The work combines machine learning with a mathematical optimization technique known as column generation. In tests using a detailed vaccine-distribution model, the hybrid approach reduced computational runtime by 79.1 percent compared with conventional column generation, while still producing solutions within 6 percent of the mathematically optimal result. The researchers say the method could transform optimization problems that are theoretically solvable but practically too large for routine use.

“We had developed an optimization model that ensures supplies of vaccines are distributed efficiently and equitably to the places they are most needed,” says Leila Hajibabai, an associate professor in NC State’s Edward P. Fitts Department of Industrial and Systems Engineering and corresponding author of the study. “However, the computational power needed to run that optimization model at a statewide level was not practical. Our goal was to develop a methodology that allows us to use that model but requires far less computing power.”

The underlying model is a maximal covering location-allocation problem. In simplified terms, it attempts to cover as much population as possible with available vaccine supplies while accounting for distribution costs, demand, transportation, and equity. The researchers previously applied the model to vaccine shipments and demand in Pennsylvania during the COVID-19 pandemic, using data from the Centers for Disease Control and Prevention and state and local health departments. The dataset included information detailed enough to support decisions at the level of individual census blocks.

That level of detail is valuable for identifying underserved communities, but it creates an enormous computational burden. “Running that model involved approximately 1.6 billion potential binary and integer decision variables,” says Ali Hajbabaie, an associate professor of civil, construction and environmental engineering at NC State and co-author of the study. Binary variables represent yes-or-no decisions, such as whether a particular allocation is selected, while integer variables can represent quantities such as shipment volumes. Evaluating billions of possible combinations can overwhelm even powerful optimization systems, particularly when planners need to update decisions repeatedly as new information arrives.

Column generation addresses this problem by avoiding the need to consider every possible decision at once. Instead, the method begins with a smaller version of the optimization problem and gradually introduces additional variables, or “columns,” when they appear capable of improving the solution. A secondary calculation, commonly called a pricing problem, searches for promising columns based on the current solution. This approach can shrink the active problem dramatically, but the search may still be expensive when the underlying decision space is exceptionally large.

The NC State researchers added machine learning to guide that search. Their algorithm learns from optimization problems solved previously and uses patterns in those solutions to predict which decisions are likely to be useful in subsequent iterations. Rather than examining every potential variable equally, the system directs column generation toward the most promising candidates. Less promising decisions can be estimated or postponed, reducing the number of optimization calculations required.

“Machine learning can learn from previously solved instances to predict promising decisions for one step of the optimization process,” Hajibabai explains. “That allows the algorithm to focus its computational effort where it is most likely to improve the solution. It reduces the search space and the number of optimization calculations in later iterations.” The machine-learning component does not replace the optimization model; instead, it acts as an intelligent filter that helps determine where the exact mathematical search should spend its time.

In computational experiments, the machine learning-guided column generation method offered the strongest overall balance between speed and solution quality among the techniques evaluated. It did not always reproduce the exact optimum, but its results remained close enough to be operationally useful. The researchers report that solutions were within 6 percent of the optimal outcome while the computational time fell substantially compared with benchmark methods. That trade-off could be especially important in fast-moving crises, when a slightly less-than-perfect plan delivered quickly may be more valuable than an exact plan that arrives too late.

The researchers emphasize that the method is not limited to vaccines. The same combination of predictive algorithms and mathematical optimization could support disaster-relief logistics, humanitarian supply chains, transportation planning, energy-system design, and other infrastructure problems involving millions of possible decisions. In each case, machine learning could identify promising regions of the search space, while optimization methods would evaluate the most important choices rigorously. The study, led by former NC State doctoral student Kuangying Li and co-authored by Ph.D. student Hiruni Niwunhella, was published open access in Sustainability Analytics and Modeling. The research was supported by the National Science Foundation under grant 2124825.

