The industrial world is rapidly maturing to cope with Industry Revolution 5.0 (IR 5.0), which necessitates an accelerated transformation in industries, particularly the metals and mining industry, to remain sustainable, resilient, and expanding. The need for deeper and more precision mining for high-value rare minerals is growing in the IR 5.0 era. One of the most important practices to remain transformative is leveraging Data Science (DS) (i.e. Science on the Data) which includes simple data analytics (DA), data mining (DM), and leveraging algorithms of Artificial Intelligence (AI). AI, in the context of mining, can be used for predictive maintenance, real-time monitoring, and automation of certain processes. I will use these popular terms DM, AI and DS, interchangeably, as DM and AI are more popular terms in industry usage.
Over the last two decades, the minerals and mining industry has made great progress in analyzing the data and adopting data analytics in some key areas across the lifecycle of the mining process. For instance, data analytics has been used for strategic decision making to adjust to market demand, exploring new business opportunities, enhancing environmental sustainability, safety and health impact, real-time monitoring and control of operational activities, reducing exploration time, and minimizing downtime and maintenance costs.
The progress of the last two decades has primarily focused on optimizing the existing workflows. However, Data Science is not just about optimizing the current workflows but it also empowers automation and innovation. So, there is much more to be done to take full advantage of Data Science along with implementation of emerging technologies like IOTs (Internet of things), AR/VR (Augmented Reality/Virtual Reality) and AI (Artificial Intelligence) at scale.
Compared to many other industry verticals, the mining industry has several unique challenges due to the risk associated with the activities; hence, getting carried over by certain practices from a low-risk industry is not advisable. For example, taking a data-driven model to production in marketing, the e-commerce industry is much easier and risk-averse. Here, we need to validate, validate, and validate before we take data-driven models to the edge of the mine. So, then, how can we accelerate the adoption and implementation of value-focused data analytics models? We can do that by addressing some of the significant challenges:
• Adaptability to emerging technology solutions
• Emerging technologies’ pace of maturity and the value generated are exponential. To leverage exponential technologies like AI, AR/VR, IOTs, and others, the industry has to be more adaptable and agile and learn from successful regulated, high-risk industries.
• Belief is that one needs all the data or large volumes of data to build meaningful AI models.
• The myth that for building data analytics and AI models one needs large volumes or all the data to begin with. Similar to mining, where minerals are exploited in a phased manner, data can be exploited in a phased manner, starting with a limited set and scaling with horizontal and vertical integration to build value-focused models.
• Culture of the organization from top to bottom
• Cultural change within an organization in various areas is required; for example, (a) organizations need to implement data democratization with proper governance rather than keeping the data locked in silos, (b) shift from scheduled maintenance practice to predictive maintenance, within the regulation framework, to reduce cost and complexity, (c) be a first mover in implementing AI-driven solutions, and (d) become technology agnostic.
• Data – Dark data, data quality
• Most organizations believe that their data quality needs to be better for leveraging data-driven models, and hence, they have stored data without generating value, thus creating a significant amount of dark data. Dark data refers to the unstructured, unanalyzed data that organizations collect but do not use. However, one can exploit the dark data through systematic, AI-driven analysis and evolve it so that future data is smart data with the highest possible quality metrics and provides real-time value for transformation.
• Edge accessibility
• Today, technology solutions are mature enough to provide analytics and insights in real-time; however, more work is needed in the mining industry to make the edge more accessible by providing new network technologies, robust computing devices, and appropriate governance practices.
• Futuristic workforce
• The industry has a trained workforce. However, new specialized skills in data analytics, cloud computing, edge computing, data science, AI, and digital transformation are needed to take full advantage of emerging technologies. While there is a shortage of specialized skilled workforce, there are alternative solutions that are more meaningful than just plain outsourcing to large consulting firms. One of them is the upskilling of existing mining industry talent with the above-mentioned skills using customized and contextualized workshops, boot camps, and projects.
As the minerals and mining industry address these challenges, they can realize
• Significantly improved processes for exploration with lower cost and reduced time also provide enhanced collaboration with ease, allowing strategic decisions to be made even faster.
• AI models for predictive monitoring and maintenance via sensors, IoT, and other data sets can predict failures well before they occur, thus reducing downtime and maintenance costs. This reassurance about the benefits of technology should make the audience feel confident about the industry’s future.
• AI applications will enable the industry to shift its focus from operational efficiency to operations efficiency, not just on industry benchmarks but against its own benchmark year over year. This shift in focus should inspire the audience about the industry’s potential for growth and improvement.
In summary, while the minerals and mining industry has made some progress in leveraging data analytics, there is a long way to go to be an IR 5.0-ready industry, and significant capital loss and inefficiencies are yet to be reduced. AI-driven applications will lead the industry to accelerate the transformation in this digital era, so they increase their focus on deep mining, high-value mining, and rare mineral mining at a reasonable cost and high value.
About the Author
Dr. Satyam Priyadarshy, Founder of Reignite Future is a transformational leader who leaves an indelible mark on the industry, educational. and research institutes, with his unique blend of scientific knowledge, technology expertise, and business acumen. He is one of the most sought-after transformational leaders in energy, oil and gas, mining and emerging technology industry. He is an adjunct/visiting Professor at number of universities/institutes in USA, India and other countries.
