BHP's AI cuts downtime, boosts iron ore output

BHP's AI strategy is built around a clear business case: not general efficiency, but a hard physical limit. The company projects the world needs to double copper production over the next 30 years to keep pace with decarbonization, even as ore grades fall and discoveries become rarer. This is why BHP's AI investments focus on optimizing existing operations rather than simply expanding.

The Escondida partnership with Microsoft, started in May 2023, uses Azure machine learning to generate hourly predictions about plant performance. While BHP has not disclosed a comprehensive figure for copper recovery improvements, it has quantified savings in water and energy: more than three gigalitres of water and 118 gigawatt hours of energy since fiscal year 2022. The company is expanding this to a second concentrator, signaling it has moved past a pilot.

For the conveyor monitoring system, the results are more concrete: roughly 1 million tonnes of additional annual iron ore production, worth an estimated $50 million, after cutting crusher downtime by 20% and related disruptions by up to 60%. The integration directly into BHP's process control systems was key to scaling it beyond a pilot. BHP is also centralizing AI capability through its Industry AI Hub in Singapore, aiming to build once and redeploy across its global portfolio.