Lei Wang
University of Cincinnati, OH·Engineering
Civil engineers must make safety decisions about buildings, bridges and roads while knowing only part of what lies underground, and communities living with natural hazards such as landslides need tools to prepare for them. Two projects connect these concerns: one teaches future geotechnical engineers to reason about uncertainty and use artificial intelligence (AI) responsibly, and the other builds partnerships with small communities in Puerto Rico to understand and reduce landslide risk.
- Fields
Engineering
Education Research
Earth & Environmental Science
- Topics
geotechnical engineering
engineering education
artificial intelligence
- Grants
- 2 known grants, through Aug 2028
What they work on
- Test how AI-integrated, uncertainty-explicit instruction in a Foundation Engineering course affects students' ability to identify, propagate and communicate uncertainty, using tools such as live uncertainty visualizations and simple Monte Carlo simulations.
- Evaluate how structured AI modules, including a Critical-AI Handbook with Think-Verify-Decide checklists and a Prompt and Privacy Clinic, influence responsible AI use and geotechnical design performance, using quantitative and qualitative analyses.
- Establish partnerships with federal and local agencies and rural, isolated and small communities in southwestern Puerto Rico to share data, resources and local knowledge about landslide hazards.
- Co-produce research questions and solutions on landslide risk propagation, perception and mitigation with community members through surveys, interviews, workshops and focus groups, and set up feedback channels for ongoing input.
More about the research
The first project studies how undergraduate engineering students can treat AI as a careful reasoning partner rather than a shortcut to answers. Geotechnical engineering, the study of soil and rock beneath structures, depends on limited and variable field data, so small changes in assumptions can change a design's safety and cost. In the Foundation Engineering course at the University of Cincinnati, the project will build an AI-supported learning environment guided by Kolb's Experiential Learning Theory and a proposed framework called AIM-UR (AI-Supported Metacognitive Framework for Uncertainty Reasoning). Tools include AI explanation cards, live uncertainty visualizations, simple Monte Carlo simulations (repeated random trials that show how uncertain inputs affect results), a course-site AI co-mentor, a Prompt and Privacy Clinic, Design Dilemma Studios and a Critical-AI Handbook with Think-Verify-Decide checklists, all built into realistic site characterization and foundation design projects. Quantitative, qualitative and mixed-methods analyses will test effects on students' uncertainty reasoning, responsible AI use and design decisions. Outputs include open instructional modules, the handbook, rubrics and decision templates.
Why it matters
Engineering firms and the public rely on geotechnical engineers to design safe foundations despite uncertain ground conditions; open teaching tools that train students to check AI outputs and reason about uncertainty could help keep infrastructure safe as AI enters practice. Separately, rural and small communities in Puerto Rico that are exposed to landslides could gain locally tailored, community-designed strategies for risk assessment and mitigation, an approach that may transfer to other hazard-prone regions.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Research Initiation: Building AI Literacy and Uncertainty Awareness for Digitally Competent Geotechnical Engineering Education
NSF · Sep 2026 – Aug 2028
Collaborative Research: Planning: CHIRRP: Understanding Risk Propagation of Landslide-Community System for Community-Centric Risk Assessment and Mitigation
NSF · Oct 2024 – Sep 2027
Closest labs
Avantika Gori
Rice University, TX·Engineering
Develop a flood-risk framework based on performance-based design, which judges solutions by how well they meet stated goals, and apply it in two rural Texas communities.
Richard Bixler
University of Texas at Austin, TX·Social Science
Identify at-risk Texas communities using social vulnerability data and mapping tools, then form reciprocal partnerships with 2-3 place-based communities to co-produce a resilience research agenda.
Marin Clark
University of Michigan, MI·Earth & Environmental Science
Develop a scientific framework that characterizes cascading land surface hazards, such as landslides that dam rivers and sediment buildup that promotes flooding, and anticipates how they will change.
The school
University of Cincinnati: 18 labs on PI Pages, $740M a year on research, 275 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.