Get Full Government Meeting Transcripts, Videos, & Alerts Forever!
Get email alerts on the Infrastructure Technology topic
No spam. Unsubscribe anytime.
York County to pilot AI analysis of sewer CCTV video to speed repairs
Summary
The county will pilot an AI‑assisted sewer CCTV analysis service to flag defects and integrate results with GIS; the tool is expected to speed triage and reduce staff time spent reviewing footage.
York County staff described a technology pilot to use artificial intelligence to analyze CCTV sewer-video footage and produce geotagged defect reports that can be integrated into the county’s GIS system.
Public works explained the routine process for sewer CCTV: crews capture long hours of video from pipe inspections and traditionally rely on staff to manually review footage for defects such as roots, cracks, offset joints and sediment. That review is time consuming and can introduce variability.
The county’s plan is to use a commercial AI product trained specifically to detect pipe defects. Staff said the vendor’s model is already tuned for sewer-pipe conditions and that the vendor will process batches of video, score defects (for example, grading issues by severity), geotag the locations and deliver a mapping-ready dataset. The county will retain ownership of the inspection video and the derived data; storage and processing will use cloud services (examples cited: AWS and Azure). Public works said no additional ongoing headcount is required to operate the service; the AI shortens staff review time and produces standardized triage lists so field crews can prioritize repairs.
IT and public‑works staff explained the workflow: capture video in the field, upload video to the vendor’s secure processing pipeline, receive a defect report with severity flags and coordinates, and use GIS integration to dispatch crews. The AI approach will allow the county to identify the most critical repairs (graded red/orange) quickly and plan lower‑priority work for later.
Supervisors welcomed the approach and asked staff to confirm data‑security, vendor training sources and how the AI model will be validated against local conditions. Staff said the vendor’s model is built on many customers’ data and that York County would supply its own footage to further tune the algorithm over time.
Ending: The board approved the consent calendar (which included the AI contract on the consent agenda), and public works will proceed under the contract terms to pilot the AI analysis, validate results and integrate outputs with the county GIS and work-order systems.
