A stochastic agent-based model to evaluate COVID-19 transmission influenced by human mobility

Abstract

The COVID-19 pandemic has created an urgent need for mathematical models that can project epidemic trends and evaluate the effectiveness of mitigation strategies. A major challenge in forecasting the transmission of COVID-19 is the accurate assessment of the multiscale human mobility and how it impacts infection through close contacts. By combining the stochastic agent-based modeling strategy and hierarchical structures of spatial containers corresponding to the notion of geographical places, this study proposes a novel model, Mob-Cov, to study the impact of human traveling behavior and individual health conditions on the disease outbreak and the probability of zero-COVID in the population. Specifically, individuals perform power law-type local movements within a container and global transport between different-level containers. It is revealed that frequent long-distance movements inside a small-level container (e.g., a road or a county) and a small population size reduce both the local crowdedness and disease transmission. It takes only half of the time to induce global disease outbreaks when the population increases from 150 to 500 (normalized unit). When the exponent of the long-tail distribution of distance k moved in the same-level container, , increases, the outbreak time decreases rapidly from 75 to 25 (normalized unit). In contrast, travel between large-level containers (e.g., cities and nations) facilitates global spread of the disease and outbreak. When the mean traveling distance across containers increases from 0.5 to 1 (normalized unit), the outbreak occurs almost twice as fast. Moreover, dynamic infection and recovery in the population are able to drive the bifurcation of the system to a “zero-COVID” state or to a “live with COVID” state, depending on the mobility patterns, population number and health conditions. Reducing population size and restricting global travel help achieve zero-COVID-19. Specifically, when is smaller than 0.2, the ratio of people with low levels of mobility is larger than 80% and the population size is smaller than 400, zero-COVID can be achieved within fewer than 1000 time steps. In summary, the Mob-Cov model considers more realistic human mobility at a wide range of spatial scales, and has been designed with equal emphasis on performance, low simulation cost, accuracy, ease of use and flexibility. It is a useful tool for researchers and politicians to apply when investigating pandemic dynamics and when planning actions against disease. Supplementary Information The online version contains supplementary material available at 10.1007/s11071-023-08489-5.

Agent-based modeling to estimate the impact of lockdown scenarios and events on a pandemic exemplified on SARS-CoV-2

“…One disadvantage of this approach is that it is unable to simulate the correct number of specific places and to define agent roles in the same way as in ABM models for specific places like Florida 31 or specific dependencies like resources and factories 34 . Furthermore, it is not capable of simulating small-world networks 33 or of analyzing the spread of the virus to other cities through human mobility 32 . This model does not focus on pharmaceutical interventions or vaccinations.…”

“…In contrast to COVASIM, our model does not consider pharmaceutical mitigation measures. In contrast to several ABM models published in 2024 30 – 34 , our model is not based on a specific region or country (USA 31 , 34 , China 32 ). Instead, it has been developed for abstract small city size communities, allowing for easy adjustments without much prior knowledge about parameters.…”

“…The current ABM models from Florida utilise synthetic maps to simulate the location of Ocala 31 or base their ABM assumptions on resource-dependent communities like Alaska, focusing on factories 34 . Other ABMs focus on the spread of COVID-19, influenced by human mobility 32 or expand their model to small-world networks 33 . The objective of our model is to be as abstract as possible at the scale of small cities, focusing on the opening or closing of specific places such as events.…”

“…The objective of our model is to be as abstract as possible at the scale of small cities, focusing on the opening or closing of specific places such as events. With the exception of the Mob-Cov 32 and our model, which is publicly available on GitHub ( https://github.com/Stegobully/Covid19_ABM ), the other models are also written in Python but are not easily publicly available. Our model focuses on events by adding parameters such as frequency or duration.…”