Model predictive control to mitigate the COVID-19 outbreak in a multi-region scenario

Area of Science:

• Epidemiology
• Public Health
• Control Theory

Background:

• The COVID-19 pandemic necessitates effective regional strategies for exiting lockdowns and managing potential resurgence.
• Existing models often focus nationally, overlooking the complexities of multi-region coordination and varying healthcare capacities.

Purpose of the Study:

• To develop an optimal control approach for post-lockdown COVID-19 mitigation in multi-region settings.
• To support policymakers in defining targeted, coordinated regional exit and restarting strategies.
• To minimize the economic and social costs of non-pharmaceutical interventions while safeguarding healthcare systems.

Main Methods:

• Utilized a non-linear Model Predictive Control (MPC) scheme.
• Integrated a modified Susceptible-Infected-Recovered (SIR)-based epidemiological model.
• Applied the methodology to COVID-19 data from Italian regions.

Main Results:

• Demonstrated an effective approach for defining regional COVID-19 management strategies.
• Showcased the ability to minimize intervention costs without violating healthcare capacity.
• Validated the model's effectiveness in a real-world multi-region scenario.

Conclusions:

• The proposed optimal control framework effectively supports coordinated regional decision-making for COVID-19 mitigation.
• This approach balances economic recovery with public health protection by optimizing non-pharmaceutical interventions.
• It provides a structured method for managing epidemic diffusion across interconnected regional health systems.