Pavement condition mobile mapping
We fused laser crack-measurement data, IMU-located imagery and IRI profiles into a per-100 m pavement condition database with automated defect proposals, halving rater review load while keeping recall above 95% across 2,800 km of arterials.
The situation.
Regional transport contractor responsible for condition grading of 2,800 km of arterial roads.
ScopeFusion of laser crack-measurement data, IMU-located imagery and IRI profiles into a per-100 m segment condition database with automated defect proposals (cracking, ravelling, potholes) and a review workstation for certified raters.
Network-wide condition refresh within 48 h of survey completion, defensible for maintenance budget hearings.
What made it hard.
Defect proposals overwhelmed raters with false positives in patched sections: we trained the proposal filter on the contractor’s own accepted grades, halving review load while keeping recall above 95%.
Common questions.
What was the scope?
Fusion of laser crack-measurement data, IMU-located imagery and IRI profiles into a per-100 m segment condition database with automated defect proposals (cracking, ravelling, potholes) and a review workstation for certified raters.
Who was the client?
Client names are withheld under confidentiality agreements. References are available on request.
What was the outcome?
Network-wide condition refresh within 48 h of survey completion, defensible for maintenance budget hearings.
Data & MLOps
Ingestion, warehousing, pipelines, model serving, drift monitoring.
Explore this capability →This is one of roughly fifty engagements we ship per year. The full record, 2011 to today, lives on the delivery history page.