The Intelligent guide to: digital twins in mining

What is a digital twin?

In its simplest form, a digital twin is a virtual simulator of mining operations; a carbon copy in the digital world where management can manipulate a huge number of variables over a given time period to see how changes will affect both up- and down-stream processes. By giving teams the ability to simulate changes in the mining process, however big or small, quickly and with no risk to the actual (rather costly) operation, the pressure associated with making mistakes is removed. In short, they are a very clever optimisation tool.

CIM Magazine ran a comprehensive article in October called How to simulate an entire operation before production. Does what it says on the tin; a good introduction to digital twins, key players in the mining space and operations that are taking advantage of the concept.

How are they currently being applied?

Mineral processing facilities or concentrators, the vast majority of which employ automated data collection through process control systems, are a natural starting point. Dr Penny Stewart who heads up the team at Petra DataScience presented a case study at the 2018 Future of Mining conference in Sydney, Australia, based on Petra’s work at PanAust’s Ban Houayxai gold-silver mine in Laos. In 2018, a machine-learning algorithm was installed at the operation to predict future metallurgical characteristics based on previous process plant performance. You can read about the project, plus the history of digital twins here.

Rio Tinto is another miner that has jumped on the data science bandwagon. The company’s new ‘intelligent mine’, Koodaideri in Western Australia, is using a digital twin to simulate its processing plant. itNews covers the basics in this article or, for a more in-depth read, and if you have an online subscription, try International Mining’s November issue which carries an interview with Rio’s managing director of planning integration and assets for iron ore, Matthew Holcz.

Anglo American was another early adopter and has been using a digital twin to improve haulage at its Los Bronces site in Chile since 2016. BloombergNEF documents the installation here.

Which companies can provide the technology?

Digital twins were developed outside of the mining sector and so, naturally, it is diversified firms that are bringing it into this space, leveraging technologies that have generated results for clients in other industries to improve mining operations.

GE is one of the big names leading the way. Creating a digital twin is part of its Digital Mine optimisation solution which is powered by the Predix industrial internet platform. Accenture also cites twins as one of the five pillars of its Digital Transformation in Mining solution. The company uses virtual and augmented reality tools to mimic physical locations for immersive employee training (useful for safety scenarios and maintenance applications), and to give greater insight into equipment assets.

Deloitte gave a nod to digital twins in its Digital Revolution report, published for the Diggers and Dealers event in 2017, and Bentley Systems, which has its fingers in a number of infrastructure pies, has been working with Atos and Siemens to offer digital twin capabilities for mining projects.

Mining Magazine ran an interview in September with mine planning and modelling specialist Maptek and software provide LlamaZOO about their collaboration in the mining space. Although you will need a subscription to read that one, LlamaZOO also has some useful info on digital twins and their applications on its website.

Research institute CSIRO has also been working with a number of companies, including Hitachi and Rio Tinto, and CRC ORE has done some interesting work in concentrator simulation.

These are just a handful of applications and solutions. If you know of any others that merit discussion, feel free to comment below.