A Transformational Turn for a Harsh Industry
Talking about ‘digital transformation’ in sectors such as banking, insurance, telecoms, and retail has become somewhat obsolete. These industries have already traveled a long way on their journey to digitalization. For example, embarking on leading-edge projects with Artificial Intelligence (AI) that facilitate a stronger business-to-customer approach. However, the mining industry typically represents a very different scenario.
Only a small number of mining businesses have truly advanced in digitalization to enhance their mining operations. And the majority of AI projects in mining have been somewhat unsuccessful, primarily due to the challenges imposed by highly variable operating conditions, legacy IT systems, inadequate network coverage, and a highly mobile workforce.
The highly mobile nature of the workforce makes the delivery of technology to help with daily decision-making very different from other industries, where the focus is mostly on desk-based users.
However, the current challenging global economic picture is forcing every mining operation to focus on margins. And this means a laser-like focus on increasing throughput and improving recovery. Hence, understanding the full potential of digital technologies such as AI to translate efficiency and productivity gains into lower costs, higher margins, and stronger, more sustainable capital returns, is increasingly being brought to the table in 2023.
According to the World Economic Forum (WEF) whitepaper, the mining sector’s digitalization has the potential to generate over US$425 billion in shareholder, customer, and environmental value over 10 years – equivalent to 3 to 4% of the industry’s overall revenue during that time – which for sure, is a big transformational turn.
Unlocking The Real Worth of Data
The next quick-growing trend in mining technology is starting to take shape; namely, the ability to unlock the worth from the mountain of data that is meticulously gathered at most mines sites. The value lies in the business insights this data can provide to decision-makers. Information such as the performance of equipment, the quality of orebodies and stockpiles, geospatial data, plant operations, and environmental performance is available. The analysis and presentation of this data (and further prediction of future behaviors) have the power to completely change mine operation management.
In our recent findings from exposure to many C-level mining executives, advancing their Digital Strategy was raised as one of the key topics for 2023 and beyond. Mining companies want to own their data and infrastructure through solutions architected from the outset to cater to their challenging operating environments, utilizing specialist software providers to unlock the potential of that data. Other themes included mine electrification, remote operating centers, and autonomous operations were all pointed out.
How to win at AI adoption
Based on the avalanche of data generated by this army of networked equipment, several miners have adopted some type of daily and monthly reporting. To improve predictive analytics and pattern identification, several mining operations have started to go one stage further and implement discrete components of machine learning and artificial intelligence.
However, at IntelliSense.io, we believe that for a process to truly become efficient you need to be able to accurately estimate the future rather than merely knowing the current or past states. And this is where traditional IoT technology and reporting falter as they are very much focused on the latter.
There is also a requirement to integrate AI outputs into people’s daily workflows. This requires the delivery of AI outputs across both desk and mobile environments, delivering decision support when and where operators need it during their daily work.
At IntelliSense.io we have gone even further with the application of ‘Scientific AI’*, which can also be referred to as a physics-inspired neural network (PNN’s). This unique technology approach plays a fundamental role in all of our solutions. Our understanding of a system or process can be represented by a model, usually expressed as mathematics or algorithms. These models can be used to
- Improve data quality (filtering)
- Predict future behavior
- Impute unobservable quantities
- Make recommendations
Now imagine if you could predict overloads in your grinding circuit or better still optimize the circuit for energy consumption. With our Grinding Application Optimization app, an AI model is used to predict, in real-time, the probability of a mill overload taking place over the next 10-20 minutes (future state). This model is trained on historical Mill feed and performance data so that it knows how the Mill responds to changes in solids feed rate, dilution water, material properties, and control variables like mill speed and ball additions.The possibility to take predictive actions translates into greater savings and more consistent throughput. Even small percentage gains provide a huge return in an industry where everything is conducted on a massive scale.
*Scientific AI is the fusion of physics models with machine learning techniques. It was developed by Intellisense.io to further increase net metal production through the use of our AI solutions
Wrapping Up
For a company to meet and exceed key performance indicators and determine whether its vital controls are operating as expected, current data capture and accurate analysis are crucial.
In order to accelerate digitalization in mines, it is essential to understand how IoT systems work and build a strategic roadmap to adopt emerging technologies such as Artificial Intelligence (AI) and Machine learning (ML) to support the digital shift that is already happening.
IntelliSense.io pre-built applications have a fast implementation framework (with low risk) that help miners in this race against time. We use available mining data, combining spatial (mine) and time-series data (plant) to drive even greater efficiency gains. On the brains.app platform, our suite of real-time decision-making applications uses Scientific AI-based technologies to optimize each process; complete from mine to market.
Interested in know more about it?
