By Lauren DeLorenzo, Assistant Editor, Mining Magazine
For an industry that boasts incredibly advanced machinery and sophisticated remote operating systems, there is a lot of room to improve the way information is stored, accessed and implemented across mining operations. Here, we look at challenges of managing big data and how miners can use these systems to drive down costs and spur forward-thinking solutions.
As miners look towards sustainability and increased efficiency outcomes, it’s no wonder that ‘big data’ has become a bit of a buzzword in recent years. In fact, use of the term in mining companies’ reports increased by 72 per cent between the first and second quarters of 2021, and rose by 150 per cent since 2016, according to a GlobalData analysis.
Big data refers to large pools of data that can be collated and analysed to identify patterns and trends. Data can be brought in from multiple sources and be used to make big decisions and identify areas for improvement and growth.
Technological advances mean equipment can be fitted with sensors, which can generate huge amounts of geoscientific, asset condition and operational data that paints a detailed picture of operations in real time.
Big data can be woven into every corner of mining operations, measuring variety, volume and velocity across the business – giving key decision makers a comprehensive overview of where productivity, safety and margins can be refined.
This type of analysis is expected to play a huge role in the mining industry’s advances towards reducing waste, enhancing productivity and remaining competitive in the industry.
But to reach the full potential of this data, miners need to invest in analytical technology that can unlock the true value of this asset, with many experts seeing data analytics as a massive game changer for the way the industry operates – from operational efficiencies, workforce management, real-time planning and so much more.
Miners are looking to cloud-based systems, automation and artificial intelligence (AI) to boost predictive capabilities, monitor operations in real time and gain a competitive advantage.
Moving to cloud platforms
Increasingly, analysts are moving away from legacy information storage systems and towards cloud-based information storage. The huge, clear benefit of this is that it means data streams can merge into one, improving agility and speed of data collection and analysis, and meaning that valuable insights can be accessed even more quickly, along the entire supply chain, sometimes in real time.
And the faster the insights are accessed, the faster they can be implemented and create savings. For example, an iron ore truck fitted with sensors can be continuously taking in data, analysing the information, and suggesting alternate, time-saving routes to the driver.
Small fuel, energy, cost and efficiency savings can quickly accumulate to have a large financial impact. Moving data to the cloud also means it is accessible from anywhere – so mining companies can access information remotely. In turn, this boosts the ability of different teams to collaborate, as they can all build on a single data point.
On the other hand, traditional information storage requires a lot more maintenance, and can be more prone to inaccuracies. Cloud-based platforms are also flexible in storage capacity, allowing operators to scale up – a huge asset for an industry that generates massive amounts of data.
IoT and predictive analytics
Advancements in industrial IoT systems have made it possible to collect data in difficult places such as underground mines, revealing insights that were previously unavailable.
Equipment fitted with sensors can provide real-time information about its operation, allowing for immediate or even predictive corrective action, increasing equipment up time. Advanced sensors can monitor vibration, corrosion, temperature and acoustic emission for analysis.
Data trends can predict when equipment repairs are necessary, and when they could be at risk of breaking down, allowing miners to increase overall machine reliability by acting on these issues before they interfere with operations. This results in increased machine lives, better repair times, better product quality, reparation planning and lower maintenance costs.
Predictive analytics are useful for a myriad of situations in the mining industry, not just for equipment maintenance. For example, previously, when a drill ran into a hard rock, the ore had to be analysed before deciding how to proceed.
Now, predictive analytics built on big data can analyse the ore swiftly, delivering information to managers in a much smaller amount of time. Predictive analytics can also help machines make decisions themselves, improving automation around the site.
Analytics can also help site managers get a better understanding of what makes some site days more productive than others, looking into the conditions surrounding them. They can then aim to replicate these conditions for greater productivity and efficiency.
Sifting through all of this data, however, is time consuming. Machine learning (ML) algorithms are able to quickly interpret and analyse data to make future predictions about the operation of the mine.
Minimising costs
Mines typically account for a third or more of operational costs for large-scale operations, and this is expected to increase as the industry navigates operational and regulatory challenges.
But by collecting data from mobile equipment and identifying trends, maintenance costs can be anticipated with precision. This means that the time and cost of maintenance and equipment downtime can be significantly reduced, increasing productivity.
Monitoring the mine’s water, ground and gas levels can also help the operation significantly reduce waste.
Safety and workforce management
Big data and digital technologies can play a huge role in empowering workers with information and in creating advanced safety mechanisms. Automated ground control systems can be used in underground or pit mining, which capture information on ground vibrations.
These systems can collect vibration data and use it to determine the structural integrity of the operation, so that if a dangerous situation such as a ground slide were to occur, the monitoring system could send out an early warning signal for evacuation. The same data could be used to develop safer drill equipment and technologies.
Data capture also allows for real-time monitoring of people, equipment, temperature and environmental factors such as dust, gas concentration and wind speed. Having these insights on hand is a groundbreaking tool to ensure the safety of workers. Monitoring of equipment operating pressure, power and speed allows for analysis that could potentially predict and avoid future risks, such as instances of near-misses.
Challenges
One of the biggest challenges in unlocking the full potential of big data is finding ways to manage and assimilate huge amounts of information from a vast array of sensors, machinery and other sources. Miners must decide which data to collect, how to prioritise it and how to create systems to implement analytics-based solutions.
Other potential roadblocks include integrating new technologies with legacy systems, IP and privacy concerns, and storage and security issues. Increased use of data will mean that digital security measures will need to be strengthened. Interruptions in operations, cyber attacks and data theft should be considered as more devices are connected, making them more vulnerable.
Workers will also need to be shown how to interpret data, and people on the ground will need to understand how data will be procured, stored, maintained and used.
The way forward
Successful integration of big data in mining operations will rely on a focused strategy that is part of the business model. Looking outside of the current portion of the value chain can spark further ideas for creating better value through data.
Going forward, digital standardisation and regulation could define how data is used within operations. These standards could define data ownership and encourage the sharing of information across the ecosystem while maintaining privacy measures.
Digital platforms can be used to improve transparency about how operations are tracking in terms of environmental concerns, sourcing, production and community engagement. Companies can implement policies to allow disenfranchised groups and communities access to data.
Building awareness of how data can be reused, and opening procurement to local tech and communications companies can provide opportunities for startups and small businesses to innovate.
Digital services could be opened up to these organisations, such as cloud hosting, software and support. Ethical management of big data includes building projects and algorithms with awareness of bias in mind, to prevent biased outcomes.
Algorithms are coded by people with conscious or unconscious bias, and one way to prevent this from translating to algorithms is to ensure that design teams are diverse, comprised of people from underrepresented backgrounds.
