Multisource Data Fusion for Real-Time and Accurate Traffic Incident Detection via Predictive Analytics

This research will investigate how data from the various traffic data sources that Massachusetts Department of Transportation (MassDOT) owns or has access to can be merged for accurate, real-time traffic incident detection, to improve travel time reliability. It will assess the current traffic incident detection methods employed by MassDOT and develop new tools for improved traffic incident detection based on available traffic data. The research will address the fusion of information from multiple sources of different temporal and spatial scales such as traffic data collected from loop detectors; information from the MassDOT Real Time Traffic Management (RTTM) system; and information available through third-party vendors (e.g., Waze, Google, INRIX). The fusion of these sources will be accomplished through evaluating the reliability of the various data sources and deploying advanced data analytical methods such as deep neural networks.


  • English


  • Status: Active
  • Funding: $150000
  • Sponsor Organizations:

    Federal Highway Administration

    1200 New Jersey Avenue, SE
    Washington, DC  United States  20590
  • Managing Organizations:

    Massachusetts Department of Transportation

    10 Park Plaza
    Boston, MA  United States  02116
  • Project Managers:

    Flanary, Michael

  • Performing Organizations:

    University of Massachusetts Lowell

    One University Drive
    Lowell, MA  United States  01854
  • Principal Investigators:

    Stamatiadis, Chronis

    Xie, Yuanchang

    Gartner, Nathan H

  • Start Date: 20210413
  • Expected Completion Date: 20221231
  • Actual Completion Date: 0

Subject/Index Terms

Filing Info

  • Accession Number: 01771455
  • Record Type: Research project
  • Source Agency: Massachusetts Department of Transportation
  • Files: RIP, STATEDOT
  • Created Date: May 11 2021 10:11AM