We leverage our experience in metallurgical engineering (test-work management, flow-sheet design, process development and optimization, and plant management) with the use of state-of-the-art data science tools (data science libraries) and machine learning algorithms (regression and classification tools, neural networks, support vector machines, anomaly detection) in order to the help metals & mining industry to tackle its modern challenges.
We use high-level prototyping languages such as Python, R and Matlab/Octave, but you do not have to worry about the jargon and the technical details of the data science process. We provide a scope of work in plain understandable English (or French and Spanish) and deliver a report focusing on solving your issue, with straightforward recommendations and support for implementation.
Specifically, the services we provide are:
Optimization of asset productivity (process optimization by a learning algorithm)
Predictive asset maintenance
Event-based downtime prevention, e.g. the prediction and avoidance of the events inducing downtime in your process, for example:
- SAG mill overload
- conveyor damages (fire, malfunction)
- crud formation in the SX plant and phase continuity reversal
- permeability loss in the leaching pads
- recovery loss due to ore variability
Refining the results of your R&D and process development results and helping you to make the most of these costs accounts (by unearthing unidentified correlations and statistically confirming hypothesis).