摘要

建筑物在各类紧急情况中面临安全威胁，而应急疏散是保障人员安全的关键手段。不过，要提高疏散效率，就需要深入研究建筑结构特点和人员行为模式。但当前仍缺乏面向应急疏散领域数字孪生技术的系统性综述。为此，本研究采用 PRISMA 研究方法，梳理分析了 2004 年至 2025 年期间的相关文献，首次提出了一个融合建筑本体、人员个体以及两者互动关系的数字孪生概念框架，该框架覆盖了从数据感知、模型更新、模拟推演到决策支持的完整闭环流程。研究表明，当前研究已在虚拟空间基础建模、单向数据映射和初步的双向交互等方面取得重要进展。但整体来看，数字孪生技术在应急疏散中的研究和应用仍处于发展阶段，大部分成果停留在 L0 到 L2 成熟度等级（注：L0-L5 为技术成熟度分级体系），而达到 L4-L5 高成熟度的实际应用还比较少见。研究指出，未来基于数字孪生的疏散系统发展需要依靠多学科交叉合作实现突破，具体包括：完善底层机理以提升数据与系统的融合能力；提高传感器精度并开发自适应的仿真、分析与预测算法；在融合新兴人工智能技术的同时解决数据伦理问题；提升计算效率以增强系统鲁棒性（即稳定性和抗干扰能力）。

Buildings face threats from various emergencies, with emergency evacuation being a key measure for occupant safety. However, enhancing evacuation efficiency necessitates detailed studies of building characteristics and human behaviors. Despite this, a systematic review of digital twin technologies for emergency evacuation is still lacking. Therefore, by collecting and analyzing literature from 2004 to 2025 using PRISMA methodology, this study first proposes a conceptual digital twin framework that integrates buildings, occupants, and their interactions, encompassing the entire loop of sensing, updating, simulation, and decision-making. The current research has made significant progresses in areas such as basic virtual modeling, one-way data mapping, and preliminary bidirectional interaction. However, studies and applications of digital twins remain in the developmental stage, with most at maturity levels L0-L2, while L4-L5 applications are still relatively scarce. It is suggested that the future development of digital twin-based evacuation systems must rely on multidisciplinary collaboration to achieve breakthroughs, including optimizing underlying mechanisms to enhance data and system integration; improving sensing accuracy and developing adaptive algorithms for simulation and prediction; integrating emerging artificial intelligence technologies while addressing data ethics; and enhancing computational efficiency to strengthen system robustness.

论文下载链接

预印本下载链接

The authors are grateful for the financial support received from the National Key Technology R&D Program (No. 2023YFC3805800) and the National Natural Science Foundation of China (No. 52378306, No. 72091512).

相关项目资助情况: