Soft sensors for conveyor belts at Samarco

Conveyor belts are the arteries of a mining operation.

Yet, unplanned stoppages and failures in belt scales are costly and frequent.

Samarco solved this challenge not with deep learning or complex architectures, but with something deceptively simple: a linear regression model.

This virtual balance sensor estimates conveyor mass flow in real time, delivering the same function as a physical belt scale — but with higher reliability and lower maintenance.

Simplicity That Works
 

While often overlooked, linear regression remains one of the most powerful tools in industrial AI/ML. In Samarco’s case, it proved accurate enough to feed APCs directly, with retraining needed only once every 4 months.

Its success shows an important lesson:

AI in OT doesn’t start with deep learning — it starts with models that work and keep working.

The Auto-ML Template: A Samarco Idea Scaled Through SIENTIA™
 

One of the key breakthroughs came not from the algorithm itself, but from how the model was operationalized.

The idea of creating a no-code template for conveyor models originated within Samarco’s own team. By incorporating this into SIENTIA™, the approach became part of the platform’s DNA:

• Drag-and-drop simplicity, so models could be built without coding.
• Standardized workflows, ensuring governance and observability by default.
• Accessibility for all skill levels: from senior process engineers to plant operators, anyone could configure and replicate the model in minutes instead of months  

 Impact on Operations
 

• Reduced unplanned stoppages from conveyor overloads.
• Reliable mass flow estimation, even when physical scales fail.
• Scalability across 30+ conveyors, multiplying the impact.
• Cultural empowerment, as more people across the organization can now build and deploy ML models safely.