Protecting healthcare workers (HCWs) during a pandemic such as the one brought on by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is critical to provide timely medical care for patients. Although prior studies have investigated HCW unavailability during the pandemic and have developed policies such as double masking, rotating shift schedules, etc., none of the studies have modeled critical parameters such as varying patient census, vaccination rates, transmission rates, and multiple hospital locations. This research models a high-risk HCW group of perioperative staff, which includes anesthesiologists, nurse anesthetists, and nurses, to investigate the impact of segregating and rotating HCW staffing shifts in a large health system with multiple locations to address staff unavailability during the COVID-19 pandemic and prepare for potential future pandemics. Using the data from one of the largest health systems in South Carolina, we developed an agent-based simulation model with susceptible, exposed, infected, and recovery compartmental model to simulate various pandemic scenarios. Over 24 scenarios with different combinations of patient census, patient transmission rates, and vaccination rates were simulated while accounting for variables like geographic segregation, interpersonal contact limits, patient census, transmission rates, provider vaccination status, hospital capacity, incubation time, quarantine period, and patient-provider interactions to identify that policies that protect HCWs from getting infected. Simulated findings indicate that restrictive policies and their rotation version of policies significantly (p-value < 0.01) reduced the peak weekly unavailability of HCWs by as much as 25% when vaccination rates were lower (<75%). Moreover, these policies significantly (p-value < 0.01) reduced the percentage of HCWs getting infected over the simulation period by as much as 60% when vaccination rates were lower (<75%). However, the benefits of these policies diminished and were statistically insignificant when the vaccination rates increased to 90%. Observations from this research indicate the importance of modeling different parameters of a pandemic, such as vaccination rates, transmission rates, patient census, and other operational information, to develop targeted policies that protect HCWs during different pandemic stages. While the findings are based on the perioperative staff population, they can be implemented or considered while studying other high-risk groups. The simulation model can also be adjusted to simulate different hospital systems and future pandemics by manipulating the respective parameters to support future pandemic preparedness.