproduced, represented as a graph where vertices are transport hubs and edges are the connecting routes.