Incorporating Large Language Models (LLMs) into Transportation Safety Analytics

Details

• Creators:
  Tang, Jinwei ; Luo, Yuebo ; Li, Tianxin ; Wang, Kai ; Ding, Caiwen
• Corporate Creators:
  New England Transportation Consortium ; University of Massachusetts at Amherst
• Corporate Contributors:
  United States. Department of Transportation. University Transportation Centers (UTC) Program ; United States. Department of Transportation. Office of the Assistant Secretary for Research and Technology
• Subject/TRT Terms:
  • Algorithms
  • Crash data
  • Data analysis
  • Equity
  • Language
  • Transportation safety
• Publication/ Report Number:
  161159
• Resource Type:
  Tech Report
• Geographical Coverage:
  United States
• Edition:
  Final Report
• Corporate Publisher:
  New England Transportation Consortium
• Abstract:
  This project focuses on enhancing transportation safety through LLMs. Its goal is to improve crash data analysis by highlighting disparities in traffic safety, particularly for vulnerable road users. Leveraging the Connecticut Crash Data Repository, the project develops an LLM algorithm to automatically analyze crash patterns and generate visual reports. This innovative approach by our interdisciplinary team combines expertise in machine learning and transportation safety, ensuring a comprehensive approach to these challenges. The project's outcomes are expected to influence policy and planning decisions, fostering safer transportation systems.
• Format:
  PDF
• Funding:
  69A3552348301
• Download URL:
  https://rosap.ntl.bts.gov/view/dot/86801/dot_86801_DS1.pdf
• File Type:
  [PDF - 749.25 KB]
• Collection(s):
  Artificial Intelligence Deployment Center of Excellence (AI COE) ; University Transportation Centers
• Main Document Checksum:
  urn:sha-512:2c70614d159b34fd9c7c63da822cd283c01468840f1bcc55e03132f14344a3f8ac5de4958ca0da11c3f5cb6852e535b35b13f041d3c32821212d26d4f6e70232