Australian miner BHP Group said an artificial intelligence solution has helped deliver almost 1 million tonnes of additional annual iron ore production at its Western Australian operations, according to comments made by the company's chief digital officer Mikko Tepponen at the Global Resources Innovation Expo in Perth.
Tepponen said BHP lost about 1,000 hours of crusher downtime over three year due to oversized rocks and foreign objects entering its processing system, as well as additional disruptions at Western Australian ports.
Tepponen said BHP responded with a computer vision solution that uses cameras and machine learning models to detect anomalies in real time. The system, integrated into the company's process control system, helps teams remove oversized rocks or foreign objects before they create a safety hazard, damage equipment or cause unplanned shutdowns.
A BHP spokesman said the solution provides annual lifting capacity of just under 1 million tonnes, equivalent to nearly US$50 million, while reducing crusher downtime by 20 percent, and associated downtime - up to 60 percent. Tepponen said BHP has not had any problems caused by oversized or foreign objects since implementing the technology in 2025.
Tepponen added that the effectiveness of the solution depends not only on the technology itself, but also on how it was deployed. He said it was developed collaboratively with front-line staff, integrated into existing systems and designed to scale from the start.
He noted that AI is seeing the same shift from testing to system-level capabilities as previously seen in autonomous transportation in the mining industry. However, he said there remains a gap between what is built and what actually scales, adding that this gap will only disappear when AI solutions are designed for scale and linked to practical operational challenges.
