An autonomous haul truck operates on a terraced open-pit mine haul road.
Autonomous mining technology is moving from pilot projects to fleet-scale deployment, but the operational case is being built less on driverless vehicles than on the minutes they remove from each production cycle.
Komatsu has commissioned its 1,000th ultra-class autonomous haul truck, while EACON says it has delivered more than 3,000 autonomous mining trucks across more than 40 active projects. CiDi has reported more than 1,900 shipments across nearly 40 mines, and SANY says its deployed autonomous fleet has logged more than 13 million kilometers.
The figures are not directly comparable. They include commissioned, delivered and contracted vehicles, and most are company-reported. Together, however, they show that autonomous haulage has reached a scale at which operators can measure its effect on queuing, shift changes, route consistency and equipment availability.
For mine managers and investors, the key question is no longer whether an autonomous truck can complete a haul cycle. It is whether a coordinated fleet can reduce idle time across the loading, hauling, dumping and maintenance system.
More Read
Why idle time is the central operating metric
A haul truck can be available for work without moving material. It may be parked during a shift change, waiting for a shovel, queuing at a crusher, held behind a disabled vehicle or stopped because the dispatch system lacks a suitable assignment.
These periods are often recorded differently across mines. A credible autonomy business case therefore needs a common baseline before equipment is converted.
The most useful measures include:
- Parked-but-available minutes: time when a truck is ready but not assigned to productive movement.
- Loading queue time: minutes spent waiting at a shovel or excavator.
- Dump and crusher queue time: time lost at the downstream production node.
- Shift-change idle time: nonproductive minutes during crew handovers.
- Intervention time: minutes requiring remote or on-site human assistance.
- Effective operating hours: hours in which the truck is available and completing productive cycles.
- Cycle-time variability: the difference between median and high-percentile cycle times.
Autonomous haulage can improve these measures through consistent spacing, route discipline and centralized dispatch. It does not automatically do so. Narrow roads, mixed traffic, weak communications and unstable loading points can transfer delays from the truck cabin to the control room.
Fleet deployment is reaching measurable scale
Recent milestones show three competing models emerging: OEM-specific systems, retrofit platforms and integrated electric-autonomous fleets.
| Company or operation | Reported milestone | Operating relevance | Evidence status |
|---|---|---|---|
| Komatsu FrontRunner | 1,000 ultra-class autonomous trucks commissioned | Demonstrates commercial scale for an OEM-specific system | Company-reported and covered by Skillings |
| Komatsu FrontRunner | More than 11.5 billion metric tonnes moved since commercial introduction | Provides a long operating history for autonomous haulage | Komatsu-reported |
| EACON ORCASTRA | More than 3,000 trucks delivered across more than 40 projects | Shows the reach of multi-OEM, retrofit and mixed-powertrain autonomy | EACON-reported and covered by Skillings |
| CiDi MetaMine | More than 1,900 trucks shipped across nearly 40 mines | Indicates high-density deployment, particularly in China | CiDi-reported and reported by MINE |
| CiDi largest reported site | More than 220 autonomous haul trucks | Demonstrates the importance of fleet-level orchestration | CiDi-reported |
| EACON, Thiess and Norton Gold Fields | Six retrofitted Komatsu HD1500 trucks | Tests autonomous day- and night-shift operation at Havana Pit | Project-reported |
| SANY Intelligent Mining | More than 300 trucks, 13 million kilometers and 41 million cubic meters moved | Provides a cross-condition operating dataset for autonomous equipment | SANY |
The figures describe different measures and should not be added together as a global fleet total.
Komatsu’s 1,000-truck milestone is important because the system has moved beyond a limited group of iron ore sites. The milestone vehicle, a 930E-5AT electric-drive truck with a 290-metric-tonne payload, was deployed at Nevada Gold Mines, extending the technology’s profile into large-scale gold mining.
