Case studyMetals & miningOptimize throughput

+4.8% tonnes-per-hour on the constraint line.

Grinding circuit advisor lifts throughput by anticipating ore hardness shifts.

Talk to an expertDownload 2-page PDF
ForecastingPrescriptive optimization
THROUGHPUT · LINE_02horizon 4h
+240 min projection · +4.8% t/h achievable
At a glancetimeframes and caveats below
+4.8%
tonnes / hour
6 months
Constraint line, audited
−22%
control swings
Same period
Std-dev of recycle load
Neutral
liner wear
Inspection cycle
No accelerated wear flagged
10 weeks
to first recommendation
Pilot
Historian access in week 2
Challenge

What was happening before.

Throughput was capped not by the mill itself but by the time it took operators to recognise a hardness shift and re-balance feed rate and water addition.

Constraints
  • Liner wear cannot increase materially.
  • Recycle load must stay within mechanical limits.
  • Operators retain manual override at all times.
Solution

What we built, on what data.

Data
  • › Historian tags: feed rate, mill power, bearing pressure, recycle load, water addition
  • › Lab assays: ore hardness, head grade
  • › Geological model: feed source plan
  1. 01
    Build an ore-hardness forecast
    Joined assay history with feed-source plan to project 30-minute hardness shifts.
  2. 02
    Train a throughput advisor
    Prescribes feed rate and water given forecast hardness and current load.
  3. 03
    Show operator the why
    Each recommendation includes the forecast that drove it, not just a setpoint.
  4. 04
    Weekly review with metallurgy
    Metallurgy reviews wear indicators and signs off on continued use.
Adoption

Who uses it, and where it fits in the day.

Used by every mill operator. Metallurgy reviews wear-indicator dashboards weekly. Mine planning shares feed-source changes through the same model.

Modules in use
ForecastingPrescriptive optimization
Results

Quantified, with timeframe and caveats.

  • R1Throughput +4.8% on the constraint line.
  • R2Control swings (std-dev of recycle load) down 22%.
  • R3No accelerated liner wear flagged on the regular inspection cycle.
THROUGHPUT · LINE_02horizon 4h
+240 min projection · +4.8% t/h achievable

Operational outcome (proof rule): Throughput on the constraint grinding line increased by 4.8% across a 6-month audited window with no liner-wear penalty observed.

Voice from the floor
"We used to chase the mill. Now we get a 20-minute heads-up that the next feed pocket is harder, and we're already where we need to be when it hits."
Mill Operator · A crew · Northern Metals
Replication

Relevant for teams trying to…

Mining and minerals teams optimising mills, crushers, or flotation circuits.

Solution: Optimize throughput →Industry: Metals & mining →
Lessons
  • Throughput models that ignore wear get switched off. Pair the KPI with a wear indicator from day one.
  • The geological forecast matters more than the control model. Spend the time on feed data first.
Get started

Talk to an expert about a similar use case.

Tell us your unit and your KPI. We'll come back with what data you'd need and a realistic timeline based on Northern Metals's rollout.

Book a demoDownload 2-page PDF