Automated decision-support for hospital admission planning

Hospitals – like any other organization – continuously aim to improve the efficiency of their most costly resources. Faced with the uncertain and highly variable nature of the care process, they try to manage the flow of patients in such a way that operational inefficiencies (overtime in the operating theatre, bed shortages in hospital wards, etc.) are minimized, while still providing care to all patients in a timely fashion. One key instrument in this management process is the admission planning of elective patients. The research question that we pose is: what automated, intelligent tools can we provide to support this decision-making process? Our current research focuses on developing decision-support models and algorithms for admission planning in hospitals. By effectively planning the inflow of elective patients, we try to improve efficiency of a hospital’s key resources (operating theatres, hospital wards and beds, and personnel). At the same time, we also anticipate the inflow of non-elective (i.e. emergency) patients – which by definition cannot be planned. This demo presents the planning of a set of elective patients over a certain period of time. The aim is to improve operational efficiency of two key resources: the operating theatre and the hospital wards. The demo shows how a metaheuristic algorithm improves an initial admission plan to have high operating theatre load, while avoiding overtime; to have a balanced patient load in the hospital wards, ensuring a stable workload; and – finally – to meet patient (room) preferences while avoiding room conflicts (e.g. male/female patient separation).status: publishe