CiDi’s reported numbers point to another path. The company’s MetaMine platform combines autonomous haulage with fleet dispatch and other mine-wide functions. Its experience includes a mixed operation in which 56 autonomous trucks worked alongside roughly 500 manned vehicles at a Chinese open-pit coal mine, according to industry reporting.
EACON’s work at Western Australia’s Havana Pit illustrates the retrofit model. Six Komatsu HD1500 trucks were converted for autonomous operation, with the project moving from day-shift operations toward night-shift haulage. The operational test is significant because it examines whether autonomy can be added to an existing fleet without requiring wholesale replacement.
SANY is combining autonomy with electrification. In August, the company announced the shipment of its first batch of pure-electric autonomous mining trucks to South America. The project includes trucks, roadside infrastructure, intelligent dispatching and lifecycle operations and maintenance services.
Control-room personnel monitor live equipment and production data.
How coordinated fleets reduce nonproductive time
The clearest productivity gain from autonomy is usually consistency rather than higher peak speed.
A human-operated fleet can produce highly variable cycle times because of differences in acceleration, braking, route selection and reaction to traffic. Autonomous systems apply the same operating rules across the fleet and can make dispatch decisions using live equipment status.
That can reduce idle time in four areas.
1. Shift-change continuity
Autonomous trucks can continue operating during crew handovers, subject to site procedures and control-room coverage. This removes one source of parked-but-available time, although the mine may still schedule pauses for inspection, maintenance or blasting.
2. Loading and dumping queues
A central dispatch system can redirect trucks before they reach a congested shovel, crusher or dump point. Consistent arrival patterns also make it easier to balance truck supply with loading capacity.
This is where fleet-level intelligence matters. As CiDi’s chief executive Albert Hu told MINE, autonomous mining requires decisions that optimize the entire fleet rather than a single vehicle. A truck that takes the shortest route may still reduce mine-wide productivity if it creates a queue at the next production node.
3. Route and speed control
Autonomous trucks can follow predefined speed profiles and maintain spacing on haul roads. That can reduce stop-start movement, particularly on repetitive routes with stable road conditions.
The benefit depends on mine design. A poorly maintained haul road or narrow intersection can still create a bottleneck, even when every vehicle is autonomous.
4. Planned maintenance
Vehicle-health monitoring can help maintenance teams identify developing faults before they become unplanned breakdowns. The result is not necessarily fewer maintenance hours; it is a shift toward scheduled intervention and more predictable availability.
The Global Mining Guidelines Group’s autonomous systems guideline recommends tracking performance, cybersecurity, lifecycle costs, commissioning and operational readiness. Those considerations are directly relevant to idle-time analysis because a truck that moves efficiently but requires frequent intervention may not improve total mine productivity.
Evidence points to utilization gains, but not a universal number
Historical operating data provide useful reference points, although they should not be treated as universal benchmarks.
Rio Tinto has previously reported that autonomous trucks in its Pilbara operations ran approximately 700 more hours per year than conventional haul trucks and delivered costs around 15% lower in a historical comparison, according to Skillings’ review of autonomous mining deployments.
The GMG guideline also cites reported operational improvements from autonomous systems, including 10% to 20% productivity gains on underground automated production drills, 7% less overbreak from automated or high-precision development drilling, and improved haul-truck utilization.
These figures describe different equipment, mines and operating conditions. They should be used as reference ranges rather than promises. A surface mine converting a large, repetitive truck fleet may achieve a different result from an underground operation with constrained headings, changing traffic patterns and frequent manual intervention.
Base, bull and bear framework
The following framework is an editorial scenario tool for evaluating a fleet deployment over the first 24 to 36 months after commissioning. It is not a forecast or a company guidance estimate.
| Scenario | Idle-time reduction versus baseline | Effective operating-hour gain | Intervention profile | Conditions required |
|---|---|---|---|---|
| Bull | 20%–30% | 15%–25% | Less than 2% of cycles require intervention | Large fleet, separated traffic, redundant communications, stable loading and dumping points |
| Base | 10%–20% | 8%–15% | 2%–5% of cycles require intervention | Partial conversion, reliable dispatch and phased workforce transition |
| Bear | 0%–10% | 0%–5% | More than 5% of cycles require intervention | Mixed traffic, weak network coverage, frequent road changes or charging constraints |
Operators can test the framework by comparing the same metrics before and after deployment:
- Establish a 90-day baseline for truck status, queueing and intervention data.
