High-Resolution Agent-Based Modeling of Campus Population Behaviors for Pandemic Response Planning (2405.11414v2)

Abstract: This paper reports a case study of an application of high-resolution agent-based modeling and simulation to pandemic response planning on a university campus. In the summer of 2020, we were tasked with a COVID-19 pandemic response project to create a detailed behavioral simulation model of the entire campus population at Binghamton University. We conceptualized this problem as an agent migration process on a multilayer transportation network, in which each layer represented a different transportation mode. As no direct data were available about people's behaviors on campus, we collected as much indirect information as possible to inform the agents' behavioral rules. Each agent was assumed to move along the shortest path between two locations within each transportation layer and switch layers at a parking lot or a bus stop, along with several other behavioral assumptions. Using this model, we conducted simulations of the whole campus population behaviors on a typical weekday, involving more than 25,000 agents. We measured the frequency of close social contacts at each spatial location and identified several busy locations and corridors on campus that needed substantial behavioral intervention. Moreover, systematic simulations with varying population density revealed that the effect of population density reduction was nonlinear, and that reducing the population density to 40-45% would be optimal and sufficient to suppress disease spreading on campus. These results were reported to the university administration and utilized in the pandemic response planning, which led to successful outcomes.

High-Resolution Agent-Based Modeling of Campus Population Behaviors for Pandemic Response Planning

Background

The COVID-19 pandemic brought a host of unprecedented challenges to various institutions. Universities needed to quickly adapt to ensure the safety of their faculty, staff, and students. This paper focuses on a case paper where high-resolution agent-based modeling (ABM) was used to understand and mitigate the spread of COVID-19 on the Binghamton University campus. While large-scale models were widely used, this paper zooms in, addressing smaller-scale settings with high detail.

Methodology

Data Collection

One of the primary challenges encountered was the lack of direct, empirical data on individual movements. To compensate for this, a variety of indirect data sources were compiled, such as:

• Course schedules and classroom allocations
• Residence hall assignments for students
• Employee office locations
• Parking lot capacities and bus schedules

Model Development

The core of the model is a multilayer transportation network where each layer represents a different mode of transport: walking, driving, and taking the bus. The agents (students and staff) were programmed to move along the shortest paths between locations. To get a detailed representation, the manual extraction of geographical data from Google Earth was used to build the campus layout including:

• Roadways
• Pedestrian paths
• Buildings and parking lots

Each agent was then assigned specific behaviors and schedules based on the indirect data. Movement simulations were done on a minute-by-minute basis, considering congestion and potential social contact.

Results

Identifying High-Risk Areas

Simulations with the model revealed several key findings:

1. High-Risk Locations: The paper identified specific zones on campus where high frequencies of close social contacts occurred, such as busy corridors and food courts.
2. Population Density Reduction: By running simulations at different population densities, it was observed that reducing the campus population to 40-45% would be optimal for mitigating disease spread without requiring a full shutdown.

Validity Check

The model showed an impressive level of accuracy when compared with real-world parking data. The correlation was found to be 0.768909, which adds a layer of confidence to the model’s predictions.

Implications and Future Prospects

Practical Implications

• Decision-Making: These simulations provided factual groundwork for campus administrators, helping them develop data-driven strategies such as transitioning some courses online.
• Public Health Education: Visualizations and presentations from the model were used to communicate the importance of social distancing and related measures to the campus community.

Theoretical Implications

The paper proves the value of detailed, high-resolution ABM in small to mid-sized settings. While larger models operate on a broader scale, this level of detail is crucial for nuanced decision-making in more confined environments.

Speculations on Future Developments

While the model created was specific to Binghamton University, it serves as an example of how similar techniques can be adapted to other campuses or small communities. Given the rapid advancements in data collection and ABM, future efforts could incorporate real-time data, making these models even more robust.

Additionally, the methodology could extend beyond pandemic planning to various applications, from traffic management to emergency evacuation planning.

Key Takeaways

The creation of a high-resolution agent-based model for Binghamton University demonstrated the power of detailed simulations in managing real-world crises. By effectively identifying high-risk areas and optimal density reductions, it provided actionable insights for campus administration, proving invaluable during an uncertain time.

Overall, this case paper serves as a testament to the practical and theoretical possibilities offered by high-resolution ABMs, especially in situations where high detail and localized context are essential for accurate predictions and effective decision-making.