CAREER: Spatial Network Database approach for Emergency Management Information Systems

应急管理信息系统(EMIS)是一种越来越重要的工具，用于了解、管理和管理与交通相关的系统，以及测试这些系统的稳定性或抗干扰性。最近，EMIS受益于志愿者地理信息(VGI)和众包作为收集用户生成的数据集的强大方法。然而，这些数据源具有挑战性，因为它们的大小、种类和更新率非常大，需要确保及时和准确地提供有用的紧急信息和对灾难性事件的响应。为先进的相关查询开发基本数据处理组件，以便在紧急情况下清楚而简明地提供关键信息，这是极其重要和具有挑战性的。本研究集中于三个相互关联的领域：1)疏散路径规划；2)资源分配；3)交通弹性。本研究以应急管理为背景，对这三个领域的创新问题进行了研究。该项目的成果对交通管理、物流、公共安全、资源分配和服务提供等广泛的社会应用具有潜在的好处，从而与NSF的使命很好地结合在一起：促进科学进步；促进国家健康、繁荣和福利。该项目的教育目标包括扩大拉美裔妇女的参与，增加本科生的研究机会，包括研究密集型课程开发，以及促进团队科学技能。该项目的目标是为应对管理信息系统的挑战找出有希望的解决办法，并开发一个先进的空间查询处理平台，以便在紧急情况下清楚而简明地提供关键信息。首先，本项目设计和开发了可以集成不同技术组件的问题解决框架，包括几何、拓扑、图论和优化技术。其次，该项目研究了空间网络的多个固有约束，并确定了查询处理的主要瓶颈。第三，该项目开发了快速、可扩展的查询处理机制，以克服这些瓶颈，并为应急管理产生简单而简明的信息。一个关键的研究挑战是识别空间网络优化问题的结构模式或最优子结构，以提高空间网络查询处理的可扩展性和效率。该查询处理框架的组件包括频繁后缀树挖掘、图化简、二部图聚类、最小多边形覆盖、图划分、谱方法、随机游走和扩展图挖掘。这些组件被集成在一起，以开发快速和可扩展的空间网络查询，并为EMIS提供简单明了的信息。该项目的成果包括数据处理工具、空间和空间网络优化算法、查询和可视化工具。该项目使用历史和实时数据集验证新的空间网络查询的性能，并提供基于网络的教育系统来增强学生的学习。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Emergency Management Information Systems (EMIS) are an increasingly important tool for understanding, managing, and governing transportation-related systems, as well as for testing the stability or vulnerability of these systems against interference. Recently, EMIS have benefitted from both volunteer geographic information (VGI) and crowdsourcing as powerful methods of collecting user-generated datasets. However, these data sources are challenging due to their very large size, variety, and update rates required to ensure the timely and accurate delivery of useful emergency information and response for disastrous events. Developing fundamental data processing components for advanced relevant queries which can clearly and succinctly deliver critical information in the case of an emergency is critically important and challenging. This research focuses on three interrelated domains: 1) evacuation route planning 2) resource assignment, and 3) transportation resilience. This research investigates innovative queries in these three domains in the context of emergency management. The outcome of this project has potential benefit to a wide range of societal applications, such as transportation management, logistics, public safety, resource assignment, and service delivery and thus aligns well with the NSF mission: to promote the progress of science; to advance the national health, prosperity and welfare . Educational objectives of this project include broadening participation of Hispanic women, increasing undergraduate research opportunities including research-intensive course development, and promotion of team science skills. The goals of this project are to identify promising solutions for addressing the challenge of EMIS and to develop an advanced spatial query processing platform that clearly and succinctly delivers critical information in emergencies. First, this project designs and develops the problem-solving framework that can integrate different technical components, including geometry, topology, graph theory, and optimization techniques. Second, this project investigates multiple inherent constraints for spatial networks and identifies main bottlenecks for query processing. Third, this project develops fast and scalable query processing mechanisms that overcome these bottlenecks and produce simple and concise information for emergency management. A key research challenge is to identify structural patterns or optimal substructures of the spatial network optimization problem that can enhance the scalability and efficiency of spatial network query processing. The components of the query processing framework include frequent suffix tree mining, graph simplification, bi-partite graph clustering, minimum polygon covering, graph partitioning, spectral method, random walk, and expander graph mining. These components are integrated to develop fast and scalable spatial network queries and to provide simple and concise information for EMIS. The outcomes of this project include data processing tools, spatial and spatial network optimization algorithms, queries, and visualization tools. This project validates the performance of new spatial network queries using historical and real-time datasets and provides a web-based educational system to enhance student learning.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.