EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

2895. Patient reallocation for waiting time reduction in emergency departments within multi-hospital networks

Invited abstract in session WA-10: ED logistics, stream OR in Health Services (ORAHS).

Wednesday, 8:30-10:00
Room: 11 (building: 116)

Authors (first author is the speaker)

1. Dario Nicola Marchese
Management, Information and Production Engineering, University of Bergamo
2. Sebastian Birolini
Department of Management, Information, and Production Engineering, University of Bergamo
3. Mattia Cattaneo
Department of Management, Information and Production Engineering, University of Bergamo

Abstract

The timely provision of emergency care has emerged as a significant challenge due to a consistent rise in the number of visits to emergency departments each year. However, there has been no corresponding expansion in hospital resources and infrastructure, resulting in a high level of overcrowding. One of the primary reasons can be attributed to non-urgent patient visits, which account for 80% of the total accesses. This work focuses on the patient reallocation strategy, illustrating how transferring non-urgent patients between EDs within the same multi-hospital network can effectively decrease waiting times. To implement this approach, we devise a novel multi-objective optimization model with an intrinsic two-stage structure, where arriving patients can be either admitted or diverted to another hospital according to system capacity and congestion, and subject to consistent vehicle routing and availability. The proposed model is applied to a real-case study involving a multi-hospital system in northern Italy, attaining an average reduction of over 35% in daily waiting times. Extensive numerical experiments are conducted to demonstrate the scalability of the model and quantify its benefits in various settings. Based on the experimental outcomes, we develop a machine learning algorithm which serves as a tool to quickly assess the potential of the reallocation in a given hospital network with minimal information and provides insights on the required fleet size.

Keywords

Status: accepted

Back to the list of papers