It's 2am at a remote pit and a haul truck drops out of the rotation with a failed final-drive bearing. The replacement exists. It's sitting in a regional warehouse 600 kilometres away, or on a purchase order with a supplier who shipped it three days ago. The part is not the problem. The problem is that the maintenance planner, the materials coordinator, the inbound logistics team, and the operations centre each hold one piece of the picture — and none of them hold all of it. By the time the part is located, expedited, received, and matched to the open work order, the truck has been down for eleven hours. Most of those hours had nothing to do with the repair.
This is the quiet failure mode of mining supply chains. The machines are instrumented, the systems are modern, and the spare is usually available somewhere in the network. What breaks is the coordination between the moment a part is needed and the moment it's installed.
Why this is getting harder, not easier
Unplanned downtime at a large mine is routinely estimated to cost in the range of $180,000 per hour once lost throughput, idle crews, and schedule recovery are counted. The instinct is to attack that number with more reliability engineering and better forecasting — and both help. But a large share of mean-time-to-repair isn't spent repairing. It's spent finding the part, confirming it's the right part, getting it to the face, and telling everyone who needs to know.
Three forces are compounding the problem. Operations are increasingly remote, so decisions about a stranded asset are now made hundreds of kilometres from the equipment, over phone and email, with a lag baked into every handoff. The system estate keeps growing — EAM, ERP, fleet dispatch, supplier portals, carrier tracking — and each new system adds another seam where information stops. And autonomous haulage and drilling generate far more events than a planner can triage by hand. The result is that coordination cost, not part cost, is becoming the constraint on uptime. Here's how to attack it directly.
1. Define "critical" before the failure, not during it
The middle of a breakdown is the worst time to discover that nobody agreed which spares are mission-critical. Build a criticality tier into your asset and materials data ahead of time: which parts, on which assets, stop production if they're unavailable. Map each tier to a target response — a Tier 1 spare for a primary haul truck or crusher should trigger a different urgency than a consumable. When the failure happens, the system already knows whether this is a wait-and-schedule event or a drop-everything event, and so does everyone downstream.
2. Connect the part to the asset to the work order
The data needed to resolve a critical-spares event almost always exists — it's just scattered. The work order lives in Maximo or SAP. The stock position lives in the inventory system. The inbound shipment status lives in a supplier portal or a carrier feed. The asset health lives in the historian. Each is correct on its own and useless in isolation. The first concrete move is to link them: an open work order should carry the part number, the part number should resolve to a live stock-and-inbound position, and the asset should be visible against both. When those three are joined, "where is my part" stops being a series of phone calls and becomes a single lookup.
3. Surface the exception when it's known, not at the loading dock
Most teams learn that a critical inbound shipment is late at the moment it fails to arrive. By then the options are bad and expensive. The fix is to treat the shipment as a tracked exception from the moment the part is needed: a missed supplier cutoff, a customs hold, a carrier that has gone silent for hours, a flight that slipped. Each of these is knowable well before the dock. The earlier a coordinator sees that a Tier 1 inbound has stalled, the more levers they still have — re-source from another warehouse, expedite a partial, or pull a unit off a lower-priority asset. Each of those levers carries a cost, and deciding when a late shipment is worth expediting is far easier while all of them are still open. Visibility only matters if it arrives while you can still act on it.
4. Give every exception an owner and a clock
The single most common reason a critical spare sits is that no one owns chasing it. It's "with logistics," or "with procurement," or "someone's looking into it." An exception without an owner is an exception that drifts. Assign ownership the instant an event is raised, attach a service-level expectation to it based on the criticality tier from step one, and escalate automatically when the clock runs out. Severity-aware routing means a stalled Tier 1 inbound reaches a supervisor in minutes, while routine events stay off the leadership radar. The goal isn't more alerts — it's that the right person is accountable for each one and knows the deadline.
5. Make maintenance, supply chain, and operations move together
A critical-spares event is never one team's problem. Maintenance owns the repair, supply chain owns the part, operations owns the production impact, and on a remote site the operations centre owns the coordination. When these functions work from separate tools and separate threads, every handoff is a delay and every status update is a meeting. The teams that recover fastest run the event on shared rails: one record of the failure, the part, the inbound status, the assigned tech, and the production decision, visible to everyone at once. Coordination stops being a sequence of relays and becomes a single, observable workflow.
6. Close the loop and measure what coordination actually costs
You can't improve what you don't instrument. Track the metrics that separate repair time from coordination time: total MTTR, but also escalation time, time-to-locate-part, and the hours your team spends manually chasing status across systems. When 30 to 50 percent of MTTR turns out to be coordination rather than wrenching, you've found where the next increment of uptime lives — and it's cheaper to recover than building more inventory or buying more spare capacity.
What to do Monday morning
Start with one asset class and one failure mode. Pick your highest-impact equipment — the haul fleet, a primary crusher, a mill — and trace exactly what happens today when a critical spare for it fails. Time each handoff. You'll almost certainly find that the repair was fast and the coordination was slow, and that the slow part lived in the gaps between systems nobody owns. That gap is the opportunity. You don't need to replace your EAM, your ERP, or your dispatch system to close it. You need the seams between them to carry information and accountability instead of dropping it.
How Orkestra fits
Closing those seams is exactly what an operational orchestration layer does — and it's what Orkestra is built for. Rather than replacing SCADA, Maximo, SAP, or your fleet systems, Orkestra sits above them and coordinates the response: detecting a critical-spares exception the moment it's known, joining the work order to live inbound status, assigning ownership with a severity-aware SLA, and driving the escalation across maintenance, supply chain, and operations without an email chain. That's the difference between seeing a problem and resolving it.
It's the same pattern that helped OIA Global cut exception resolution time by 60% in a single quarter — applied to the systems that run a mine. The constraint on uptime is coordination, and coordination is solvable.
Control your orders — Book a demo.
