Stockpile Volumes · Pit Progression · Reclamation

From Drone Capture to Stockpile Volume Before the Crew Lands

Mining operations survey weekly. Each survey is 500M+ points. GeoDataConverter classifies, generates DEM/DSM, computes stockpile and cut/fill volumes, and runs M3C2 epoch comparison — automatically, end-to-end.

The Problem

Where the time goes today

Mining survey teams produce data faster than they can process it. Reconciliation slips. SMCRA paperwork piles up. Pit-wall stability waits behind volume reports.

2–3 days per site for traditional ground stockpile survey

Weekly cadence creates a continuous processing backlog

±10–15% accuracy from manual methods vs. ±2% drone-based

SMCRA reclamation reporting assembled by hand each quarter

The Workflows

Mining workflows on the platform

Fourteen workflows mapped to the operational cadence of an active pit — weekly volume, monthly progression, quarterly reclamation, annual integrity.

Workflow Name & what it automates Time saved
#24Stockpile Volume
Stockpile, pit, and mound volumes from LiDAR/DEM with configurable base surface
8 hr / site
#23Cut/Fill Analysis
Cut and fill comparison between two terrain epochs with per-cell delta raster
#26Terrain Change Detection
Elevation change quantification between temporal DEM pairs with significance mask
#276M3C2 LiDAR Epoch Comparison
Statistically rigorous distance computation with LOD95 confidence — engineering-grade
#270Deep-Learning LiDAR Classification
RandLA-Net semantic segmentation: ground, bench, road, stockpile, equipment, vegetation
#04Path Profile / Haul Road Grade
Elevation profiles along haul road centerlines to verify 8–12% grade limits
#33NDVI Reclamation Monitoring
Vegetation health time series for SMCRA revegetation compliance
#287LiDAR QA/QC Suite
USGS LBS-compliant quality assessment per survey — point density, voids, classification
#305Construction Equipment Tracking
MobilityDB trajectory storage with geofence + idle detection for shovels, haulers, drills
#280Temporal Analytics Dashboard API
TimescaleDB REST API for stockpile volume / pit progression trend queries
The Math

ROI vs. manual processing

Modeled on a Tier-2 surface mine with weekly stockpile surveys across 8 active pads and quarterly SMCRA reclamation reporting.

Metric Manual / current tooling GeoDataConverter
Stockpile survey processing time8 hours12 minutes
Annual processing labor (52 cycles × 8 pads)$291,200$36,000
M3C2 LOD95 confidence on volumeNot producedPer-survey
SMCRA reclamation report assembly5–8 days1 day
Differentiator

M3C2 with LOD95 — the geotechnical-grade answer to "is the pit wall moving?"

Workflow #276 implements the Multiscale Model-to-Model Cloud Comparison algorithm with per-point Level of Detection at 95% confidence. This isn't raster differencing — it's the published algorithm geotechnical engineers cite in stability reports. Pair it with #305 equipment tracking and you can correlate displacement signals against blast events, rainfall, and equipment proximity.

Free 4-Week Stockpile Pilot

Send us 4 consecutive weeks of stockpile survey data. We'll process every survey through the automated pipeline and return volumes, cut/fill, and an M3C2 trend report.

  • 4 weeks of weekly stockpile cycles processed
  • Volumes, cut/fill, and M3C2 epoch comparison
  • Side-by-side benchmark vs. current method
  • QA/QC report card per survey