Agent-based epidemiological modeling of COVID-19 in localized environments

Area of Science:

• Epidemiology
• Computational modeling
• Public health

Background:

• Epidemiological models inform public health practices but often lack localized precision.
• Modeling disease spread in specific environments like offices or campuses is crucial for effective risk mitigation.
• Current models may not adequately capture the nuances of localized disease transmission.

Purpose of the Study:

• To propose a highly configurable agent-based simulation (ABS) framework for localized epidemiological modeling.
• To provide administrators and policymakers with actionable preparedness information.
• To simulate the impact of interventions on disease spread within specific facilities.

Main Methods:

• Development of a configurable agent-based simulation (ABS) framework.
• Incorporation of detailed control over COVID-19 epidemiological characteristics.
• Application of the ABS to a research lab environment as a proof of concept.

Main Results:

• The ABS framework demonstrates configurability for localized environments.
• Simulation results offer insights into disease risk and intervention effectiveness.
• The model can inform facility-level decision-making and risk management.

Conclusions:

• Agent-based simulation is a valuable tool for localized epidemiological risk assessment.
• The proposed ABS framework can support evidence-based public health decisions in specific settings.
• Future work will focus on enhancing the ABS framework's extensibility and decision support integration.