RoadAI road surface condition scores may differ from past manual ASTM survey results

If a RoadAI-based road surface condition score for a road section looks unexpectedly high or low compared to the previously reported or projected values, check if one or any combination of the following cases apply:

  1. The road segment has limited survey coverage. The short section analyzed by RoadAI does not represent correctly the average condition of the road.
    Reports exclude road sections with less than 30 % coverage.
  2. The previous ASTM survey may have been sample-based. The sample chosen from this road did not represent the actual condition of the road as a whole.
  3. The road was treated after the previous survey. The projection may not take the repairs into account due to, for example, missing maintenance data in the asset management system.
  4. The road is made of concrete. Currently, RoadAI does not support ASTM reporting for concrete road distresses.
  5. The previous road surface condition score is heavily impacted by a distress type that is not supported by RoadAI (for example, weathering).
  6. The RoadAI result is impacted by sub-par video quality due to poor recording conditions (for example, direct sunlight towards the camera, a reflection from the windshield, foggy weather, or similar)

The best way to investigate the potential reasons is to:

  1. Create a Road segment group report that uses the same section division as the asset management system uses (for example, street name).
  2. Confirm that the road is properly covered by RoadAI (see case 1) by checking the coverage percentage in the report.
  3. Open the video hyperlink in the report and drive the road virtually to investigate if cases 2 to 6 could explain the difference.

If none of the reasons above seem to explain the difference, contact your RoadAI customer success manager. Include a link to the road section in RoadAI and describe why the road surface condition score value is either higher or lower than what it should be.