Effectively Implementing Machine Learning within Office of Materials Technology

Every year the Maryland State Highway Administration (SHA) invests millions of dollars into testing the states geomaterials to optimize engineering designs. There is a significant opportunity for cost savings by leveraging historic material testing data with predicative machine learning models to provide estimated values as well as gaining a better understanding of historic data. Every year MDOT invests millions of dollars into testing geomaterials and thus massive amounts of engineering datasets as well as other data such as roadway and construction data have been accumulated over a long time period. There are many advantages of the machine learning-based approaches for field inspection/testing and ARAN imagery data modeling and prediction. Significant cost savings can be achieved by leveraging historical datasets and integrating with machine learning enabled work process. If implemented, not only the material and condition characteristics of highway, and other transportation infrastructures can be estimated in the early phase of the project, but also scheduling/construction and maintenance can be optimized by data driven decision-making assistance enabled by accurate prediction with cutting edge machine learning methods and continual self-supervised learning of future incoming data.

Language

• English

Project

• Status: Completed
• Funding: $194945
• Contract Numbers:

  SPR21B4E

• Sponsor Organizations:

  Maryland Department of Transportation State Highway Administration

  707 North Calvert Street
  Baltimore, MD  United States  21202
• Managing Organizations:

  Maryland Department of Transportation

  State Highway Administration, Office of Policy and Research
  707 North Calvert Street
  Baltimore, MD  United States  21202
• Project Managers:

  Herrera Riggs, Saskia

• Performing Organizations:

  University of Maryland College Park, Department of Civil and Environmental Engineering

  ,    
• Principal Investigators:

  Zhang, Yunfeng

• Start Date: 20210428
• Expected Completion Date: 20230428
• Actual Completion Date: 20230509

Subject/Index Terms

• TRT Terms: Databases; Highways; Implementation; Information processing; Machine learning; Materials
• Identifier Terms: Maryland Department of Transportation
• Subject Areas: Data and Information Technology; Highways; Maintenance and Preservation; Planning and Forecasting;

Filing Info

• Accession Number: 01866358
• Record Type: Research project
• Source Agency: Maryland State Highway Administration
• Contract Numbers: SPR21B4E
• Files: RIP
• Created Date: Dec 1 2022 10:55AM