Date of Graduation
8-2024
Document Type
Dissertation
Degree Name
Doctor of Philosophy in Engineering (PhD)
Degree Level
Graduate
Department
Industrial Engineering
Advisor/Mentor
Eksioglu, Sandra
Committee Member
Proano, Ruben
Second Committee Member
Eksioglu, Burak
Third Committee Member
Shen, Haoming
Keywords
COVID-19; Machine learning; Resource allocation; Stochastic programming; Uncertainty; Vaccine hesitancy
Abstract
Infectious disease outbreaks highlight the urgent need for effective strategies to distribute vaccines and allocate critical healthcare resources to contain the disease and reduce its negative impacts on the population. Managing these allocations is a significant challenge, especially in marginalized communities facing uncertainty in healthcare demand and logistical constraints. This dissertation addresses these challenges by investigating factors that influence dynamic changes in vaccine hesitancy (VH) and its implications for disease spread and healthcare resource demand. It develops optimization models for vaccine distribution and resource allocation under uncertainty, validated with data from the COVID-19 pandemic in the U.S. The first study uses machine learning (ML) models to identify key factors influencing VH at the county level, considering both static (e.g., demographics) and dynamic (e.g., social media sentiment) factors. A metric, VHb, is proposed to track changes in VH behavior over time. Data from between January and October 2021 is used to build and validate this metric. The goodness of variance fit method identifies five distinct county clusters based on VHb. The ML (i.e., Random Forest) model indicates that factors like Google search trends and political affiliation can effectively predict VH trends at the county level. These results reveal that strict vaccination policies can be effective in increasing vaccination rates in counties with low resistance to vaccines. Furthermore, the results reveal significant variation in VH across the U.S., complicating decisions about resource allocation. This motivates the subsequent studies to enhance allocation strategies under stochastic VH trends. The second study proposes a compartmental multi-stage stochastic programming (MSP) model to optimize the allocation of scarce healthcare resources (e.g., ventilators) during a pandemic. The model captures the impact of stochastic VH on disease spread and the need for critical resources. This compartmental model simulates dynamic changes in the population in response to VH over time and space, enhancing the ability of the MSP to adapt resource allocation strategies to the evolving spread of disease and population behavior. Our numerical analysis highlights the significant impact of resource deployment timing and size on managing disease spread and mortality rates. The results emphasize the importance of advanced planning, effective management of critical resources, and balancing equity and effectiveness in resource allocation. The timing of vaccination plays a crucial role in the stochastic behavior of VH. The third study improves the previous compartmental MSP model to optimize vaccine allocation, aiming to better understand the impact of vaccination strategies on VH. The model accounts for how the timing of vaccination influences VH behavior over time and determines optimal vaccination sequences for different population groups. Various prioritization strategies were analyzed to understand their impact on reducing deaths and infections. The study finds that flexible and adaptive strategies, such as the "Available-for-All" vaccination policy, are key to minimizing deaths and infections. Early vaccination of groups with high transmission rate, rather than strictly prioritizing by age or occupation, can significantly enhance public health outcomes. The findings and methodologies presented in this dissertation offer robust tools for managing VH and optimizing resource allocation during health crises, ultimately improving public health outcomes during future pandemics and health emergencies.
Citation
Bui, H. T. (2024). Leveraging Machine Learning and Stochastic Programming to Address Vaccine Hesitancy in Public Health Resource Allocation. Graduate Theses and Dissertations Retrieved from https://scholarworks.uark.edu/etd/5499