University of Twente Student Theses


Predicting utilization in a multi-provider ambulatory allied healthcare organisation

Poll, T.A. van der (2022) Predicting utilization in a multi-provider ambulatory allied healthcare organisation.

[img] PDF
Abstract:This study aims to develop a prediction model that is able to predict session (combination of date, practitioner, location) utilization accurately, within a time horizon of one month, such that it influences a newly proposed multi-objective online scheduling method in order to reduce the variance in session utilization. A reduction in session utilization variance should help the new scheduling method to reduce the waiting time for new patients. Large growth in requests for patient appointments has lead to unacceptable waiting time for new patients (62.9% within 21 days in 2019). Therefore the decision was made to analyze and alter the scheduling process, with the main goal to reduce waiting time for new patients. We use a LightGBM prediction model and find that it reduces the test rmse to 0.041. We simulate the prediction model in order to validate it's performance in relation to the newly proposed scheduling method. We find that the LightGBM model significantly reduces the variance in utilization. However this comes at the cost of other objectives. We recommend to implement the LightGBM model, and carefully investigate implementation in practice.
Item Type:Essay (Master)
Faculty:BMS: Behavioural, Management and Social Sciences
Subject:31 mathematics, 50 technical science in general, 85 business administration, organizational science
Programme:Industrial Engineering and Management MSc (60029)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page