Patient-to-room assignment planning in a dynamic context

The present contribution proposes an extension to the patient as- signment (PA) planning problem in a dynamic context. Two ILP-models have been developed for optimizing this day- to- day planning problem. The rst considers nding the optimal assignment for newly arrived patients, whereas the second also considers future, but planned, arrivals. The performance of both models is compared to each other on a set of benchmark instances. The relative performance with respect to a known lower bound is also presented. Furthermore, the eect of uncertainty on the patients' length of stay is stud- ied, as well as the eect of the percentage of emergency patients. The results show that the second model provides better results under all conditions, while still being computationally tractable.