Pavement-Maintenance-GPT: Optimizing Pavement Maintenance Decisions Using Generative AI
Pavement maintenance has significant impacts on transportation safety, mobility, and asset management. However, current decision-making for prioritizing pavement repairs often relies on subjective assessments, manual reviews, or disparate datasets, causing inefficiencies, increased vehicle emissions, and occupational health risks. This project proposes “Pavement-Maintenance-GPT,” a large language model designed to optimize repair prioritization decisions using high-fidelity Ground Penetrating Radar (GPR) video log data. Pavement-Maintenance-GPT leverages advanced generative AI to simulate expert decision-making, synthesizing pavement condition data into actionable maintenance strategies. The LLM will be trained on historical data, expert judgments, and GPR metrics to replicate and enhance human diagnostic capabilities. By providing precise and efficient repair recommendations, the model significantly improves mobility efficiency by reducing unnecessary lane closures while decreasing vehicle idling and associated fuel consumption. Aligned with CHEM’s focus areas of “Occupational Health and Efficient Mobility,” this research addresses occupational hazards by minimizing workers’ exposure to construction-related risks through optimized maintenance schedules. Additionally, it promotes efficient mobility by lessening disruptions, thereby improving overall roadway mobility. The anticipated outcomes include improved resource allocation, reduced vehicle emissions, enhanced safety for workers and road users, lower lifecycle costs, and greater transportation system performance. Pavement Maintenance GPT represents a transformative advancement, providing transportation agencies with a robust, scalable, and sustainable solution for optimized infrastructure maintenance, directly contributing to safer, healthier, and more efficient mobility.
Language
- English
Project
- Status: Active
- Funding: $75,000.00
-
Contract Numbers:
69A3552348329
-
Sponsor Organizations:
Office of the Assistant Secretary for Research and Technology
University Transportation Centers Program
Department of Transportation
Washington, DC United States 20590 -
Managing Organizations:
1111 Rellis Parkway
Bryan, Texas United States 77807 -
Project Managers:
Ocon, Monica
-
Performing Organizations:
Texas A&M Transportation Institute
Texas A&M University System
3135 TAMU
College Station, TX United States 77843-3135 -
Principal Investigators:
Choi, Kunhee
- Start Date: 20250501
- Expected Completion Date: 20261031
- Actual Completion Date: 0
- USDOT Program: University Transportation Centers
- Source Data: 02-12-TTI
Subject/Index Terms
- TRT Terms: Artificial intelligence; Data analysis; Decision support systems; Ground penetrating radar; Pavement maintenance; Video
- Subject Areas: Data and Information Technology; Highways; Maintenance and Preservation; Planning and Forecasting;
Filing Info
- Accession Number: 01976229
- Record Type: Research project
- Source Agency: Center for Advancing Research in Transportation Emissions, Energy, and Health (CARTEEH)
- Contract Numbers: 69A3552348329
- Files: UTC, RIP
- Created Date: Jan 13 2026 3:10PM