Incorporating Large Language Models (LLMs) into Transportation Safety Analytics

This project pioneers the integration of Large Language Models (LLMs) into transportation safety analytics to advance equity and improve data accessibility. Leveraging the Connecticut Crash Data Repository (CTCDR), the research team developed a novel system that enables users—regardless of technical background—to perform complex crash data analysis and generate visual reports using natural language queries. This LLM-powered interface streamlines data retrieval, reduces the need for specialized training, and facilitates the identification of safety disparities impacting vulnerable road users and underserved populations. By automating data cleaning, classification, and visualization, the system offers a cost-effective and scalable solution for state Departments of Transportation and policymakers seeking to meet federal safety and equity goals. The project demonstrates the transformative potential of AI in democratizing access to transportation data and shaping more informed safety interventions.

Language

• English

Project

• Status: Completed
• Funding: $106000
• Contract Numbers:

  69A3552348301

• 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:

  University of Massachusetts, Amherst

  Department of Civil and Environmental Engineering
  130 Natural Resources Road
  Amherst, MA  United States  01003
• Performing Organizations:

  University of Connecticut, Storrs

  Connecticut Transportation Institute
  270 Middle Turnpike, Unit 5202
  Storrs, CT  United States  06269-5202
• Principal Investigators:

  Ding, Caiwen

  Wang, Kai

• Start Date: 20240101
• Expected Completion Date: 20241231
• Actual Completion Date: 0
• USDOT Program: University Transportation Centers Program
• Subprogram: University Transportation Centers

Subject/Index Terms

• TRT Terms: Algorithms; Crash data; Data analysis; Equity; Language; Transportation safety
• Geographic Terms: Connecticut
• Subject Areas: Planning and Forecasting; Safety and Human Factors; Transportation (General);

Filing Info

• Accession Number: 01904458
• Record Type: Research project
• Source Agency: New England University Transportation Center
• Contract Numbers: 69A3552348301
• Files: UTC, RIP
• Created Date: Jan 12 2024 10:46AM