For Mining Operations Leaders at Surface Mines
Rithmik discovered a fault costing ~$45,000 in weekly production that no OEM alarm flagged.
$45,000 15% slower 0 alarms
Rithmik is a prescriptive analytics platform for surface mining operations.
01 — Context / Problem 02 — Anomaly Detected
A Hitachi EH3500 haul truck at a copper mine has been underperforming its fleet since a routine preventive maintenance event on December 4. The truck is still hauling, but it is dragging cycle times across the haul loop, and the OEM onboard monitoring systems have triggered no alarms. Average loaded uphill speed had dropped from 13 to 11 km/h, a 15% reduction, ranking the truck last in the fleet on effective resistance and engine power.
No OEM alarms fired at any point. The anomaly surfaced exclusively through Rithmik's relative fleet ranking and signal deviation models, which compare each truck against its peers across the same haul loop.
03 — Signal Analysis
04 — Failure Mode Diagnosis
Failure Mode
88%
Evidence
Category Component
05 — Outcome
$45,000 a Week of Production Protected
Every week, the fault went undiagnosed, the operation was losing roughly $45,000 in production. A single slow truck does not operate in isolation; it backs up the loader, queues every truck behind it on the dump cycle, and drags effective cycle times across every unit on the same haul loop. Rithmik flagged the fault, traced it back to the December 4 PM event, and delivered the diagnosis with recommended corrective action.
Equipped with the diagnosis and the underlying signal evidence, the operator was able to take targeted corrective action and restore the truck to fleet performance. The weekly bleed of production stopped.
Rithmik showed their AI solution could reveal early indicators of genuine equipment issues beyond what was available from the manufacturer's alarms.
— Superintendent of Reliability
Canadian Mining Operation