Integrating Machine Learning and Optimization With Spatiotemporal Techniques to Develop a Methodology for Assessing Rural Resilience

Details

• Creators:
  Ozguven, Eren Erman ; Moses, Ren ; Takyi, Samuel ; Antwi, Richard
• Corporate Creators:
  Florida A&M University-Florida State University (FAMU-FSU). College of Engineering. Department of Civil Engineering ; Rural Safe Efficient Advanced Transportation (R-SEAT) Center
• Corporate Contributors:
  United States. Department of Transportation. University Transportation Centers (UTC) Program ; United States. Department of Transportation. Office of the Assistant Secretary for Research and Technology
• Subject/TRT Terms:
  • Machine learning
  • Traffic lanes
  • Lane closure
  • Geospatial data
  • Disaster resilience
  • Satellite communication
  • Infrared imagery
  • Rural areas
  • Geographic information systems
  • Hurricanes
  More +
• Resource Type:
  Tech Report
• Geographical Coverage:
  United States ; Florida
• Edition:
  Final Report: 06/01/2023 - 04/05/2025
• Corporate Publisher:
  Rural Safe Efficient Advanced Transportation (R-SEAT) Center
• Description:
  This project develops an integrated methodology to assess and enhance the resilience of rural communities against natural disasters, with a focus on hurricanes and tornadoes in Florida. By combining Geographical Information Systems (GIS)-based spatiotemporal analysis, machine learning, and optimization techniques, the project addresses the unique challenges faced by rural areas, which are often underserved and disproportionately affected by extreme weather events. The methodology leverages high-resolution satellite imagery, remote sensing, and geospatial data to evaluate the impacts of disasters on transportation infrastructure and impacted populations, such as the elderly. The project builds on three key components: (1) a machine learning-based framework for assessing roadway closures and accessibility using satellite imagery, as demonstrated in the analysis of Hurricanes Ian and Idalia, which achieved 89% accuracy in classifying roadway conditions; (2) the development of indices such as the Road Closure Impact Index (RCII) and Roadway Vulnerability Index (RVI) to quantify infrastructure disruptions and vulnerabilities; and (3) an integrated GIS and image processing approach for post-tornado debris detection, utilizing spectral indices like the Normalized Difference Vegetation Index (NDVI) to identify high-impact areas and inform recovery efforts. The outcomes of this project provide actionable insights for prioritizing infrastructure restoration, improving situational awareness, and guiding resource allocation during disaster response. By focusing on Florida's Panhandle as a test bed, the project extends knowledge on rural resilience and informs strategic adaptation plans. Aligned with USDOT priorities, this work supports advanced asset management by leveraging cutting-edge technologies to enhance the resilience of transportation systems in disaster-prone regions.
• Format:
  PDF
• Funding:
  69A3552348321
• Download URL:
  https://rosap.ntl.bts.gov/view/dot/94139/dot_94139_DS1.pdf
• File Type:
  [PDF - 6.87 MB]
• Collection(s):
  University Transportation Centers
• Main Document Checksum:
  urn:sha-512:3eca25f6ceb48f8c1e8c41a7ae3084be361fda90704c265bc162f68617f380383c2ca47de0cbff46eddf5502c97420baa6f334351bcd0d0fdf9cc96a32bcff95