Subject of Research: Computational optimization of vaccine distribution and large-scale resource allocation

Article Title: Machine Learning-Guided Column Generation for a Maximal Covering Location-Allocation Problem

News Publication Date: 30-Jul-2026

Web References: https://www.sciencedirect.com/science/article/pii/S2667259626000196?via%3Dihub; https://www.sciencedirect.com/science/article/pii/S1093968726000939; https://pubsonline.informs.org/doi/10.1287/trsc.2022.1134

References: Li, Kuangying et al., “Machine Learning-Guided Column Generation for a Maximal Covering Location-Allocation Problem,” Sustainability Analytics and Modeling, DOI: 10.1016/j.samod.2026.100069

Keywords: vaccine distribution, machine learning, column generation, mathematical optimization, computational modeling, logistics, resource allocation, supply chains, disaster relief, healthcare planning

Lead author Dane Minnick, a social work and public health expert, explains that despite their widespread use, coalition-based strategies have struggled to achieve measurable, sustainable outcomes at the population level on a national scale. “The existing framework has failed to keep pace with advances in prevention science and the explosion of digital technologies,” Minnick notes. This creates significant barriers to scalability and cost-effectiveness, raising doubts about returning adequate investment given limited public health resources.

Instead, the study advocates for embedding prevention strategies directly within everyday systems such as schools, healthcare settings, and digital platforms. By integrating evidence-based prevention into teacher and provider training, enriching young people’s environments with routine programming, and expanding early risk screening, prevention efforts can become more proactive and widespread. The researchers emphasize the need for a centralized digital repository of free, scientifically validated prevention tools accessible to communities nationwide.

Importantly, the analysis highlights the promise of leveraging cutting-edge digital innovations—mobile apps, wearable devices, and personalized digital interventions—to tailor support, monitor behavioral risk markers, and intervene before substance misuse escalates. These technology-driven approaches offer unprecedented scalability and personalization, addressing a critical gap in traditional prevention models.

This reframed method aims not only to enhance accessibility and consistency but also to alleviate the burden on local coalitions that often struggle with staffing shortages and funding instability. The study posits that a shift towards these integrated, tech-enabled modalities could revolutionize public health prevention, making it more agile, cost-effective, and impactful at a systemic level.

As substance misuse remains a pervasive public health crisis, this analysis presents a timely and influential blueprint for modernizing prevention infrastructure. By embedding prevention into the fabric of institutions and leveraging digital tools, the United States can enhance its capacity to reduce substance misuse and its associated societal harms.

This research signals a critical evolution in public health strategy, prioritizing scalability, sustainability, and scientific rigor. It challenges policymakers and practitioners to rethink and redesign prevention efforts for the digital age, where real-time data and personalized interventions can transform outcomes on a national scale.

Subject of Research: People
Article Title: Reframing Substance Misuse Prevention: a RE-AIM Analysis of Federal Infrastructure and Future Directions
News Publication Date: July 9, 2026
Web References: http://dx.doi.org/10.1007/s11121-026-01920-4
References: Prevention Science, June 30, 2026
Image Credits: USF

Over the past quarter-century, data reveals that more than half of the Gates Foundation’s US$5.5 billion in donations to WHO has been earmarked for vaccine-centric initiatives and polio eradication efforts. This targeted funding contrasts sharply with the relative neglect of other pressing health challenges that WHO identifies as strategic priorities. Such a financial focus not only narrows the scope of WHO’s interventions but also potentially sidelines emerging and critical health issues demanding urgent attention.

The Gates Foundation, currently the second-largest contributor to WHO’s revenue streams—accounting for 9.5% of its total funding between 2010 and 2023—plays a decisive role in shaping the organization’s agenda. While the United States remains WHO’s largest funder, recent developments, including the U.S. government’s announced withdrawal from WHO as of January 2026, signal a future where the organization’s financial stability could be jeopardized further. Compounding this uncertainty, Germany and the United Kingdom rank as the third and fourth largest donors, respectively, though their funding patterns differ in scope and intent.

Notwithstanding the considerable influence held by these donors, a paucity of rigorous investigations into how such funds are specifically allocated within WHO programs has existed. The present analysis stands as one of the first systematic efforts to track and quantify the Gates Foundation’s grants to WHO from 2000 through 2024. By extracting comprehensive grant data directly from the Foundation’s records, researchers meticulously cataloged 640 grants totaling $5.5 billion to WHO, revealing nuanced insights into funding priorities and discrepancies.

