Adaptive Pandemic Management Strategies for Construction Sites: An Agent-Based Modeling Approach

Newswise — In the field of construction site management, sudden pandemics pose severe threats to worker health and project progress—construction workers face high infection risks due to labor intensity and close physical proximity, leading to project delays, cost increases, and quality declines. However, existing research mostly focuses on statistical analysis of pandemics’ negative impacts, lacking adaptive management strategy development for construction sites. Conventional post-causal inference methods (e.g., archival analysis, questionnaires) cannot provide scenario-specific and counterfactual insights, while traditional contact list-based epidemic simulation models are incompatible with construction sites’ semi-open outdoor spaces and unique worker movement patterns.

Therefore, a research team composed of Chengqian LI (Hunan University), Qi FANG (Central South University), Ke CHEN, Zhikang BAO, Zehao JIANG, and Wenli LIU (Huazhong University of Science and Technology) has carried out a research entitled “Adaptive Pandemic Management Strategies for Construction Sites: An Agent-Based Modeling Approach”.

This study employs agent-based modeling (ABM) to fill the research gap, simulating worker movement patterns and epidemic spread under different risk scenarios and management strategies. First, the team constructed an ABM integrating two core modules: worker daily activity simulation and epidemic transmission dynamics. For activity simulation, they used RFID trajectory data, hierarchical clustering (to identify spatiotemporal movement patterns), and social force models (to simulate movement under time-triggered, activity-triggered, and emergency-triggered drivers). For epidemic transmission, they adopted the SEIAR (Susceptible-Exposed-Infectious-Asymptomatic-Recovered) model to define worker health states, and considered droplet (distance < 1.8 m) and aerosol transmission routes. Monte Carlo simulation was used to handle randomness, ensuring robustness.

Simulation experiments on a Changsha construction site (90 workers, 8 job types) validated three management strategies: 1) Mask-wearing: Universal mask-wearing reduced average infections by ~50%, but partial compliance had negligible effects. 2) Group balance: Dividing workers into job-specific subgroups (with staggered meals/work zones) lowered infection rates from 71.1% to 51.1%, especially reducing cross-infection between high-mobility groups (e.g., carpenters). 3) Entry controls: Screening returning workers daily cut infection rates by 88%. Combing all three measures reduced construction delays by 82.6% (from 18.4 d to 3.2 d). Additionally, infection heatmaps identified dormitories and canteens as high-risk areas.

The paper “Adaptive Pandemic Management Strategies for Construction Sites: An Agent-Based Modeling Approach” is authored by Chengqian LI, Qi FANG, Ke CHEN, Zhikang BAO, Zehao JIANG, and Wenli LIU. Full text of the open access paper: