The importance of an optimal solution for disaster evacuation has recently raised attention from researchers across multiple disciplines. This is not only a serious, but also a challenging task due to the complexities of the evacuees’ behaviors, route planning, and demanding coordination services. Although existing studies have addressed these challenges to some extent, mass evacuation in natural disasters tends to be difficult to predict and manage due to the limitation of the underlying models to capture realistic situations. It is therefore desirable to have on-demand mechanisms of locally-driven computing and data exchange services in order to enable near real-time capture of the disaster area during the evacuation. For this purpose, this paper comprehensively surveys recent advances in information and communication technology-enabled disaster evacuations, with the focus on fog computation and communication services to support a massive evacuation process. A numerous variety of tools and techniques are encapsulated within a coordinated on-demand strategy of an evacuation platform, which is aimed to provide a situational awareness and response. Herein fog services appear to be one of the viable options for responsive mass evacuation because they enable low latency data processing and dissemination. They can additionally provide data analytics support for autonomous learning for both the short-term guidance supports and long-term usages. This work extends the existing data-oriented framework by outlining comprehensive functionalities and providing seamless integration. We review the principles, challenges, and future direction of the state-of-the-art strategies proposed to sit within each functionality. Taken together, this survey highlights the importance of adaptive coordination and reconfiguration within the fog services to facilitate responsive mass evacuations as well as open up new research challenges associated with analytics-embedding network and computation, which is critical for any disaster recovery activities.

灾难疏散的最佳解决方案的重要性最近引起了多个学科研究人员的关注。由于疏散人员行为、路线规划和协调服务要求的复杂性，这不仅是一项严肃的任务，而且是一项具有挑战性的任务。尽管现有研究在一定程度上解决了这些挑战，但由于底层模型无法捕捉现实情况，自然灾害中的大规模疏散往往难以预测和管理。因此，希望有本地驱动的计算和数据交换服务的按需机制，以便在疏散期间能够近乎实时地捕获灾区。以此目的，本文全面调查了信息和通信技术支持的灾难疏散的最新进展，重点是雾计算和通信服务，以支持大规模疏散过程。多种工具和技术封装在疏散平台的协调按需策略中，旨在提供态势感知和响应。在这里，雾服务似乎是响应式大规模疏散的可行选择之一，因为它们能够实现低延迟数据处理和传播。他们还可以为短期指导支持和长期使用的自主学习提供数据分析支持。这项工作通过概述全面的功能并提供无缝集成来扩展现有的面向数据的框架。我们回顾了每个功能中提议的最先进策略的原则、挑战和未来方向。总之，这项调查强调了雾服务中自适应协调和重新配置的重要性，以促进响应式大规模疏散，并开辟与分析嵌入网络和计算相关的新研究挑战，这对任何灾难恢复活动都至关重要。