Generating synthetic population for simulating the spatiotemporal dynamics of epidemics

• DOI: 10.1371/journal.pcbi.1011810
• Publication Date: 2024-02-12T18:35:32Z
• Authors: Kemin Zhu; Ling Yin; Kang Liu; Junli Liu; Yepeng Shi; Xuan Li; Hongyang Zou; Huibin Du
• Keywords: Agent-Based Model

Agent-based models have gained traction in exploring the intricate processes governing spread of infectious diseases, particularly due to their proficiency capturing nonlinear interaction dynamics. The fidelity agent-based replicating real-world epidemic scenarios hinges on accurate portrayal both population-wide and individual-level interactions. In situations where comprehensive population data are lacking, synthetic populations serve as a vital input models, approximating demographic structures. While some current synthesizers consider structural relationships among agents from same household, there remains room for refinement this domain, which could potentially introduce biases subsequent disease transmission simulations. response, study unveils novel methodology generating tailored By integrating insights microsample-derived household structures, we employ heuristic combinatorial optimizer recalibrate these subsequently yielding that faithfully represent agent relationships. Implementing technique, successfully generated spatially-explicit encompassing over 17 million Shenzhen, China. findings affirm method’s efficacy delineating inherent statistical relationship patterns, aligning well with benchmarks at city subzone tiers. Moreover, when assessed against stochastic Susceptible-Exposed-Infectious-Recovered model, our results pinpointed variations can notably alter projections, influencing peak incidence rate its onset.