The main objective of this paper is to provide a simulation-based decision-support tool for the healthcare industry. This tool will help hospital management decide on resource utilization, in particular bed allocation, for the next few months. With it, hospitals could predict admissions and see how newly implemented policies impact the patient’s flow.
To illustrate the tool’s ability to monitor resource utilization in a recovery unit during the pandemic, the paper uses a case study.
This project is a collaboration with Saint-Etienne University Hospital. The hospital is composed of 1,795 hospitalization beds in 60 departments, located on 3 different geographical sites across the city.
The researchers produced a prospective analysis of patients’ arrivals in the year 2019 and the first semester of 2020. Using data from the current coronavirus pandemic, they ran an experiment to demonstrate how this simulation could be used to assess the model’s response to an unexpected variation of arrivals.
The model for healthcare resource utilization considered two main sources of patients:
- Elective admissions patients.
- Non-elective and emergent patients
For elective admissions patients, the model used data available in the information system of the Saint-Etienne University Hospital. For non-elective patients, there wasn’t any information apart from the previous admissions. To model such patients, the researchers built a case-mix of previous emergent patients and randomly picked up patients in this case-mix.
The strength of this data-driven healthcare resource utilization simulation is its ability to account for a major part of the hospital. Also, given a thorough data analysis, it can be used to test different organizations of the hospital in what-if scenarios, or can be applied to other healthcare centers.