Key Takeaways

  • Crusher circuit optimization is one of the more measurable AI wins on this list: reported gains of 1-5% throughput and 10-15% energy reduction from AI setpoint optimization.
  • Imubit’s Closed Loop AI Optimization uses reinforcement learning to write optimal setpoints directly to control systems in real time, rather than relying on a static linearized process model.
  • ABB Ability for Mining is the more established alternative, though it depends more on linearized process models that can struggle with the genuinely non-linear behavior of crushing circuits.
  • Marked “Yes” — this is a production capability with measured results at deployed sites, not a lab demo.

TL;DR

Optimize your crusher circuit with closed-loop AI control — Imubit uses reinforcement learning to continuously adjust setpoints against live plant data, reporting real throughput and energy gains over traditional control approaches like ABB Ability.

How Do I Optimize Crushing With AI?

Crushing circuits are genuinely hard to control well manually: feed rate, crusher gap setting, and ore hardness all interact non-linearly, and operators historically tune setpoints based on experience and periodic manual adjustment rather than continuous optimization. That gap between “what the circuit could achieve” and “what manual tuning actually achieves” is where AI setpoint optimization earns its keep.

Imubit’s Closed Loop AI Optimization approach is built around reinforcement learning: the system learns from actual plant data and writes optimal setpoints directly back to the control system in real time, continuously adapting as ore conditions change rather than running a fixed rule set. This matters because crushing circuits are non-linear — feed hardness, particle size distribution, and equipment wear all shift the optimal setpoint over time, and a reinforcement-learning agent that’s continuously interacting with the live process can track that drift in a way a static model can’t. Reported results from concentrators deploying this kind of approach include 1-5% throughput increases and 10-15% energy cuts, which are meaningful numbers at plant scale.

ABB Ability for Mining is the more established incumbent in this space, offering broader digital mining services beyond just crusher optimization, but its process-optimization approach tends to lean on linearized models of the circuit — a reasonable simplification for many control problems, but one that can leave performance on the table specifically in crushing/grinding circuits where the underlying physics is genuinely non-linear. If you’re evaluating between the two, the practical question is less “which brand” and more “does this vendor’s optimization approach actually model the non-linearity in my specific circuit, or does it assume linearity for tractability.”

Try It With Geocluster

Benchmarking reinforcement-learning-based process control against traditional linearized optimization for your specific crushing circuit is exactly the kind of applied AI evaluation worth doing with real data before committing. The Geocluster Research Harness helps teams run that research systematically.