Rio Tinto’s Pilbara Machine: How Automation Built the Lowest-Cost Iron Ore System

Every tonne of Pilbara iron ore is a logistics problem disguised as a mining problem. The ore is abundant and the geology is generous; what separates a profitable producer from a marginal one is how cheaply it can move hundreds of millions of tonnes of rock across a thousand kilometres of desert and onto a ship. Rio Tinto’s answer, built over fifteen years and several billion dollars, was to remove humans from as much of that chain as possible. This article explains how the system works, what it actually delivers financially, and why the company’s hardest problems have turned out to be social rather than technical.

What does Rio Tinto actually own in the Pilbara?

An integrated portfolio of 18 iron ore mines spread across roughly 70,000 square kilometres of Western Australia, connected by two dedicated rail lines to port terminals at Dampier and Cape Lambert. Rio Tinto owns the mines, the rail and the ports outright — it is not a customer of anyone else’s infrastructure, which is the single most important fact about its cost position.

Production in 2025 was 327.3 million tonnes on a 100% basis with shipments of 326.2 million tonnes, broadly flat year on year after cyclones disrupted the first quarter and record mining rates from April onwards recovered the shortfall. Total group iron ore sales, including the Iron Ore Company of Canada and initial Simandou volumes, reached 342 million tonnes.

The Pilbara Blend is a marketing product rather than a single orebody. Ore from many pits with different iron, silica, alumina and phosphorus content is blended to a consistent specification that steel mills can buy without re-engineering their furnace mix. That blending capability is a commercial asset in itself — and following a 2025 product strategy review, Rio Tinto adjusted certain Pilbara Blend specifications as its mine sources age.

How did Rio Tinto become the most automated miner in the world?

It started with haul trucks in the late 2000s and worked outward across the whole value chain. Today autonomous haul trucks move ore from pit to crusher without a driver, autonomous drills prepare blast patterns, and the AutoHaul programme — completed in 2018 and the first fully autonomous long-distance heavy-haul rail network in the world — runs the trains.

The coordinating layer is the Operations Centre near Perth airport, roughly 1,500 km from the mines. Controllers there monitor and direct equipment across every site, port and rail line in real time. Instead of a dozen isolated mine control rooms, the entire system is optimised as one network, which is how a delay at one port berth can be absorbed by re-scheduling trains from three different mines.

Gudai-Darri, opened in 2022, was designed from the ground up around this model — autonomous trucks and trains, a data-analytics backbone, and its own solar farm. Its real significance is as a template: the marginal cost of adding a fully automated mine to an existing automated network is far lower than automating an existing mine.

Does automation actually lower costs?

Yes, but less dramatically and less immediately than the headlines suggested. Pilbara unit cash costs were US$23.5 per wet metric tonne in 2025, slightly higher year on year because of inflation, a higher work index as pits deepen, and cyclone recovery costs — partially offset by productivity gains. Automation does not exempt a miner from inflation; it changes the slope of the cost curve over time.

The clearer benefits are in utilisation and consistency. Autonomous trucks do not take shift breaks, drive to a consistent speed profile, brake identically, and generate maintenance data continuously. Tyre wear, fuel burn and haul-road degradation all become predictable and therefore manageable. Across 300+ million tonnes a year, a few percent of extra availability is worth more than most cost-cutting programmes.

There is also a safety dividend that shows up indirectly in cost. Removing people from haul roads and rail corridors removes the most common category of serious incident in open-pit mining. Rio Tinto’s all-injury frequency rate was 0.37 in 2025, and the Safe Production System now deployed across dozens of sites is explicitly designed to link safety performance to production consistency rather than treat them as competing objectives.

What did Juukan Gorge cost Rio Tinto?

In May 2020 Rio Tinto legally destroyed two rock shelters at Juukan Gorge in the Pilbara that contained evidence of continuous human occupation stretching back roughly 46,000 years, in order to access higher-grade ore. The blast was permitted under Western Australian heritage legislation dating from 1972, and it was catastrophic for the company.

