Modularity maximization to design contiguous policy zones for pandemic response

Area of Science:

• Epidemiology
• Operations Research
• Geospatial Analysis

Background:

• The COVID-19 pandemic highlighted the need for effective public health policies.
• Geospatial approaches can enhance traditional epidemiological responses to pandemics.
• Optimal methods for defining policy coordination zones based on population movement are lacking.

Purpose of the Study:

• To develop and evaluate optimization models for identifying effective policy coordination zones.
• To base these zones on natural human movement patterns for improved public health policy.
• To address the challenge of defining optimal jurisdictions for pandemic response.

Main Methods:

• Developed a mixed-integer quadratic-programming model to maximize community modularity.
• Ensured contiguous jurisdictions within identified coordination communities.
• Implemented heuristic and column-generation algorithms to solve the optimization problem.

Main Results:

• The proposed models and algorithms effectively identify coordination communities in computational experiments.
• Case study in North Carolina and South Carolina demonstrated superior policy coordination potential.
• Detected communities outperformed existing geopolitical boundaries for pandemic response.

Conclusions:

• Optimization models based on human mobility offer a superior approach to defining public health policy coordination zones.
• This data-driven method enhances pandemic preparedness and response strategies.
• The findings provide a framework for more effective, geographically relevant public health interventions.