05 January 2018
As part of our FutureSmart MiningTM approach to innovation and sustainability, we have developed four aspirational concepts that we are working to achieve. In this article, we take a look at the Intelligent Mine…
Picture a mine where data science and machine learning combine to create a truly smart mine, transforming vast quantities of data into predictive intelligence (from sensor to boardroom).
This is the vision behind our Intelligent Mine.
Huge advances in computational power from advanced “big data” analytics to machine learning have the potential to deliver a quantum leap in the value across the value chain by turning data into insight about the probability of future events.
Mining is data-rich, from the temperature of truck tires to the speed of impellers on a pump, so the potential applications are truly immense: and, in the face of ever increasing sources of more complex data (internet of things, automation, real-time monitoring), we need to measure what matters to support more informed decision making.
This means being able to consolidate our existing data, and interrogate it using advanced analytics to find new patterns and trends, and to answer business questions that will deliver value - from mine to market and sensor to boardroom.
By taking this to the next level, these decisions feed back into the automation process and work with machine learning algorithms, improving with each and every iteration. So the benefits are exponential.
With each application being a significant project in its own right, the focus, for now, is on where we have well-defined questions, and readily available data. We are looking, in particular, at opportunities to optimise excavation. Our latest work is concerned with identifying mineral and material properties using analytics for real-time drilling analysis, hyperspectral core imaging, and geological modelling software using 3D and VR to generate and interpret predictive data models. We are also developing customised learning algorithms to predict control parameters required by our plants. Another important application is condition monitoring and predictive maintenance (learning from the airline industry in particular).
The logical end point is a fully integrated, systemised and ultimately self-learning operation that will help remove the uncertainty and the huge variability that characterises mining today.
Picture this: The Modern Mine
Picture this: The Concentrated Mine
Picture This: The Waterless Mine