Dynamic Staffing for Healthcare Contact Centers: A Simulation-Optimization and Forecasting Approach to Enhance Patient Experience and Efficiency

Abstract

Healthcare contact centers play a critical role in the patient care pathway, handling tasks including appointment scheduling, cancellations, patient registration, and records verification. Optimal staffing is vital to enhance service quality, patient safety, and experience. However, variability in call and referral volumes, influenced by factors including time of day and patient needs, complicates staffing. Traditional Erlang models, often used for contact center staffing, assume Poisson arrival rates and Exponential service times and fail to account for variability in agent performance and volume. This study developed a simulation optimization model combined with a forecasting model to create a dynamic staffing policy for SHC's Enterprise Contact Center. The call and referral volumes forecasted using the time series models were inputted to the simulation model that mimics the call center operations, capturing key metrics such as service levels, queue lengths, answer and abandonment rates, and average days to first touch. The optimization algorithm identified the optimal staffing assignments to meet performance targets while minimizing the staffing needs. Our optimized staffing policy adapts dynamically, switching employee assignments between handling patient calls and referrals based on forecasted volumes. The policy also adjusts staffing levels throughout the day and week to account for peak and off-peak times, ensuring efficient resource allocation. Results indicate that, compared to the current practices, this adaptive approach significantly improves the performance (+4% answer rate, +5% referrals processed, -40% average speed to answer, while optimizing workforce utilization (-11% overtime hours). This policy can be applied to other healthcare systems facing similar challenges.