University of Twente Student Theses


Predicting hospital bed census due to planned surgeries using queueing models

Jong, B. de (2022) Predicting hospital bed census due to planned surgeries using queueing models.

[img] PDF
Abstract:Nurses are scheduled for work according to a prediction for the number of inpatient patients. The size of the prediction interval plays a key role here. A model has been created to predict the number of patients that are inpatient due to planned surgeries and to analyze its prediction interval. This has been done by using two approaches: firstly by applying an M/M/∞ queueing model and secondly by applying an M/G/∞ queueing model. We have applied the M/G/∞ model to a normal and log-normal service distribution, where an additional update rule is introduced once a patient undergoes surgery. The schedule, or blueprint, of these surgeries has been either deterministically made in advance, stochastically made in advance or the schedule can be altered up until the moment of surgery. For both the M/M/∞ model and the M/G/∞ model, we have simulated both the deterministic and stochastic blueprint for various parameters. Additionally, for the M/G/∞ model, we have used a log-normal service distribution, where we compare the additional update rule to the case where no additional update rule is used. We see that the introduction of the additional update rule does not benefit the quality of the prediction. The size of the prediction interval of the models does not decrease with absolute certainty, but the simulations show that it decreases almost always when time progresses.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:02 science and culture in general
Programme:Applied Mathematics MSc (60348)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page