Mine operators in Ghana: a cost-per-tonne you can trust
A cost-per-tonne you can trust.
You own the mine. You carry the cost-per-tonne, the royalty exposure, and the Minerals Commission relationship.
Darikoda turns your fleet, fuel, and production into a live cost-per-tonne and a compliance-ready operating record. Captured at source.
You own the mine. You carry the cost-per-tonne, the royalty exposure, and the Minerals Commission relationship. Darikoda turns your fleet, fuel, and production into a live cost-per-tonne and a compliance-ready operating record.
The numbers the mine runs to
The numbers that separate a mine making money from a mine paying for idle assets.
Cost-per-tonne and cost-per-ounce as live numbers.
The golden metrics. Per haul road, per truck, per shift, per operator. Live, not lagged. Under the new sliding-scale royalty regime, the curve is steeper at high gold prices, which means every cedi of avoidable operating leakage hurts more.
Mechanical above 85%, physical above 90%, utilisation above 75%.
The targets that separate a mine making money from a mine paying for idle assets. Per machine, not per fleet. The asset pulling the average down is named, not averaged into anonymity.
Stripping ratio visibility.
The relationship between waste moved and ore exposed, tracked over time, with cost attribution. Without it, the pit plan and the cost plan drift apart silently.
Fuel integrity across the bowser-to-engine chain.
You store it. You dispense it. You account for it. One of the largest cost lines in the operation. Variance over your threshold opens an issue automatically with operator, machine, time and quantity attribution.
OTR tyre life as real money.
Six per truck. USD 30 to 60K each. USD 180 to 360K per set. Attribution per operator, haul road, loading style turns tyre life from a fixed cost into a controllable cost.
Minerals Commission compliance under L.I. 2431.
The sixth-edition procurement-list framework is part of how the operating record is built. Ghana-specific compliance is the first context, not a localisation afterthought.
What changes when the record is structured
Three scenarios where the operating record changes the unit economics.
Old way
Two haul trucks run the same route with the same load. Month-end reconciliation lands. One reads GHS 15/tonne. The other reads GHS 8/tonne. The pattern has been live for six weeks. The GM did not see it.
On Darikoda
Live cost-per-tonne per truck, per haul road, per shift, per operator. The diverging truck shows up the day it diverges. The fault, the operator, the loading style, the haul-road grade. All attributed in the same record.
Per-truck cost
live, not lagged. The number you can take to a workshop conversation.
Old way
Diesel dispensed at the mine bowser. Recorded on paper at the dispense. Recorded again at month-end on the workshop spreadsheet. Variance over 5% is normal and absorbed. The dispense clerk turns over twice a year and the records reset.
On Darikoda
Every dispense captured against operator, machine, project, quantity and time at the bowser. Reconciled against shift consumption automatically. Variance above your threshold opens an issue, not a memory.
5% recovery
USD 5K back to margin per month on a USD 100K monthly diesel spend.
Old way
Six tyres per truck, USD 30-60K each. Two trucks burn through sets every nine months. Two trucks last fifteen. The pattern is operator and haul-road dependent but no one has the per-asset, per-operator history to prove it.
On Darikoda
Tyre life per truck, per operator, per haul road, per loading style. The pattern surfaces in the second month. Coaching, route changes, and operator rotation become evidence-led, not speculative.
USD 180-360K
per truck per tyre set. Attribution is where the saving lives.
Different scenarios. Same underlying gap. Same closing move.
Three engines applied to a full mine
The operating record, framed for the mine owner's seat.
ENGINE 01
Cost-per-tonne, cost-per-ounce, live.
Per haul road, per truck, per shift, per operator. Visible the moment work happens, not at month-end after the variance has disappeared into the ledger. The royalty curve becomes a managed exposure rather than an absorbed shock.
ENGINE 02
Availability targets the operation runs to.
85% mechanical, 90% physical, 75% utilisation, tracked per machine. The asset pulling the average down is named. The intervention happens against a specific machine and a specific operator, not against a fleet aggregate.
ENGINE 03
Fuel integrity at the bowser.
Every dispense captured with operator, machine, project, time and quantity attribution. Reconciled against shift consumption. Variance over threshold opens an issue automatically. Five percent recovery on a USD 100K monthly diesel spend is USD 5K back to margin every month. One line item. (Illustrative model, your number from the audit.)
Proof point
If two haul trucks run the same route with the same load and one costs GHS 15/tonne and the other costs GHS 8/tonne, would you know? And what would you do?
Built so the cost-per-tonne survives commodity-price drift.
A full mine in Ghana lives or dies on the integrity of the record between pit, bowser, and royalty return. Six operational commitments make that record survive the trip.
- Every field write saves locally first and syncs when signal returns. Pit edges and stripping benches do not wait for 4G.
- Every transaction has a sync state: saved, queued, synced, failed. No silent gaps in the daily production record.
- Every fuel event, hour-meter, and fault is attributed to operator PIN, machine, project, and time. Royalty reporting traces back to a structured source.
- Shared tablets at the bowser use PIN-level operator attribution. The dispense clerk, the operator, and the supervisor each leave a clean trail.
- Failed syncs at the pit edge become visible issues, not silent gaps in the Minerals Commission record.
- Finance, operations, and the Minerals Commission see the same source record. The royalty return is built from the operating ledger, not reconstructed alongside it.
If you do nothing
The cost of one more quarter on the reconstructed record.
Numbers below are illustrative. The Operational Audit produces your specific number against your fleet, your production volume, and your royalty exposure.
Scenario 01
GHS 7/tonne variance hidden in the fleet average.
On 100,000 tonnes hauled per month, a GHS 7 per-tonne gap between best and worst truck is GHS 700,000 absorbed per month into the cost-per-tonne average.
Annualised, that is GHS 8.4M. One truck identified, one operator coached, one haul road re-planned: the recovery starts there.
Scenario 02
5% fuel variance absorbed without challenge.
USD 200,000 monthly diesel × 5% absorbed = USD 10,000 leaking per month. One line item.
USD 120,000 per year. The structured bowser ledger is the recovery path.
Scenario 03
Tyre life one cycle short on one OTR truck.
One full tyre set USD 180,000-360,000. One truck rotating sets nine months instead of fifteen costs roughly USD 90,000-180,000 per year in accelerated replacement.
Per-operator and per-haul-road attribution recovers the gap within one rotation cycle.
Scenario 04
Royalty exposure on a 3% cost-per-ounce drift.
Under the sliding-scale royalty regime, a 3% cost-per-ounce drift on a 50,000-ounce annual production at USD 2,400/oz = USD 3.6M absorbed before the audit closes the year.
Live cost-per-ounce visibility turns the royalty curve from absorbed shock into managed exposure.
Most mine GMs recognise two or three of these. The audit identifies the biggest line for your operation.
Inside a typical month
What a Darikoda month looks like at a full mine.
From daily production capture in week one, to the royalty exposure becoming a managed line by quarter two.
Week 1
Daily production capture goes live.
Hauls per truck per shift, tonnes loaded, fuel dispensed at the bowser, hour-meter readings, fault reports. All PIN-attributed, all time-stamped, all syncing from the pit edge.
Week 2
First cost-per-tonne dashboard surfaces.
Per truck, per haul road, per operator. The variance between best and worst is named. Not by feel, not by quarterly reconciliation. By live evidence.
Week 3
Workshop conversations shift.
Mechanical availability per machine. The asset pulling the average down is named. The intervention happens against a specific truck and a specific failure mode, not against a fleet aggregate.
Week 4
Monthly Minerals Commission pack built from the record.
Production, fuel, royalty calculation, plant register. Structured per asset, per shift, per operator. The return is an export, not a reconstruction.
Quarter 2
Royalty exposure becomes a managed line.
Cost-per-tonne and cost-per-ounce trend lines surface against the royalty curve. Pricing decisions, hedging conversations, and capital-allocation calls become evidence-led instead of estimate-led.
A note from Theo
“The pit produced more than the workshop record reflected.”
The mines I worked alongside at Caterpillar and Unatrac in Ghana shared the same operational pattern. The pit produced more than the workshop record reflected. The bowser dispensed more than the fuel ledger could explain. The royalty return was reconstructed at quarter-end from three spreadsheets and a long memory. Operationally these mines were sound. Financially they were absorbing 5-10% leakage that nobody could quite name. The operating record is the layer that names it. Per-truck, per-operator, per-shift, captured at the event. The royalty curve becomes a managed exposure rather than an absorbed shock. That is what we built for full mine operators in Ghana.
Theo Ilori
Founder, Darikoda. UCL MSc Mechanical Engineering. Formerly GE precision turbines, Caterpillar/Unatrac Ghana & Nigeria.
Mine operator FAQ.
The questions other Ghana mine owners ask in the first call.
How does Darikoda handle the new sliding-scale royalty regime?
Since March 2026, gold royalties rise as commodity prices rise. The cost-per-tonne and cost-per-ounce visibility Darikoda surfaces is the operational lens that turns the royalty curve from an absorbed shock into a managed exposure. Every cent of cost-per-ounce drift matters more when the curve is steeper at high prices.
Are you Minerals Commission compliant?
Darikoda is a software product. Ilori Streamline Ltd, the company behind Darikoda, is a UK-registered entity (Company No. 16378536) with active Ghana operating presence and works alongside Ghana-registered partner engineering firms where the Minerals and Mining Local Content Regulations require it. The L.I. 2431 sixth-edition procurement-list framework is part of how we built the product, not an afterthought.
What if our fleet is mixed-OEM?
Mixed fleets are the norm in Ghana. Darikoda is OEM-agnostic. The operating record is captured at the operational layer (operator, project, shift, fuel event, downtime cause) so the asset record is consistent whether the machine is a CAT, Volvo, Komatsu or Hitachi. Telemetry integration where it exists is additive, not required.
If two haul trucks run the same route with the same load and one costs GHS 15/tonne and the other costs GHS 8/tonne, would you know?
On the current month-end cycle, probably not. On Darikoda, the variance surfaces within hours of the divergence, on a per-truck per-route basis. What you do with the gap is the operational decision. The data is the foundation.
How does the platform handle remote pit sites on patchy 4G?
Every field write saves locally first and syncs when signal returns. Operators do not wait for connectivity to record a fuel event, an hour-meter reading or a fault report. Failed syncs become visible issues, not silent gaps. Sites with Starlink plus patchy 3G/4G are well within the operating envelope.
Want the leakage map for your mine?
30 minutes on WhatsApp. You keep the map regardless of next steps.
Patterns described here are drawn from extensive field audits and industry research across Ghana's mining, construction, roadworks, and quarry sectors. No specific operator is named or identifiable.