Abstract
Hemodialysis treatment requires patients to visit the hospital multiple times per week, with each session lasting several hours. Planning these treatments is challenging and requires interdependent decisions on nurse staffing, patient scheduling and daily assignment of patients to chairs and nurses. These decisions are complicated by heterogeneous patients acuity levels, diverse chair requirements, spatial constraints, and the need to maintain equitable nurse workloads. In collaboration with the hemodialysis center at the Centre Hospitalier de l'Université de Montréal (CHUM), a major teaching hospital in Montreal, we develop a sequential optimization approach that integrates long-term nurse staffing and patient scheduling followed by a multi-objective daily nurse-to-patient-to-chair assignment. The staffing and scheduling model balances workload across appointment periods, while the daily assignment model explicitly incorporates patient acuity, patient-chair requirements, and spatial clustering of chairs. Computational experiments based on real-world data show that schedule optimization reduces the maximum workload per appointment period, while optimized daily assignments decrease peak nurse workload by approximately 30% and substantially reduce non-contiguous chair allocations. Sensitivity analyses highlight trade-offs between workload equity, patient-chair alignment, and spatial efficiency. The proposed framework provides actionable insights for managing recurrent infusion services and is readily adaptable to other ambulatory care settings characterized by regular demand and shared critical resources.