A striking finding is that over 80% of the Foundation’s grants to WHO, totaling approximately $4.5 billion, have been targeted at infectious diseases, with nearly 60% specifically focusing on polio eradication. Vaccine programs, a substantial component of these efforts, accounted for more than half of the Gates funding at roughly $2.9 billion. Such concentrated investments underscore a donor preference towards traditional infectious disease interventions, often at the expense of addressing broader health system capacities or chronic, non-communicable diseases.

In stark contrast, important areas such as non-communicable diseases (NCDs), which WHO identifies as accounting for 74% of global deaths with the highest burden in low- and middle-income countries, receive negligible Gates Foundation funding through WHO. Specifically, less than 1% of the Foundation’s dollars supported NCD initiatives. Similarly, water and sanitation projects and health systems strengthening together constituted mere fractions of the total, with just $11.8 million (0.2%) and $37.4 million (0.7%) allocated, respectively. This reveals a disconnect between donor priorities and the comprehensive health needs outlined in WHO’s strategic framework.

The financial architecture of WHO further compounds these challenges. The organization’s revenue is bifurcated into assessed contributions from member states—which are calculated based on a country’s wealth and population—and voluntary contributions or extra-budgetary funds from both member states and non-state actors. Notably, approximately ninety percent of WHO’s income consists of voluntary contributions, nearly all of which come with earmarking conditions. These stipulations allow donors to dictate the usage of funds, effectively steering the organization’s program implementation and goal setting.

This earmarked funding model constrains WHO’s operational flexibility, forcing a prioritization of donor interests over its own strategic vision. According to the study authors, insufficient assessed contributions have compelled WHO to accept earmarked donations, thereby skewing resource allocation disproportionately towards projects favored by major donors while underfunding vital, though less glamorous, areas. Such a system risks perpetuating imbalances and undermines the efficacy of global health initiatives, particularly for emerging health threats.

The implications of this funding dependence are particularly pressing given the potential exit of the United States, WHO’s largest funding source. Should the withdrawal proceed as scheduled in 2026, WHO’s already precarious reliance on voluntary, earmarked contributions will intensify. The consequent funding vacuum could exacerbate governance challenges and further restrict WHO’s ability to implement a balanced and evidence-informed health agenda responsive to evolving global health dynamics.

Critical reflection on the role of non-state actors such as the Gates Foundation also tempers the narrative. While it is easy to fault these donors for wielding disproportionate influence—potentially undermining WHO’s independence—this study rightly highlights the role of member states in perpetuating the status quo. For over four decades, governments have resisted increasing assessed contributions commensurate with the scale and complexity of WHO’s mission, effectively transferring power to donors through structural underfunding.

WHO has long advocated for a fundamental reconfiguration of its financing framework, calling for increased flexible and sustainable funding that aligns with strategic priorities rather than donor-driven projects. Without such reforms, the organization risks remaining hostage to external interests, hampering its capacity to comprehensively address the multifaceted challenges that define modern global health crises. The study’s findings serve as a stark reminder that sustainable global health leadership hinges not only on resources but on preserving institutional independence through sound and equitable financing mechanisms.

In a world grappling with interconnected health threats—ranging from pandemics to the rising tide of chronic diseases—ensuring WHO’s financial model facilitates rather than constrains its mandate is paramount. As global health governance enters an era marked by geopolitical shifts and emerging priorities, the lessons drawn from this study underscore the urgent need for member states to reaffirm their commitments and for WHO to reclaim its role as an impartial steward for all facets of global health.

Subject of Research: Not applicable
Article Title: Who’s leading WHO? A quantitative analysis of the Bill and Melinda Gates Foundation’s grants to WHO, 2000-2024
News Publication Date: 28-Oct-2025
Web References: http://dx.doi.org/10.1136/bmjgh-2024-015343
Keywords: Health and medicine, Health care policy, Research priorities