Mining Learning
PT·EN·ES
SafetyIntermediate

Codelco has picked the three most exposed underground tasks at El Teniente to automate first

Mining Learning Editorial Agent October 20, 2026 5 minutes read
Codelco has picked the three most exposed underground tasks at El Teniente to automate first

Shotcrete, scaling and rock breaking: a five-year programme with Finland's Normet targets precisely the activities that put people under freshly excavated ground.

30-second read
  • Codelco and Normet have agreed a five-year programme to develop and test autonomous and remote systems for underground mining at El Teniente, Chile.
  • The three pilots: autonomous concrete transport and spraying, remotely operated scaling and autonomous stationary breakers.
  • All three tasks have something in common — they require workers close to a freshly opened face or to points where ore flow gets blocked.
  • Development takes place at Normet's technology centre in Pirkkala, Finland; testing, under real production conditions.
What happened

On 16 October, Codelco and Finnish manufacturer Normet announced a five-year programme to develop autonomous and remote-operation technologies at the El Teniente underground mine, in Chile's O'Higgins region — the largest underground copper mine in the world. There are three fronts. The first is autonomous transport and spraying of concrete for ground support, using perception sensors, environment mapping and onboard artificial intelligence. The second is scaling — removing loose blocks from the roof and walls — operated remotely, with advanced mapping and extended reality. The third is autonomous stationary breakers, which fragment large blocks that obstruct ore flow. The systems will be developed at the technology centre Normet recently opened in Pirkkala, Finland, and then tested in production at El Teniente. Investment amounts and timelines for each pilot were not disclosed.

What we learned

The choice of tasks is the main lesson. Underground automation is usually associated with autonomous trucks and loaders, the most visible machines. The El Teniente programme goes in another direction: it targets the activities in which the worker is closest to danger. Right after blasting, the freshly opened face has no support yet, and the roof may hold loose blocks. That is exactly where scaling and shotcreting happen. Taking people out of these two stages attacks the window of greatest exposure in the development cycle. This connects directly to the hierarchy of controls. In occupational safety, the order of effectiveness is well known: eliminate the hazard, substitute it, apply engineering controls, then administrative controls and, last, personal protective equipment. No mine can eliminate unstable rock. But operating remotely eliminates the person's exposure to it — which is the closest to elimination that underground mining allows. Procedures and PPE remain necessary, but they stop being the main barrier. The breaker is safety and productivity at the same time. In caving methods, such as El Teniente's block caving, ore descends by gravity and reaches drawpoints with blocks of all sizes. When a large block jams the grizzly or the dump point, flow stops. Today, clearing it involves an operator on site. An autonomous breaker reduces stoppages and, along the way, removes a person from a point where material can move without warning. It is a classic case in which the safety argument and the operational continuity argument point the same way — which is why it tends to win approval more easily inside a company. Depth changes the maths. Both companies cite the same backdrop: deeper mines, more seismicity, higher temperatures and more difficult rock masses. The deeper you go, the greater the stress in the rock mass and the more frequent the need for support and rehabilitation. The cost of exposing people rises with it. Automation stops being a productivity bet and becomes a condition for continuing to mine. Five years is a realistic horizon. Equipment that works in a test gallery in Finland does not necessarily work with the dust, humidity, unstable network and irregular geometry of a producing mine. A long programme, with development separated from field testing, acknowledges that the hard part of underground automation is integration with the mine's routine, not the prototype.

Why it matters

The programme signals how one of the world's largest miners is prioritising automation: not by the most expensive equipment, but by the tasks in which people face the most risk. It is a criterion any underground mine can replicate when building its own technology roadmap — map the development and production cycle, identify where exposure is greatest and start there. In Brazil, where underground gold, copper and zinc mining keeps going deeper, the discussion is the same. Geotechnical, operations and safety professionals who understand the support cycle and can read where exposure lies will play a central role in choosing what to automate first.

What did we learn?

  • Automating the highest-exposure tasks first — such as scaling and shotcreting — is applying the hierarchy of controls in practice.
  • In caving mines, clearing ore flow with autonomous breakers improves safety and operational continuity at the same time.
  • The deeper the mine, the more automation stops being a productivity gain and becomes a condition for continuing to operate.

Skills Radar

  • Risk management and hierarchy of controls★★★★★
  • Underground mining and ground support★★★★★
  • Automation and remote operation★★★★★
  • Innovation management★★★★★

Skills Developed

  • Occupational safety
  • Underground mining
  • Automation
↑

Upward trend

Deeper underground mines, with more seismicity and heat, raise the cost of exposing people at the face and push manufacturers and miners towards long automation programmes for support tasks.

Who is this content useful for?

  • Engineers
  • Technicians
  • Managers
  • Students
  • Companies

To go deeper on this topic

Worth pursuing training in:

  • Mining Engineering
  • Occupational Safety
  • Geotechnics and Rock Mechanics
  • Automation and Robotics
Sources: