[Paper Review] Optimal Resource and Demand Redistribution for Healthcare Systems Under Stress from COVID-19

Felix Parker, Hamilton Sawczuk+3 more|arXiv (Cornell University)|Nov 6, 2020

Healthcare Operations and Scheduling Optimization29 references19 citations

TL;DR

This paper proposes a mixed-integer linear programming framework for optimal redistribution of patients and resources across hospitals during pandemic surges, minimizing required surge capacity and overflow. Using robust optimization and real-world data from New Jersey, Texas, and Miami, the model achieved at least 85% reduction in required surge capacity compared to observed outcomes, demonstrating significant operational efficiency and feasibility under uncertainty.

ABSTRACT

When facing an extreme stressor, such as the COVID-19 pandemic, healthcare systems typically respond reactively by creating surge capacity at facilities that are at or approaching their baseline capacity. However, creating individual capacity at each facility is not necessarily the optimal approach, and redistributing demand and critical resources between facilities can reduce the total required capacity. Data shows that this additional load was unevenly distributed between hospitals during the COVID-19 pandemic, requiring some to create surge capacity while nearby hospitals had unused capacity. Not only is this inefficient, but it also could lead to a decreased quality of care at over-capacity hospitals. In this work, we study the problem of finding optimal demand and resource transfers to minimize the required surge capacity and resource shortage during a period of heightened demand. We develop and analyze a series of linear and mixed-integer programming models that solve variants of the demand and resource redistribution problem. We additionally consider demand uncertainty and use robust optimization to ensure solution feasibility. We also incorporate a range of operational constraints and costs that decision-makers may need to consider when implementing such a scheme. Our models are validated retrospectively using COVID-19 hospitalization data from New Jersey, Texas, and Miami, yielding at least an 85% reduction in required surge capacity relative to the observed outcome of each case. Results show that such solutions are operationally feasible and sufficiently robust against demand uncertainty. In summary, this work provides decision-makers in healthcare systems with a practical and flexible tool to reduce the surge capacity necessary to properly care for patients in cases when some facilities are over capacity.

Motivation & Objective

• To address the inefficiency of reactive, facility-level surge capacity creation during pandemic surges by enabling system-wide load balancing.
• To minimize total required surge capacity and patient overflow across hospitals through optimal redistribution of patients and critical resources.
• To incorporate operational constraints such as nurse availability, transfer costs, and care path variations into a practical decision-support framework.
• To ensure solution robustness against demand uncertainty using robust optimization techniques.
• To provide a flexible, scalable, and publicly available tool for healthcare systems to proactively manage capacity during extreme demand events.

Proposed method

• Formulates a series of linear and mixed-integer programming models to optimize patient and resource redistribution across hospital networks.
• Incorporates robust optimization to handle demand uncertainty, ensuring feasible solutions under varying forecast scenarios.
• Models include operational constraints such as nurse capacity limits, transfer costs, and distinct patient care pathways.
• Uses publicly available hospitalization data from New Jersey, Texas, and Miami to calibrate and validate models retrospectively.
• Employs Gurobi 9.0.3 for efficient solution of large-scale optimization problems.
• Develops an interactive web platform (https://covid-hospital-operations.com/) to visualize and explore model outcomes in real time.

Experimental results

Research questions

• RQ1 Can optimal redistribution of patients and resources reduce the total surge capacity required during a pandemic surge compared to facility-level responses?
• RQ2 How effective is a robust optimization approach in maintaining solution feasibility under demand uncertainty in healthcare systems?
• RQ3 To what extent can operational constraints such as nurse availability and transfer costs be integrated into a scalable redistribution model without sacrificing performance?
• RQ4 How does the proposed model compare to real-world outcomes in terms of overflow reduction and capacity utilization?
• RQ5 Can the model be flexibly adapted to different healthcare systems and applied in real-time planning for future pandemic waves?

Key findings

• The proposed models reduced required surge capacity by at least 85% compared to observed outcomes in New Jersey, Texas, and Miami during the first wave of the pandemic.
• The framework successfully balanced patient loads across hospitals, minimizing overflow and improving system-wide operational efficiency.
• Robust optimization ensured solution feasibility under demand uncertainty, enhancing reliability in real-world deployment.
• The inclusion of operational constraints such as nurse availability and transfer costs did not compromise performance, demonstrating practical feasibility.
• The models outperformed no redistribution by a large margin, even when accounting for complex real-world limitations and data uncertainty.
• The publicly available code and interactive web platform enable real-time exploration and adoption by healthcare systems for strategic capacity planning.