Verified Business Case · Public Administration & Safety·5 min read

Your trucks already drive every street, every week. They're just not telling you what they see.

Curbside dumping costs councils up to $750,000 a year to clean up. The fix isn't more rangers, it's getting signal out of the cameras you're already paying for.

How we turned a council's existing garbage truck cameras into an automated curbside dumping detection system.

Cleanup Cost Exposure

Up to $750K

Annual curbside dumping spend addressed

Detection Source

Existing Trucks

No new hardware, no new routes

Response Time

Sharply Reduced

From weeks of reports to next-route pickup

Run the numbers on your own bottleneck

See what this looks like inside your operation.

Plug in your headcount, salary, and weekly admin hours. See the financial bleed and the recoverable budget in real time, then bring the number to the review.

Run the calculator

The Structural Flaw

Curbside dumping, the piles of household rubbish and mattresses people leave on the street outside scheduled pickup, is a chronic problem for local councils. Cleanup alone can run up to $750,000 a year, on top of damage to infrastructure and the environment. For this council, the operational pain was worse than the dollar figure. Regular waste trucks couldn't collect the dumped piles because they weren't on the route sheet. A separate manual dispatch service handled them, but only after someone reported the dump, which meant rangers and residents were the detection layer. Reports could take weeks to surface, dispatch ran reactively instead of being route and load optimized, and by the time a truck arrived, the pile was usually bigger.

The Execution Engine

Blackbook AI built a computer vision model on top of the cameras the council's garbage trucks already carry. We trained the model on an annotated dataset drawn from historical footage from smart-enabled trucks and in-cab triggers, so it learned to recognize rubbish piles and mattresses in real curbside conditions, not staged ones. As the trucks run their normal routes, the model processes the camera feed and uses the truck's GPS and GNSS data to triangulate the exact location of each detected pile. An API then pushes those detections back to the council as street addresses ready for pickup. No new vehicles, no new routes, no new reporting burden on rangers or residents. The trucks that are already on every street every week become the detection network.

Deployed Stack

Computer Vision ModelCustom Annotated Training DatasetGPS / GNSS Location TriangulationAWSMicrosoft AzureDetection API to Council Systems

Verified Outcomes

  • Significant cost savings by letting dispatch run on route and load optimization instead of one-off reactive callouts.
  • Response time to dumped piles dropped from weeks of waiting on reports to next-route pickup.
  • Coverage improved across the council area because every truck on every route is now a detection point.
  • Reliance on rangers and residents to report dumping is sharply reduced, freeing them for higher-value work.
  • Council gained a real data feed to run targeted anti-dumping campaigns and measure whether they're actually working.

From Robbie's desk

Get the operator brief.

A short, no-fluff read from our VP of Growth on the bottlenecks worth watching, the AI execution patterns that are actually shipping, and the numbers behind them.

Stop guessing at scale.

Most consultancies want to sell you a 12-month transformation roadmap. We don't. Every engagement starts with a blunt 15-minute operational review directly with our VP of Growth. We look at your bottleneck, work out where it's actually breaking, and tell you if an AI execution path makes sense. No pitches. Just operator to operator.