ABSTRACT

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.

Please report errors in award information by writing to: awardsearch@nsf.gov.