Citation: Moyaux, T.; Liu, Y.; Bouleux, G.; Cheutet, V. An Agent-Based Architecture of the Digital Twin for an Emergency Department. Sustainability 2023, 15, 3412. https://doi.org/10.3390/ su15043412 Academic Editors: Robert Pellerin, Samir Lamouri and Pascal Forget Received: 10 January 2023 Revised: 25 January 2023 Accepted: 8 February 2023 Published: 13 February 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). sustainability Article An Agent-Based Architecture of the Digital Twin for an Emergency Department Thierry Moyaux 1,† , Yinling Liu 2,† , Guillaume Bouleux 1,† and Vincent Cheutet 1, * ,† 1 University Lyon, INSA Lyon, University Jean Monnet Saint-Etienne, University Claude Bernard Lyon 1, University Lumière·Lyon 2, DISP UR4570, 69621 Villeurbanne, France 2 University Lorraine, CNRS, CRAN UMR 7039, University Lorraine, 54000 Nancy, France * Correspondence: vincent.cheutet@insa-lyon.fr; Tel.: +33-4-72-43-75-37 † These authors contributed equally to this work. Abstract: The concept of Digital Twin (DT) seems promising to improve the management of patient pathways in Emergency Departments (EDs). This article proposes an agent-based architecture of a DT designed for that purpose. The core of this DT is its Information System (IS), which is regularly synchronised on the IS of the Physical Twin (PT). The agents model the ED’s resources (equipment and staff) and patients in the DT and update this information in the DT’s IS. This article shows how such a DT may operate in three modes: (0) “Digital Shadow” to monitor the ED’s current state in real time, (1) “Synchronised DT” to monitor the ED’s current and future states according to a predictive simulation, and (2) “Exploratory DT” in order to perform Monte Carlo simulations of various future states. Mode (1) is the main contribution. This proposition is illustrated in a simulation of the ED in order to demonstrate the capabilities. Keywords: architecture; digital twin; agent-based simulation (ABS); patient pathway 1. Introduction Improving patient flows in organisations or care pathways has been a topic of increas- ing interest for decades. The multitude of actors and their interactions are such that it is still difficult to adequately address this issue. The hospital and its gateway, the Emergency Department (ED), is therefore the main system in which a significant improvement of the patient pathway may be envisaged. Today, with the renewed interest in Artificial Intelligence methods, a new era in the management of hospital systems organisation and patient flow is arising. This digital transformation is often likened to that of Industry 4.0. Of all the concepts and approaches encompassed in the Industry 4.0 paradigm, Digital Twin (DT) is a particularly promising path to improve ED performance. It offers faster access to data captured on the floor and provides a digital support framework enhanced with simulations and data analysis [1]. An abundant literature exists on DT applied to various domains [2,3]. One definition is that “DT consists of a virtual representation of a production system that is able to run on different simulation disciplines, characterised by the synchronisation between the [digital] and [physical] systems, by means of data and connected smart devices, mathematical models and real time data elaboration”[4]. DTs are used when there is a need for (i) a control as close as possible to the physical system and (ii) an anticipation of its behaviour, through the integration of simulation. This work is based on Kritzinger et al.’s one [5] which defines a DT as a digital system with an automatic data flow both from and to the PT. In their definitions, the PT and physical system are synonymous, but not DT and digital system. For instance, a Digital Shadow is a digital system monitoring a PT that is not a DT. The same definitions as [5] are used here, except for DT which is refined as “Synchronised DT” and “Exploratory DT”. In the aforementioned work [5], the data flows between the DT and the PT are au- tomatic, implying an absence of decisions taken by a stakeholder. Nevertheless, due to Sustainability 2023, 15, 3412. https://doi.org/10.3390/su15043412 https://www.mdpi.com/journal/sustainability