Susu Xu, an assistant professor in the Department of Civil and Systems Engineering, is a recipient of the National Science Foundation’s CAREER Award, which recognizes early-stage scholars with high levels of promise and excellence.
Rooted in mobile sensing, machine learning, and urban computing, Xu’s research enables smart urban infrastructure systems that address fairness in resource allocation and accessibility and enhance rapid disaster response. Her five-year award will support her project “Open-source Probabilistic AI Infrastructure for Dynamic Cascading Disaster Impacts Modeling and Assessment,” which focuses on developing AI-driven software to model and evaluate the cascading impacts of natural disasters, with the goal of improving emergency management practices.
“No-notice and short-notice natural hazards, like earthquakes and hurricanes, often cause cascading impacts, but deciphering the location, severity, and compounding impacts of disasters is a challenge for end users, like emergency management agencies and researchers,” Xu said. “This project proposes a new cyberinfrastructure that models dynamic, cascading impacts to improve disaster response and resilience.”
The project proposes a new probabilistic AI infrastructure—a type of AI that incorporates uncertainty and probability into a decision-making process—designed to advance fast, flexible, and scalable modeling of variables in disaster-induced hazards. Data used in the model will include information from a variety of sources, like photos, sensors, and videos, spanning a wide range of scales and levels of detail.
Xu said, “The probabilistic AI model will integrate geophysical and civil infrastructure knowledge, making disaster-focused modeling and inference easy to use for hazard researchers and practitioners.”
Xu’s project aims to overcome barriers in data, modeling, computation, and programming to provide accurate and high-resolution estimates of disaster-induced hazards at a regional scale, which can be readily adopted by decision-makers to improve emergency response and community resilience.
The project includes education and outreach activities, like the co-creation of high school teaching modules with art and STEM teachers and recruitment of high school and college students from diverse backgrounds to assist with research.
“I’m focusing on attracting first-generation college students and underrepresented minorities to computational science. I want to increase their awareness of natural hazards, foster their interest in the field, promote diversity and inclusion in the field, and prepare high school and college students to address disaster resilience challenges,” said Xu.