- Separate planned stoppages from avoidable idle time.
- Measure performance by route, loading point, dump point and shift.
- Track both median and high-percentile cycle times.
- Record network outages, manual takeovers and maintenance-related stoppages.
- Recalculate cost per tonne after infrastructure, training and software-support costs.
This approach prevents a fleet from appearing productive simply because it completes more cycles while waiting longer at the crusher or consuming more control-room resources.
Infrastructure can determine the result
Autonomy changes the mine’s cost structure. Hardware remains important, but communications networks, positioning systems, control rooms, traffic rules and digital mine models increasingly determine whether trucks can operate continuously.
Electrification adds further complexity. Battery-electric trucks compete for charging capacity, which means dispatch systems must coordinate production schedules with energy availability. A fleet may reduce driver-related idle time but create new queues at charging stations if the infrastructure is undersized.
The issue is particularly relevant to copper and critical-minerals operations, where mines are under pressure to increase output while reducing diesel use and emissions. First Quantum Minerals, for example, says its trolley-assist infrastructure supports 128 trucks across its Kansanshi and Sentinel copper mines in Zambia and saved an estimated 29 million liters of diesel and 45,000 tonnes of carbon dioxide equivalent in 2025, according to MINE’s reporting. That is an electrification example rather than an autonomous-haulage result, but it shows how energy infrastructure can become part of the productivity calculation.
Fleet scale and road design shape the economics of autonomous haulage.
Workforce transition is part of fleet availability
Autonomous mining reduces exposure to driving tasks but does not make the operating model labor-free.
Mines require control-room supervisors, network technicians, sensor and vehicle maintenance specialists, safety-assurance teams, data analysts and software-support personnel. The transition can protect employment where workers are retrained and redeployed, but it may also reduce the number of conventional cab-based roles.
Workforce readiness affects idle time directly. If a mine lacks trained personnel to respond to exceptions, calibrate sensors or maintain communications infrastructure, intervention minutes can rise quickly.
What operators should monitor next
The next milestones will be less about the number of trucks delivered and more about repeatable performance in difficult operating environments.
Operators should ask whether a system can:
- maintain productivity during night operations;
- integrate manual and autonomous equipment safely;
- operate across multiple truck brands;
- limit intervention rates as the fleet expands;
- coordinate charging with production;
- preserve availability during network interruptions;
- demonstrate lower cost per tonne after full lifecycle costs;
- support workforce transition without weakening operational readiness.
The deployment data now available show that autonomous mining technology has reached commercial scale. Komatsu’s 1,000-truck milestone, CiDi’s reported 1,900-plus shipments, EACON’s multi-OEM footprint and SANY’s operating dataset all point to a market moving beyond isolated demonstrations.
The more difficult test is whether mines can convert scale into dependable utilization. The operators that succeed will treat autonomy as a fleet and mine-planning decision, not simply as a truck upgrade.
LinkedIn snippet
Autonomous mining is moving into fleet-scale deployment, but the strongest productivity case is built on reduced queueing, shift-change idle time and more consistent cycle times; not driverless trucks alone. Our analysis tracks reported milestones from Komatsu, CiDi, EACON and SANY and provides a base/bull/bear framework for measuring operational gains.
X snippet
Autonomous mining’s next test is fleet performance. Komatsu, CiDi, EACON and SANY are scaling deployments, but the key metrics are idle minutes, queueing, intervention rates and effective operating hours. Read the evidence-led framework.
Mine-wide autonomy depends on coordinated haulage, processing and digital infrastructure.
