Epidemic Dynamics and Intervention Measures in Campus Settings Based on Multilayer Temporal Networks.

Abstract

This study simulates the spread of epidemics on university campuses using a multilayer temporal network model combined with the SEIR (Susceptible-Exposed-Infectious-Recovered) transmission model. The proposed approach explicitly captures the time-varying contact patterns across four distinct layers (Rest, Dining, Activity, and Academic) to reflect realistic student mobility driven by class schedules and spatial constraints. It evaluates the impact of various intervention measures on epidemic spreading, including subnetwork closure and zoned management. Our analysis reveals that the Academic and Activity layers emerge as high-risk transmission hubs due to their dynamic, high-density contact structures. Intervention measures exhibit layer-dependent efficacy: zoned management is highly effective in high-contact subnetworks, its impact on low-contact subnetworks remains limited. Consequently, intervention measures must be dynamically adjusted based on the characteristics of each subnetwork and the epidemic situations, with higher participation rates enhancing the effectiveness of these measures. This work advances methodological innovation in temporal network epidemiology by bridging structural dynamics with SEIR processes, offering actionable insights for campus-level pandemic preparedness. The findings underscore the necessity of layer-aware policies to optimize resource allocation in complex, time-dependent contact systems.