Integrated patient-to-room and nurse-to-patient assignment in hospital wards

What the paper says

Abstract Assigning patients to rooms and nurses to patients are critical tasks within hospitals that directly affect patient and staff satisfaction, quality of care, and hospital efficiency. Both patient-to-room assignments and nurse-to-patient assignments are typically agreed upon at the ward level, and they interact in several ways, such as jointly determining the walking distances nurses cover between different patient rooms. This provides the motivation to consider both problems jointly in an integrated fashion. This paper presents the first optimization models and algorithms for the integrated patient-to-room and nurse-to-patient assignment problem. We provide a mixed integer programming formulation of the integrated problem that considers the typical objectives from the single problems and additional objectives that can only be properly evaluated when integrating both problems. Moreover, motivated by the inherent complexity that results from integrating these two NP-hard and already computationally challenging problems, we devise an efficient heuristic for the integrated patient-to-room and nurse-to-patient assignment problem. We conduct extensive computational experiments on both artificial and real-world instances to evaluate the runtime and quality of the solution obtained with the heuristic. The artificial instances are generated by a parameterized instance generator for the integrated problem that is made freely available.