Machine Learning and Artificial Intelligence for Underground Mining: FAQ

Artificial Intelligence in Underground Mining Newtrax

In the news, the exciting fields of Machine learning (ML) and Artificial Intelligence (AI) keep announcing breakthroughs around the world and in various industries.

The mining industry has also joined in and recently started to explore and seek the benefits of this new ‚Äúgeneral-purpose technology‚ÄĚ with ML & AI enhancing safety and productivity in underground hard rock mines.

Newtrax has been actively investing in resources towards building an AI knowledge base and expertise for applications in underground hard rock mines. With this, we have created a short ‚ÄúFAQ‚ÄĚ to demystify some of the common questions that the industry might be asking.

Is AI really useful or is it only hype?

It is a reality. Today, AI is helping society in applications like recognizing cancer tissue faster than experts, finding criminal patterns in tons of financial transactions, doing speech and video recognition for converting huge video banks into structured information, chatbots that speak and understand speech to become better assistants, and the list goes on.

What is the difference between Artificial Intelligence and Machine Learning?

Machine Learning is actually simply a sub branch of Artificial Intelligence, which also includes a sub branch of its own called Deep Learning. Artificial Intelligence also includes both Operations Research and Heuristics which can also be considered AI.

The illustration below explains it well:

AI Definitions

Are ML and AI relevant to the mining industry?

Mine managers are accustomed to analyzing the ROI before investing in technology. The same goes for paying for the services of an ML or AI bot. The expense must pay for itself within a given period. In the field of AI, that decision is usually easy to make, especially for proven algorithms.

ML and AI are very relevant to the mining industry, in fact, there are numerous practical ways ML and AI can be used within mine operations.

  • Re-optimizing in real-time the shift plan of mine activities for overall mine productivity upon occurrence of an unplanned event.

 

  • Self improving predictive algorithm avoiding breakdown of mobile equipment based on continuous monitoring of vehicle sensor activity. This alone can save countless hours of equipment downtime.

 

  • Predicting in real-time activities that can become bottlenecks

 

  • Finding patterns in events leading to hazards, accidents or fatalities which also enables mine operators to evaluate and understand where their safety standards are lacking.

 

  • Finding patterns in productivity variance. This can help in operations planning and time management since it is highlighting areas where improvement is immediately possible.

Should mining companies care about IoT and AI?

Definitely. According to a recent white paper by World Economic Forum, compared to other industries, especially customer-facing ones, the mining and metals sector is considered to have lower levels of digital utilization. However, For those organizations that can move from being digital laggards to digital first movers, the value is real.

A recent article by Techemergence noted that the nature of the industry means mining companies are hyper focused on any way to improve productivity and efficiency. So it is not surprising that some mining companies are becoming very aggressive about exploring artificial intelligence, machine learning, and autonomous equipment to find ways to improve efficiency.

Autonomous equipment and smart equipment have been well tested and are being rapidly expanded to new mine operations. Many of the biggest mining companies in the world have been significantly growing autonomous haulers fleets for a few years which is a strong proof that they are applications which provide a positive return on investment.

Mining is relatively unique in that it requires big capital investments in extremely expensive pieces of large equipment, and a main way to be competitive is a ruthless focus on efficiency. This is going to always make the industry a logical choice for pushing the envelope when it comes making equipment smarter and applying algorithms to make better decisions.