This dissertation develops and tests a new data-driven framework for short-term roadway pluvial flash flood (PFF) risk estimation at the scale of road segments using crowdsourced navigation data and a simplified physics-based PFF model. Pluvial flash flooding (PFF) is defined as localized floods caused by an overwhelmed natural or engineered drainage system. This study develops a data curation and computational framework for data collection, preprocessing, and modeling to estimate the risk of PFF at road-segment scales. A hybrid approach is also developed that couples a statistical model and a simplified physics-based simulation model in a machine learning (ML) model to rapidly predict the risk of roadway PFF using Waze alerts in real-time.
Civil and environmental Engineering
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License
Safaei-Moghadam, Arefeh, "Leveraging Crowdsourced Navigation Data In Roadway Pluvial Flash Flood Prediction" (2022). Civil and Environmental Engineering Theses and Dissertations. 24.