Enhancing FDOT Standard Specifications and Construction Guidance with Large Language Models

The main objective of this research is as follows: (1) perform an industry survey of large language models (LLM) that meet Florida Department of Transportation (FDOT) artificial intelligence (AI) policy objectives; (2) test an AI system that can rapidly extract and deliver accurate information from FDOT Standard Specifications for Road and Bridge Construction; (3) document the performance of the LLM, including accuracy in citing source materials, frequency of hallucinations, and robustness is use of industry jargon; (4) enable real-time querying of the FDOT specifications, both in the field and office, through mobile or web applications; and (5) ensure the AI system can interpret the specifications accurately and provide users with clear, concise, and legally sound interpretations of FDOT requirements.

Language

• English

Project

• Status: Active
• Funding: $136,900.00
• Contract Numbers:

  BEJ23

• Sponsor Organizations:

  Florida Department of Transportation

  Research Center
  605 Suwannee Street MS-30
  Tallahassee, FL  United States  32399-0450
• Managing Organizations:

  Florida Department of Transportation

  Research Center
  605 Suwannee Street MS-30
  Tallahassee, FL  United States  32399-0450
• Project Managers:

  Chung, Oliver

• Performing Organizations:

  Florida Department of Transportation

  Research Center
  605 Suwannee Street MS-30
  Tallahassee, FL  United States  32399-0450
• Principal Investigators:

  Wetherell, Ryan

• Start Date: 20250626
• Expected Completion Date: 20260630
• Actual Completion Date: 0
• USDOT Program: Real-Time Transit Infrastructure and Rolling Stock Condition Assessment Demonstration Program

Subject/Index Terms

• TRT Terms: Artificial intelligence; Bridge construction; Information dissemination; Road construction; Specifications
• Identifier Terms: Florida Department of Transportation
• Subject Areas: Bridges and other structures; Construction; Data and Information Technology; Highways;

Filing Info

• Accession Number: 01959242
• Record Type: Research project
• Source Agency: Florida Department of Transportation
• Contract Numbers: BEJ23
• Files: RIP, STATEDOT
• Created Date: Jun 26 2025 1:36PM