Author:

Abstract:

The present dissertation focuses on developing operational decision support models and algorithms for hospital admission planning and scheduling. The aim is to increase efficient usage of key hospital resources by supporting human planners at hospital admission offices with automated tools for their daily and weekly decision making. Three planning processes concerned with admission scheduling of patients are considered: assignment of admitted patients to hospital rooms, determination of admission dates for elective surgical patients, and scheduling surgical cases in operating rooms.The planning process of assigning patients to hospital rooms and wards is the subject of two studies. Firstly, a reactive and an anticipatory decision support model are presented for daily decision making on patient-to-room assignments. It is shown that the anticipatory model is better than the reactive model under various conditions. The reactive model can be seen as an idealized version of current hospital practices, implying that current decision making can be improved and efficient usage of a diverse set of hospital rooms can be increased. Secondly, the Red-Blue transportation problem (Red-Blue TP) is introduced as an abstraction of the patient-to-room assignment problem. A complexity and computational study on the Red-Blue TP provide insights into the difficulty of patient-to-room assignment planning under a gender separation policy.The third and fourth studies concentrate on the admission scheduling process and operating theatre scheduling process for surgical patients. For the admission scheduling process, the aim is to support human planners in determining when patients should be admitted such that expected operating theatre costs and patient waiting time are minimized, while considering limited bed availability. A stochastic optimization model and a heuristic algorithm are presented, that serve as the basis for developing admission scheduling strategies. It is shown that, when given sufficient planning flexibility, stochastic optimization models may improve on deterministic decision models by considering the variance in bed usage and operating theatre usage. However, this improved performance is at the expense of patient friendliness, quality of care and throughput.Finally, for the operating theatre scheduling process, a general and flexible decision support model is presented capturing many considerations encountered in practice. It aims to support human planners in determining a schedule for performing surgical cases in the operating theatre while considering a variety of resources by means of a generalized resource model. Additionally, the model's objectives are to increase throughput and the efficient usage of the operating theatre and its resources. A heuristic algorithm is developed to solve the model which scales well with problem size.