Real-time decision-support using simulation (RtS) for operational responsiveness in urgent and emergency care: Bridging the gap between conventional simulation and digital twins
提出一种实时决策支持仿真方法，通过融合实时数据与历史分布，帮助急症护理网络在需求激增时快速恢复急诊部正常运营，对医疗管理者和运筹学从业者有用。
Conventional simulation models in healthcare are typically parameterised using historical data, limiting their ability to support responsive, real-time operational decision-making. They are therefore poorly suited to dynamic, high-pressure environments such as urgent and emergency care, where decisions must reflect the current system state. In this paper, we introduce real-time decision-support using simulation (RtS), a methodology that integrates real-time operational data into a discrete-event simulation model to enable context-aware, short-term decision-making. The model integrates partial real-time data feeds with historically derived distributions using a novel data fusion and model initialisation approach. By continuously aligning the simulation with the evolving system state, RtS enhances situation awareness and enables the evaluation of adaptive interventions whose implementation influences subsequent system behaviour. We implement the technical component of our RtS framework through a proof-of-concept model of an Urgent Care Network comprising multiple emergency and urgent care facilities. We evaluate resource-neutral operational policies designed to support short-term adaptation under conditions of rising demand. Our results demonstrate that, under the modelled high-demand conditions, normal emergency department operations can be restored within a 4-hour period through the implementation of two adaptive policies. By operationalising RtS in a service-based setting and demonstrating its effectiveness relative to conventional simulation, our findings provide healthcare stakeholders and OR practitioners with both the technical foundations and the scholarly framing needed to engage in mainstream discussions on deploying RtS as a step-change solution toward simulation enabled digital twins in complex sociotechnical systems.