To improve quality and delivery of care, operations need to be coordinated and optimized across all services in real-time. We propose a multi-stage adaptive robust optimization approach combined with machine learning techniques to achieve this goal. Informed by data and predictions, our framework unifies the bed assignment process across the entire hospital and accounts for present and future inpatient flows, discharges as well as bed requests – from the emergency department, scheduled surgeries and admissions, and outside transfers. Based on historical data from a large academic medical center, we demonstrate that our optimization model can be solved in seconds for a 600-bed institution, reduces off-service placement by 23% on average, and boarding delays in the emergency department and post-anesthesia units by 52% and 24% respectively. We also illustrate the benefit from using adaptive linear decision rules instead of static assignment decisions. All together, holistic hospital optimization offers a unique opportunity to revitalize healthcare delivery with optimization and data at the core.
Citation
Dimitris Bertsimas, Jean Pauphilet (2023) Hospital-Wide Inpatient Flow Optimization. Management Science 70(7):4893-4911. https://doi.org/10.1287/mnsc.2023.4933