AI-Based Environmental Code Checking Tool for Sustainability Best Management Practices of Infrastructure Construction Projects [Research Brief]

Details

• Creators:
  Kim, Joseph J. ; Chavan, Pooja D.
• Corporate Creators:
  San Jose State University. College of Business. Mineta Transportation Institute ; California State University, Long Beach
• Corporate Contributors:
  State of California SB1 2017/2018, Trustees of the California State University Sponsored Programs Administration ; United States. Department of Transportation. University Transportation Centers (UTC) Program
• Subject/TRT Terms:
  • Artificial intelligence
  • Sustainable transportation
  • Life cycle analysis
  • Infrastructure
  • Greenhouse gases
  • Best practices
  • Construction projects
  • Emissions testing
• Publication/ Report Number:
  2524
• Resource Type:
  Brief
• Geographical Coverage:
  United States
• Edition:
  Research Brief
• Corporate Publisher:
  San Jose State University. College of Business. Mineta Transportation Institute
• Description:
  This research develops, implements, and evaluates an AI-based environmental compliance checking system aligned with the Envision sustainability rating framework. The goal is to automate lifecycle greenhouse gas (GHG) assessment, reduce human error in data extraction, and improve the transparency and reproducibility of Envision credit evaluations. To accomplish this, the research combines case study analysis, NLP, life-cycle assessment (LCA), and Envision rating system's scoring into a single automated workflow. This research adopts the LCA approach of Dequidt (2012) to quantify and compare emissions against baseline scenarios. Publicly available documentation was used to develop infrastructure-specific NLP training datasets. A case study using real-world bridge lifecycle GHG data tested extraction accuracy, unit normalization, and LCA computations. This project developed a custom spaCy-based NER model to extract three key pieces of information in groups from unstructured text, such as metadata, materials and activities, and quantities. Extracted values were standardized and transferred into an LCA module that applied emission factors from established databases to compute total, annualized, and intensity-based GHG metrics. These results were mapped to Envision Credit CR1.2 to determine performance tier alignment. The system takes documents as input, extracts relevant environmental data, calculates emissions, and automatically determines the project's Envision performance level.
• Format:
  PDF
• Download URL:
  https://rosap.ntl.bts.gov/view/dot/93578/dot_93578_DS1.pdf
• File Type:
  [PDF - 295.61 KB]
• Collection(s):
  University Transportation Centers
• Main Document Checksum:
  urn:sha-512:4c13c98cb2e39911a392d37db3e5e2b486a7783bd721ab95973c98f74629dbf276a946066338e7a608d978d2a4c6156d39c9ffd45e1275f4de093d57ff9163a8