Optimizing Healthcare Worker Availability During Infectious Disease Outbreaks with an MDP Model: Simulation-Based Validation and Sensitivity Analysis for Workforce Resilience

Healthcare workers (HCWs) face elevated risks during infectious disease outbreaks, which can undermine their ability to provide essential care. To mitigate these risks, we develop a Markov Decision Process (MDP) model designed to optimize HCW availability by reducing infection exposure through realistic staffing policies. Unlike traditional Susceptible-Exposed-Infected- Recovered (SEIR) models, our approach focuses on generating implementable policies that enhance workforce resilience during outbreaks. We validate the model through a comprehensive simulation study, first confirming its alignment with findings from existing literature to establish credibility, and then assessing the practical effectiveness of the optimal policy obtained from the MDP model in a detailed simulation environment. Additionally, we examine the model’s sensitivity to varying infection probabilities and reward function coefficients, revealing valuable insights on the resulting number of new infections and workforce availability. By extending the model and simulations to incorporate different number of HCW groups, we further assess the model’s adaptability under diverse outbreak scenarios.  The proposed MDP framework provides a robust, implementable approach for managing HCW availability, offering a critical tool for healthcare system resilience during infectious disease outbreaks.