Mining, Resources and Energy Impact of AI and ML

I wanted to share some experiences I've had working in the Data and AI field at #Microsoft for the past few years and focused on the Mining, Resources and Energy (MRE) sector where I am an advisor to our top customers strategic and enterprise customers.

In this article I will cover a quick perspective around AI and Machine Learning in the MRE Industry. I'll just call it AI going forward - yes, I know there is a difference. I have seen first-hand the dramatic impact that AI and can have on this sector. I hope I can bring some perspective to you and encourage you to start experimenting with real world solutions.?

1. My most shocking finding is that implementing AI technologies and at least proofing it is very cost effective. Yes, it is pocket change compared to other investments. The cost to implement AI in the MRE industry has drastically decreased in recent years, making it accessible to a much wider range of organizations. I've often seen executives skeptically eyeing the total cost of the project because it looks unbelievably small!
2. The return on investment is fast. Many of you in the industry may be familiar with how slow-moving?analytics projects typically takes a long time to return on investment, as the data must be ingested, modelled and analyzed to identify opportunities and solutions. The process may also involve much user acceptance validation, training, and a slow adoption curve. Results may take weeks or months to be realized and outcomes are highly dependent on the organizational culture. Most AI-driven projects, on the other hand, can see a return on investment much more quickly. It typically stands on the shoulders of those analytics projects, but not always. With AI much is automated and working in the background resulting in it being completed in a fraction of the time. The AI system can generate actionable insights quickly, and the results can be implemented and measured quickly, leading to a much faster return on investment. For example, AI-driven predictive analytics can provide insights on equipment maintenance and operational costs, as well as identify opportunities for cost savings. In predictive maintenance we are seeing massive returns because the MRE industry has much high value equipment.?
3. Use cases are everywhere, but especially in the areas that matter most like safety, sustainability, efficiency and cost savings. So, AI can help operations to better manage and analyze large datasets and drive action faster. With the amount of data that my customer's operations generate, traditional methods of analysis can become tedious and time-consuming. AI-driven analytics can process this data quicker and more accurately, helping to identify trends and patterns to help streamline operations. For example, AI can also be used to help with safety by identifying potential safety risks searching through human written or digital reports, video streams from cameras or telemetry data from IoT devices. This can be used to develop or refine safety protocols that can be implemented quickly. This can ultimately help to reduce the number of accidents on sites. What use cases have you come across?
4. Not everyone is a fan, and some will throw out the baby with the bathwater. There are those who see AI as a way to displace human workers, leading to job losses and unemployment - this is defiantly not what is need in the current climate of uncertainty. I would not shy away from these discussions. Machines have always threatened that they can be more efficient, cost-effective, and reliable than human labour. But we have always had the choice around WHERE we use these tools that can amplify what we as humans want to achieve. I've seen this at a customer where they could reduce the amount of a certain skilled worker for an automated AI system that is over 45% more efficient and never sleeps or gets tired, only better. They chose not to go ahead - good for them. ?
5. There are some real problems I worry about. I often wonder where the threshold of return would be to overturn some ethical considerations. In the previous example in point 4 - would they have made a different decision if the returns where at 500%? In addition, many that are concerned that AI could lead to a lack of transparency in the industry. At Microsoft, we have built some good tools to unpack its inner workings better - but these tools need to be used. There are many other things I worry about of course, but mostly I worry about the willingness of leaders and regulators to enforce and keep good controls regarding ethical issues. What do you think are the long-term consequences of using AI to influence our judgement?

The impact of AI in the mining, resources and energy industry is vast and growing. As the cost to implement this technology continues to decrease, I believe we will continue to reap the benefits of AI technology while exercising leadership over the areas we choose to implement it.

In the near future, I expect AI to be a key part of MRE operations, helping to drive an impact on safety, sustainability, efficiency and cost savings.