Measured rarely.
Surveys are expensive, so they happen quarterly at best.
How much is in that stockpile?
Measure the pile, not the paperwork. Every stockpile. Every survey. Every variance.
See it worked throughBook stock says one thing.
The yard says another.
Stockpiles are measured a few times a year, by survey crews or by eye. In between, the book figure drifts from what is actually on the ground, and nobody knows by how much until the audit.
Estimated by eye.
Between surveys, volumes are guessed from truck counts and experience.
Variances found at audit.
Write-offs appear at year end, long after anything could be done about them.
What the AI reads.
Drone and camera images turned into depth and volume, set against what the books say should be there.
In
- Drone surveys
- Fixed cameras
- Weighbridge records
- Book stock
- Site plan
- Material densities
At work
- Sense Reads drone photos and fixed camera views.
- Understand Builds depth, surface and volume for each pile.
- Check Checks each volume against book stock and weighbridge records.
- Anticipate Flags piles whose volume is drifting from the books.
- Operate Sends the variance report with the images and the calculation.
Out
One stockpile survey. Every volume shows its surface.
Every pile measured.
Every variance explained.
Site and finance teams see what is actually in the yard, how it compares with the books and where the difference probably went.
4 match · 1 small variance · 1 large variance
| Pile | Material | Status | Measured vs book | Survey |
|---|---|---|---|---|
| Main yard | ||||
| SP-1 | Sand | Matches | 8,240 m³ vs 8,310 m³ (−0.8%) | Drone 07:10 |
| SP-2 | Gravel 20 mm | Matches | 5,960 m³ vs 6,020 m³ (−1.0%) | Drone 07:10 |
| SP-3 | Crushed aggregate | Large variance | 12,480 m³ vs 13,900 m³ (−10.2%) | Drone 07:10 |
| Lower yard | ||||
| SP-4 | Gravel 10 mm | Matches | 3,110 m³ vs 3,160 m³ (−1.6%) | Camera 4 |
| SP-5 | Topsoil | Variance | 2,030 m³ vs 2,117 m³ (−4.1%) | Camera 5 |
| SP-6 | Rail ballast | Matches | 4,480 m³ vs 4,450 m³ (+0.7%) | Drone 07:14 |
Measured
12,480 m³ from this morning's drone flight: 1,860 photos at 4 cm ground resolution.
Survey SV-0714 · SP-3 surfaceBook
13,900 m³ in stock: production minus despatches since the April survey.
ERP · item AGG-40Trucks
The weighbridge shows 212 loads out; despatch notes record 196.
Weighbridge log · despatch notesDensity
The books convert tonnes at 1.52 t/m³; the latest lab test gives 1.60.
Lab report LR-221
1Large
1Small
4Match
Adjustments to review
- Large varianceSP-3: update the density factor, reconcile 16 loads, investigate 420 m³ Site manager
- VarianceSP-5: re-survey after rain; the camera view was partly blocked Survey technician
- MatchesSP-1, SP-2, SP-4, SP-6: no adjustment Finance
Sign-off
- Site manager Pending
- Finance controller Pending
The AI measures and explains. Finance decides what to book.
Illustrative example. Site, piles, volumes and records are fictional.
Built for the yard.
Yards are dusty, remote and busy. The AI works from the cameras and flights you already have.
Uses what you already fly.
Works from your existing drone surveys and fixed cameras.
Depth & 3D ↗Shows its workings.
Every volume comes with the surface it was measured from and the calculation.
How Check works ↗Runs on site.
Processing can run on a site server, with results sent when there’s a connection.
Deployment options ↗Same records.
Other problems.
The same surveys answer other questions too.
What runs out before the next delivery?
Stock shortages
Anticipate shortages before they become a problem.
See how →Which animals passed, and are they hurt?
Wildlife monitoring
Identify species, spot injuries and flag poachers from camera traps.
See how →What here isn’t safe?
Safety checks
Flag what’s unsafe or incomplete.
See how →Bring us your last drone survey.
We’ll turn it into volumes and show you how they compare with your books.
Talk to us →LondonDiscovery AI Limited · United Kingdom
HyderabadReDscvry Technology Private Limited · India
ChennaiReDscvry Technology Private Limited · India