The chief executive and two senior executives left, the chairman subsequently departed, a federal parliamentary inquiry produced damning findings, and Western Australia rewrote its Aboriginal cultural heritage legislation. Rio Tinto rebuilt its agreements with Traditional Owner groups, added heritage veto rights into mine planning, and made cultural heritage a board-level risk.

The financial cost was modest; the strategic cost was enormous. Access to land in Australia now depends on relationships that take years to establish and moments to destroy. For any resources company modelling project risk, Juukan Gorge is the case study that shows a legally compliant decision can still be an existential governance failure.

How is Rio Tinto diversifying away from iron ore?

Three ways. Copper is the largest: the Oyu Tolgoi underground project in Mongolia completed development and is ramping toward becoming one of the world’s largest copper mines before the end of the decade, and drove much of the 8% copper-equivalent production uplift the group reported in 2025.

Lithium is the newest. The acquisition of Arcadium Lithium closed in March 2025, giving Rio Tinto brine and hard-rock lithium assets across Argentina, Australia and North America, and it was followed by agreements in Chile with Codelco and ENAMI. The group has since restructured into three product groups — Iron Ore, Aluminium & Lithium, and Copper — which is the clearest possible signal of where management expects growth to come from. Australia’s hard-rock lithium producers are covered in detail in our guide to the ASX lithium sector.

The third is high-grade iron ore outside Australia. Simandou in Guinea, developed through the SimFer joint venture with Chinese partners and the Guinean government, made its first shipment in December 2025 and is targeting 60 million tonnes a year of high-grade ore. Rio Tinto is, in effect, hedging the eventual decline of its own Pilbara grades by building the resource that competes with them.

What did the 2025 results tell investors?

That the diversification is working financially, but slowly. Group underlying EBITDA rose 9% to US$25.4 billion, underlying earnings were stable at US$10.9 billion, and operating cash flow reached US$16.8 billion. The board maintained a 60% payout ratio, giving an ordinary dividend of about US$6.5 billion.

Inside those numbers, iron ore did the heavy lifting but went backwards: segment underlying EBITDA fell 11% to US$15.2 billion as the average realised Pilbara price dropped to US$90.0 per dry metric tonne. Operating cash flow from iron ore alone was US$10.6 billion, with free cash flow of US$6.1 billion. Copper and aluminium growth offset the price decline — which is precisely the outcome the diversification strategy was designed to produce.

Guidance for 2026 puts total iron ore sales at 343–366 million tonnes including 5–10 million tonnes from Simandou, with medium-term Pilbara system capacity of 345–360 million tonnes a year underpinned by the replacement mine programme. Simon Trott, previously head of the iron ore business, became chief executive in August 2025 and has framed the agenda around a “stronger, sharper, simpler” operating model.

How does Rio Tinto compare with BHP and Fortescue?

On iron ore alone the three are remarkably close in cost, and all three are far below the marginal seaborne producer. The differences are in portfolio shape and grade. BHP has rotated hardest into copper and potash and now earns more from copper than iron ore. Rio Tinto is more evenly spread across iron ore, aluminium, copper and lithium, and carries the industry’s largest aluminium business. Fortescue is the pure-play challenger, with lower-grade ore that sells at a discount and a far more concentrated risk profile.

Grade is the quiet differentiator. Rio Tinto and BHP both sell blended products near the 62% iron benchmark; Fortescue’s hematite sits below it and attracts a discount that widens whenever steel mills prioritise productivity over cost. As Pilbara ore bodies age and average grades drift down, all three face the same slow squeeze — which is exactly why Simandou’s high-grade ore matters strategically.

For finance teams benchmarking capital intensity, the useful comparison is capital spend per tonne of sustained capacity. Replacement mines are not growth capital; they are the cost of standing still in a business where every pit eventually runs out. Treating them as discretionary is the most common analytical error made about Pilbara producers.

Frequently Asked Questions