LiDAR Stockpile Volume Measurement: How It Works and Why Mines Are Switching

A mid-size coal mine in Queensland, Australia used to shut down its primary haul road for four hours every month so a survey crew could walk the stockyard with a total station. Each survey covered 18 piles and produced volume numbers with 3–5% uncertainty. The shutdown cost the mine roughly $28,000 per month in lost hauling capacity.

They replaced the total station with a drone-mounted LiDAR system in 2022. Surveys now take 45 minutes with zero production shutdowns. Volume accuracy tightened to 1–2%. Annual savings from eliminating shutdowns alone exceed $330,000.

That's the financial equation driving the shift to LiDAR in mining inventory management. It's not just about better accuracy. It's about speed, safety, and keeping the operation running while you measure it.

The Mechanics: How LiDAR Measures a Pile

LiDAR measures stockpile volume the same way it measures anything else — by firing laser pulses and recording the round-trip time of each reflection. The sensor builds a dense 3D point cloud of the pile surface, and software calculates the volume between that surface and a reference plane below.

The process has four steps:

Data collection. The LiDAR sensor scans the stockpile from one or more positions. Depending on the setup, this could be a fixed scanner on a tower, a drone flying overhead, or a handheld unit walked around the pile. Each laser pulse returns a distance measurement along with the pulse's angular position, producing a point in 3D space. A sensor scanning at 200,000 points per second (like the Livox M360) generates millions of data points per minute.

Point cloud processing. Raw point cloud data gets cleaned — removing noise, filtering outliers, and classifying points as stockpile surface vs. surrounding terrain. Ground points are separated from pile material.

Volume calculation. Software defines a reference surface (the ground plane beneath the pile) and calculates the volume between that reference and the pile surface. The reference surface choice matters enormously — using a flat plane when the actual ground slopes introduces systematic error. Most mining operations use a custom terrain surface captured before material placement, or interpolate from surrounding ground points.

Tonnage conversion. Volume gets multiplied by the material's bulk density factor to estimate tonnage. A 5% volume error on a 100,000-tonne stockpile at $80/tonne means a $400,000 discrepancy in inventory value. Getting the volume right is the first step; keeping density factors current through periodic material testing is the second.

Three Deployment Approaches

Fixed-mount scanning

A LiDAR sensor installed on a permanent structure — a tower, building roof, or purpose-built pole overlooking the stockpile area — provides continuous or scheduled scanning without human involvement.

Fixed-mount systems work well for a defined number of stockpiles in a known location: ore stockpiles at a crusher, salt barns, aggregate yards, or ROM (run-of-mine) pads. The sensor scans automatically at set intervals (daily, weekly, or on-demand), and volume changes between scans are tracked over time.

The trade-off: fixed sensors have a limited field of view. A single unit can typically cover a 180° arc up to 50m, depending on pile geometry. Large stockyards with many piles may need multiple sensors or accept incomplete coverage on distant piles.

Drone-mounted LiDAR

A LiDAR sensor flown on a drone combines the coverage of aerial survey with the 3D precision of laser scanning. The drone flies a programmed path over the stockyard, collecting a dense point cloud of every pile in a single flight.

Drone LiDAR has become the dominant approach for mining stockpile surveys because it hits the sweet spot of coverage, accuracy, and speed. A typical survey of 10–20 piles takes 30–60 minutes of flight time. Volume accuracy of 1–3% meets or exceeds the requirements for inventory reconciliation and financial reporting.

The sensor weight matters. The Livox M360 at 408g is light enough for most survey-grade drones. Its 200kHz scan rate produces dense point clouds even at moderate flight speeds. The 12–32V input range works with standard drone power systems. IP67 protection handles the dust and occasional moisture exposure that's normal at mine sites.

Where drone LiDAR outperforms alternatives: irregular pile shapes with overhangs or steep faces, stockyards where terrain is uneven, and sites where ground access is restricted due to safety or operations.

Handheld / pole-mounted scanning

For smaller stockpiles, indoor storage, or situations where drones can't fly (under covered areas, near airport approaches), a LiDAR sensor on a pole or handheld rig provides a practical alternative. An operator walks around the pile while the sensor continuously scans, building a point cloud over 5–15 minutes.

Handheld systems are less common in open-pit mining but see use in indoor aggregate storage, salt barns, and construction material yards. The accuracy is comparable to drone and fixed-mount approaches (1–3%) if the operator maintains good scan coverage.

LiDAR vs. the Alternatives

LiDAR vs. Total Station

Total station surveying has been the standard for decades. A surveyor places the instrument at a known position, then walks the pile perimeter taking discrete point measurements. For simple piles on flat pads, this works. For complex stockyards, it falls short.

LiDAR vs. Photogrammetry

Drone photogrammetry uses overlapping photos processed into 3D surface models. It's cheaper than LiDAR (a good camera costs $2,000–$5,000 vs. $5,000–$25,000 for LiDAR) and produces visually intuitive orthophotos.

The key difference: photogrammetry creates surfaces from visual feature matching. When the ground beneath a pile is not visible in any photo — because the pile covers it completely — the software interpolates, and that interpolation introduces error. LiDAR pulses penetrate gaps in loose material and reach the actual ground surface, giving a true reference.

Under most conditions, photogrammetry achieves 2–5% volume accuracy. LiDAR typically hits 1–3%. The gap narrows for well-shaped piles on flat pads. It widens for irregular piles, sloped terrain, or dusty conditions where photo quality degrades.

Dust is worth emphasizing. Mining environments generate airborne dust that coats camera lenses and reduces photo contrast. LiDAR sensors rated IP67 are sealed against particulate ingress, and the laser pulses penetrate light dust more effectively than visual cameras can see through it.

The Economics: When the Numbers Add Up

The financial case for LiDAR in mining rests on three pillars:

Eliminated shutdowns. Total station surveys often require partial or full production shutdowns to keep survey crews safe near active haul roads and loading areas. Drone LiDAR doesn't. At operations where a single hour of shutdown costs $5,000–$10,000, the savings from eliminating monthly shutdowns alone can justify the LiDAR system within a year.

Third-party survey cost reduction. Many mines contract survey firms for monthly or quarterly stockpile surveys. A third-party total station survey of 15–20 piles might cost $5,000–$15,000 per visit. An in-house drone LiDAR system amortizes over 12–24 months of avoided contractor costs.

Inventory accuracy improvement. A 2% improvement in volume accuracy on a stockyard holding $50 million in material at any given time represents a $1 million reduction in inventory uncertainty. For operations where inventory numbers feed directly into financial reporting, loan covenants, or royalty calculations, this accuracy improvement has real monetary consequences.

Accuracy Pitfalls to Watch

LiDAR doesn't guarantee accuracy by itself. The system is only as good as the process around it.

Reference surface consistency. Changing the reference surface between surveys — switching from flat-plane to terrain-surface, or adjusting the ground elevation — introduces errors that look like real volume changes. Document the reference surface method and stick with it across surveys.

Ground control. Without accurate ground control points (GCPs) or RTK positioning, the point cloud may shift or scale incorrectly. Even small positional errors compound over large piles. Use surveyed GCPs or RTK/PPK-enabled drones.

Density factor currency. Volume × density = tonnage. If the density factor is wrong, the tonnage is wrong regardless of volume accuracy. Test material density quarterly or when material sources change.

Point cloud gaps. Steep pile faces, overhangs, and areas directly below the drone flight path may have sparse point coverage. Check the point cloud for gaps before calculating volume. Supplement with ground-level scans where needed.

What to Look for in a Mining LiDAR Sensor

• Scan rate and point density: 200kHz+ produces dense enough coverage for reliable volume calculation at typical drone speeds.
• IP rating: Mining means dust, vibration, and exposure. IP67 is the practical minimum.
• Dual-echo capability: In dusty environments, dual-echo LiDAR (like the M360-D) receives returns from both dust particles and the actual pile surface, maintaining accuracy where single-echo systems degrade.
• Weight: Under 500g fits most survey-grade drones without payload trade-offs.
• Power range: 12–32V covers both drone battery systems and fixed-mount power supplies.
• Temperature range: -10°C to +60°C covers most mining environments. Extreme operations (arctic mines, desert quarries) need to verify against site conditions.

Volume accuracy figures and cost estimates in this article reflect published industry sources and case studies as of July 2026. Actual results depend on site conditions, survey methodology, and material characteristics. Consult with a survey technology provider for site-specific assessments.

© 2026 SmartBotParts. All rights reserved.