Stop Ore Loss: Optimize Stockpiles and Prevent Misclassification

At NTWIST We've worked with mining operations all over the globe, and a pattern is evident frequently: valuable ore is misclassified, misrouted and then mismanaged before it reaches the mill. For open-pit operations with high throughput the misclassification usually results in one spot, and that is the incorrect stockpile.

It's not just a matter of inefficiency. It's valued destruction. Recoverable ore becomes trash; metallurgical models become confused downstream systems are affected by poor quality feed. We'll explain the causes of this problem and how much it costs you, and how you can do it.

The True Cost of Ore Misclassification

Ore misclassification can occur when incorrect blast movement models, insufficient dig boundaries or outdated geophysical assumptions lead to valuable material being categorized as low grade or even waste. This problem is most common in open-pit mines, where blast movement isn't rectified by offset vectors, or models for digital reconciliation.

According to a research study published on ResearchGate Failure to account for blast-induced movements can result in as much as 20% loss of ore as well as 20 percent incorrect classification in copper mining operations which can seriously affect mill feed accuracy as well as grade forecasting (ResearchGate 2018).

When the material is placed onto the wrong stockpile, the material is lost. You may never reclaim it. In fact, the entire chain of production that includes crushing and blending to reconciliation and metallurgy is now based on bad data.

Stockpile Optimization Requires Real-Time Intelligence

A lot of mines depend on shovel guidance systems, as well as grade control drilling for the assignment of material. In reality, they often function in isolation, and fail to form an intelligent, continuous tracker loop for ore. As a result, dig blocks are treated in a static manner and do not account for blast movements, operator-override or routing issues caused by equipment.

As Propeller Aero states the stockpile's reliability and accuracy improves drastically when you switch into automated monitoring systems instead of manual including drone-based volume scanning and 3D dig block modeling and real-time dispatch information (Propeller 2023).

But digitizing your stockpiles won't be enough. You must incorporate that data downstream. AI-driven platforms such as NTWIST's GeoMet-AI combine geological models with blast offset correction as well as feedback loops for dispatch, allowing mines to adjust material classification and stockpile allocation in real-time.

Why Most Mines Misclassify

It's important to be clear that there is no way to dump precious ore. However, most misclassifications occur because the systems aren't interconnected or because models of geology aren't in sync with operational decision-making. Common reasons include:

• Blast movement not considered: Misplaced dig lines because of uncorrected displacement
• Classification that is simplified: Dig blocks defined with bins of wide grade, without resolution
• System that is disconnected: Dispatching, Grade Control, and Reconciliation do not provide live feedback

A truckload classified for being "marginal" could contain pockets of premium ore. And once it's disposed of, it's virtually inaccessible.

The Solution: AI-Powered Routing and Feedback

To prevent misallocation of stockpiles, you require an approach that goes around the world beyond simple classifications. NTWIST's method uses predictive models that are trained on the performance of historical orebodies' blast displacement and operational data that is live. Based on these inputs, AI can recommend not just the best place to dig but the best location to send each truck - all in real-time.

The routing information is returned to downstream models, continually increasing reconciliation accuracy and making it possible to blend proactively. The result? higher mill head grades and fewer losses in recovery and better alignment between operations and geology.

3 Things You Can Do Today

1. Begin to measure the gaps Compare expected mill feed with actual quality and find out who is contributing to the stockpile.
2. Check out the assumptions of your model for blast: If your model isn't making allowances for the movement of your blast, you're already behind.
3. Link your devices: If your dispatch and reconciliation, geology, or reconciliation teams have different platforms, bridge the gap.

Conclusion: Misclassified Ore is Not Just a Mistake - It’s a Metric

At NTWIST we consider ore misclassification as well as misallocation of stockpiles as key performance indicators that are worth keeping track of. If your systems aren't capable of detecting these losses in real-time, it's because you're placing value in the wrong spot and it could not be able to come back.

The best part? With the right information, you can make better decisions about routing to maximize value and bring back confidence in the production data you have. We're here to help lead to the change.

Frequently Asked Questions

What is the definition of ore misclassification mining?

Ore misclassification happens when the materials are incorrectly classified or assigned to the incorrect grade, category, or stockpile.

What is the reason for confusion in classification?

Common causes are blast movement as well as inaccurate dig boundaries. outdated geological models and inoperable systems.

What can miners do to reduce loss of mine?

Mines can cut down on loss of ore by improving the modeling of blast movement by enhancing grade control, linking operational systems, and utilizing real-time tracking of materials.

What is stockpile optimization?

Optimizing stockpiles is a method of controlling materials' placement, tracking, mixing, and recovery to maximize the value of ore and enhance the quality of mill feed.

How can AI assist in managing stockpiles?

AI can analyze operational, geological, and historical data to improve the classification of materials, make recommendations for routing decisions, and help manage stockpiles in real-time.

About the Author

Pramodh Chandrashekar is an industry and technology expert and Senior Researcher at NTWIST. His expertise includes industrial AI, mining and manufacturing technology, digital transformation, and operational optimization.

He focuses on applying AI and data-driven technologies to improve operational efficiency and business performance.

LinkedIn: Pramodh Chandrashekar

Explore Mine-to-Mill Optimization