Physician Assignment and Scheduling Under Licensure Constraints in Emergency Telemedicine

This study addresses operational and tactical challenges in emergency teleneurology, where remote physicians provide real-time consultations to stroke patients across multiple healthcare facilities. At the operational level, we develop and evaluate five dynamic physician-to-patient assignment policies within a discrete-event simulation framework that accounts for stochastic demand, licensure constraints, and non-preemptive service logic. A novel predictive policy, Blast Prediction (PRED), estimates the likelihood of system overload (“blast”) to improve responsiveness and workload balance. Results show that predictive, system-aware assignment strategies significantly enhance service levels without increasing staffing. At the tactical level, we propose a mixed-integer linear programming (MILP) model and a hybrid genetic algorithm (HGA) to solve the Physician Roster Problem (PRP), incorporating individual physicians’ credentialing portfolios, productivity indices, and facility-level factors affecting performance. The model adapts to productivity variations across shifts (day/night) and aims to minimize system costs while aligning staffing with fluctuating patient demand. Using a real-world case study from a U.S. telemedicine provider serving hospitals in multiple states, we demonstrate that integrating productivity and site-level performance factors into scheduling substantially reduces staffing shortages and improves service levels. Together, the simulation-based assignment framework and the optimization-based scheduling model provide a comprehensive decision-support approach to enhance efficiency, equity, and responsiveness in emergency telemedicine operations.