Al Analyzer for Revealing Insights of Traffic Crashes

The developed tool is a software solution that utilizes artificial intelligence (AI) to automatically analyze traffic crash narratives, assisting traffic safety engineers in their work. By processing a substantial volume of crash narratives, the software extracts factors associated with crash severity, providing valuable insights for comprehensive crash understanding and data-driven decision making. Through its ability to minimize manual intervention, the tool efficiently analyzes crash narratives and generates potential contributors to crash severity in the form of phrases. The research comprised two stages. In the initial stage, an AI analysis approach was developed by implementing an AI text classifier that balances predictive performance and computational complexity. This approach was combined with Explainable-AI techniques to identify phrases that correlate with severe crashes. Subsequently, the AI analysis approach was integrated into a user-friendly web-based software tool, simplifying the extraction of insights from crash narratives. In the second stage, the software was validated by comparing its results with those obtained from classical statistical analysis on quantitative crash data. Feedback from safety analysts at partner agencies was incorporated to refine the software. To facilitate the implementation of the solution, the software was released under an open-source license, allowing transportation agencies to freely download and analyze their own crash narratives. Additionally, a webinar, website, and documentation were created to showcase the tool, guide its usage, and support the integration of evolving techniques. The research findings were also shared through transportation conferences. The proposed solution delivers significant value to the transportation community by empowering analysts to utilize crash narratives as a valuable data source for traffic safety analysis. Analysts can identify potential contributors to crash severity without manually reading each crash narrative. This improved identification of contributing factors can assist researchers and policy makers in designing targeted countermeasures to enhance safety. The final report is available.

• Record URL:
  http://apps.trb.org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=5191

Language

• English

Project

• Status: Completed
• Funding: $82899
• Contract Numbers:

  Project 20-30, IDEA 231

• Sponsor Organizations:

  Safety Innovations Deserving Exploratory Analysis (IDEA)

  Transportation Research Board
  500 Fifth Street, NW
  Washington, DC  United States  20001

  National Cooperative Highway Research Program

  Transportation Research Board
  500 Fifth Street, NW
  Washington, DC  United States  20001

  American Association of State Highway and Transportation Officials (AASHTO)

  444 North Capitol Street, NW
  Washington, DC  United States  20001

  Federal Highway Administration

  1200 New Jersey Avenue, SE
  Washington, DC  United States  20590
• Project Managers:

  Jawed, Inam

• Performing Organizations:

  University of Nevada at Las Vegas

  ,    
• Principal Investigators:

  Park, Jee

• Start Date: 20210701
• Expected Completion Date: 20230630
• Actual Completion Date: 0

Subject/Index Terms

• TRT Terms: Artificial intelligence; Crash analysis; Crash causes; Crash severity; Literature reviews; Software; State of the art; Traffic crashes; Traffic safety; Validation
• Subject Areas: Data and Information Technology; Highways; Safety and Human Factors;

Filing Info

• Accession Number: 01776582
• Record Type: Research project
• Source Agency: Transportation Research Board
• Contract Numbers: Project 20-30, IDEA 231
• Files: TRB, RIP
• Created Date: Jul 14 2021 5:38PM