Azure AI, machine learning and cloud tech unlock value at largest copper mine in the world.

In a move toward the digital mines of the future, mining giant BHP has forged a partnership with computing giant Microsoft to improve copper recovery at the Escondida mine in Chile.

"We expect the next big wave in mining to come from the advanced use of digital technologies," said BHP Chief Technical Officer Laura Tyler.

Churning out more than 1 million metric tons of copper per year, Escondida is the world's largest producer of this critical energy metal.

The amount of copper produced at Escondida, however, pales in comparison to the quantities forecast to be needed for the electric vehicles, wind turbines, and other technologies that will power the low-carbon energy future.

When calculating the quantity of metals required for the envisioned future that trades in pipelines for powerlines to deliver global energy needs, the World Bank estimated in 2019 that roughly 550 million tons (1.1 trillion pounds) of copper would be needed over 25 years, which is roughly equivalent to all copper that has been mined over the previous 5,000.

In order to squeeze as much of this energy metal out of the Escondida mine as possible, BHP is collaborating with Microsoft to leverage the global mining company's newly deployed Azure Machine Learning operations platform, along with other Microsoft computing and cloud services such as Azure Synapse Analytics and Azure Data Lake Storage.

"We are excited to partner with BHP on this transformative project that demonstrates the power of AI, machine learning and cloud technologies," said John Montgomery, corporate vice president, AI Platform at Microsoft.

Machine learning assistance

Alongside the modern development practices and new working methods adopted at Escondida, the Microsoft Azure AI, machine learning, and cloud technologies, have enabled BHP to optimize a concentrator circuit at the operation responsible for extracting the copper mineral from mined ore.

BHP is using real-time plant data from the copper concentrators and Azure Machine Learning to make hourly predictions and create machine learning-assisted recommendations for its Escondida operations team.

The mine is now expanding the technology to a second concentrator at Escondida.

Microsoft

BHP is using real-time data from copper concentrators and Azure Machine Learning to make machine learning-assisted recommendations for its Escondida operations team.

Tyler foresees a wider need for the powerful digital technologies that are helping the Escondida team to generate more value from an existing resource.

"As grade declines at existing mines and fewer new copper deposit discoveries are made, next-generation technologies like artificial intelligence, machine learning and data analytics will be used to unlock more production and value from our existing mines," she said.

The project marks the first time BHP has used Azure Machine Learning for its Americas operations and has led to a broader collaboration with Microsoft to make continual improvements to BHP's copper concentrators with AI and new ways of working.

"This partnership is an example of Microsoft's commitment to helping our customers maximize the value of their existing investments so they can unlock operational efficiencies and make step changes to their business models," said Montgomery.