NSF Convergence Accelerator: Symposium on Big Data and AI-Driven Disaster Management for Planning, Response, Recovery, and Resiliency

本项目支持组织和举办研讨会，以扩大希望更有效地利用大数据和人工智能工具更有效地准备、应对和恢复自然灾害的利益相关者之间的合作。仅在2017年，美国就经历了16次以上的自然灾害，每次造成的损失都超过10亿美元，许多预测表明灾害会更频繁、更严重。正因为如此，组织和研究努力的方向是改进计划、即时反应和对自然灾害的恢复能力。然而，这些努力往往是基于灾害类型（例如野火与山体滑坡）而孤立的。另一个挑战是理解和利用技术能力的爆炸式增长，这些技术能力提供了如此多的新数据源，包括物理（如传感器、物联网工具、无人机）、网络（如灾害数据库、公开政府数据）和各种模式（文本/图像/视频）的社交（如微博流）数据。本次研讨会将确定将大数据、人工智能和机器学习工具转化为适用于所有类型灾害管理的可操作知识所需的未来研究和组织间合作领域。本次研讨会将汇集30-40名利益相关者，包括从业人员（如急救人员、地方/州/联邦政府）、学术界（包括计算机、社会、物理科学领域）。以及工程)和工业（营利性和非营利性）讨论从规划到响应到恢复的所有阶段的灾难恢复能力的信息和数据技术需求。研讨会将确定主要的研究挑战/机会，以及研究领域在短期内转化为实际应用的潜力。这些见解将有助于描述该领域作为NSF融合加速器未来主题的潜力。大数据与人工智能（AI）工具相结合，可以提高备灾和救灾的生产力、效率和决策能力，但大数据和人工智能也与我们社会面临的许多其他挑战相关。因此，本次研讨会还应提供一个示例，说明如何将不同的数据和工具集成到规划和决策支持中。这项工作建立在NSF其他项目的投资基础上，包括关键弹性相互依赖基础设施系统和过程（CRISP）、灾害和灾害跨学科研究（Hazard-SEES）和智能连接社区（S&amp；CC），以及许多其他美国联邦机构的投资。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

This project supports efforts to organize and host a workshop to expand collaboration between stakeholders wishing to more effectively use big data and artificial intelligence tools to more effectively prepare for, respond to, and recover from natural disasters. The United States experienced more than sixteen natural hazards causing greater than a billion dollars of impact each in 2017 alone, and many predictions point to more frequent and more severe disasters. Because of this organizational and research efforts are directed toward improving planning, immediate response, and resiliency against natural disasters. However, those efforts are often siloed based on the type of disaster (e.g. wildfires vs. landslides). An additional challenge is understanding and using the explosion of technological capabilities that provide so many new data sources, including physical (e.g., sensors, internet-of-things tools, drones), cyber (e.g., hazard databases, open government data), and social (e.g. microblog streams) data of various modalities (text/images/videos). This workshop will identify areas of future research and inter-organizational collaboration that are needed to transform big data, artificial intelligence, and machine learning tools into actionable knowledge for all types of disaster management. This workshop will bring together 30-40 stakeholders including practitioners (e.g. first responders, local/state/federal government), academia (including fields of computer, social, physical sciences. and engineering), and industry (for-profit, non-profit) to discuss the information and data technology needs for all phases of disaster resilience, from planning to response to recovery. The workshop will identify major research challenges / opportunities and the potential for the research areas to transition to practical use in the short term. These insights will help describe the potential for this area to serve as a future topic for the NSF Convergence Accelerator. Big data harnessed with artificial intelligence (AI) tools are needed to improve productivity, efficiency, and decision making in disaster preparedness and response, but big data and AI are also relevant to many other challenges our society is facing. Therefore, this workshop should also provide an example of how disparate data and tools may be integrated for planning and decision support. This effort builds upon investments made by other programs at NSF, including Critical Resilient Interdependent Infrastructure Systems and Processes (CRISP), Interdisciplinary Research in Hazards and Disasters (Hazard-SEES), and Smart & Connected Communities (S&CC), as well as investments by many other U.S. federal agencies.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.