With the total operating expenses of the top mining companies worldwide reaching USD $15 billion, efficient extraction of raw materials is one of the biggest buzzwords right now. While it is undeniable that the human contribution to the mining sector is irreplaceable, McKinsey estimates that by 2035, the age of smart mining achieved through autonomous mining using data analysis and digital technologies like artificial intelligence (AI) will save between $290 billion and $390 billion annually for mineral raw materials producers.
Artificial intelligence is the ability of a digital computer or computer-controlled robot to perform tasks commonly associated with intelligent beings, such as learning from past experience. The systems powered by artificial intelligence use different algorithms to organize and understand vast amounts of data, with the purpose of making optimal decisions. But how can AI be supportive of the rapidly growing mining industry and drive towards smart mining? The following sections will cover some examples to answer that question.
PreScouter has a partnership with SRI International, a non-profit AI organization aligned with PreScouter’s goals of identifying new opportunities, developing products, and creating custom solutions. Contact us here or email us at firstname.lastname@example.org with your AI-related request and an AI specialist will be in touch to help.
Gold exploration needs:
Historically, major gold discoveries have outpaced exploration spending; however, in 2012, spending peaked at USD $6.1 billion. And a total of four deposits were discovered between 2013 and 2015, whereas before 2012, the industry found an average of 10 per year.
Future discoveries are likely to be deeper and more complex ore bodies that are hidden by overburden and other geological structures. Machine learning has shown to be a promising tool in processing the overwhelming amount of data collected during exploration and is key to finding these more complex deposits. The way it accomplishes that is by:
- Using training data (maps with known deposits and characteristics)
- Data is cleaned, transformed, interpreted, and then used to train machines in order to predict targets
- Targets are identified with high potential for mineralization or extraction
Goldspot Discoveries Corp uses artificial intelligence for improving mineral exploration by ingesting different data from which to discover potential gold deposit locations. The company, through AI, advanced analytics, and simulation modeling, has the potential to add more than $11 billion in additional value to the mining industry.
Ensuring remote exploitation error elimination through data analysis:
If artificial intelligence can support the user to better understand the environment and the terrain where exploitation is to begin, the exposure of frontline mineworkers to dangerous situations is going to be reduced. This has been an idea brought up in the oil & gas industry as well. With the support of big data, companies can save 80% while locating new mines, compared to the traditional methods with all the complicated and expensive bureaucratic processes that precede field exploration. Additionally, the derived data can be used in later reclamation programs, where re-establishment of the natural environment of the area is required.
But we should always keep in mind that in order to achieve valuable exploration data analysis, we need to overcome one of the biggest challenges in machine learning in the exploration field: achieving relevant and clean data sets.
Autonomous mining vehicles inside the mine that can be remotely operated and managed:
Logically, mining safety metrics will be significantly improved by the use of artificial intelligence. This is one of the reasons that approximately 50% of Rio Tinto’s haul truck fleet is already operating autonomously. On the other hand, it is important to take into consideration aspects such as the need for a high level of cyber security.
Imagine some very expensive mining equipment like an autonomous excavator being hacked. It could be easily turned into a mass of metal by simply crashing it into a mountain. The high level of knowledge of complicated electronics is a requirement in order to maintain these kinds of vehicles. And for sure, an internet connection is needed, as data collection is essential for machine learning, autonomous mining, and most of the time, mining is happening far away from the civil world and electricity.
How can autonomous personnel tracking sharpen safety during the mining process?
Imagine wearable sensors to continuously monitor worker behavior, generating real-time data that is then automatically analyzed to spot problematic behavioral trends and recommend highly focused remedial training in a simulator-based environment. Or, with the support of cameras that capture videos during the mining process, automatically monitoring mine-site personnel for safety, security, and process analysis purposes. ThoroughTec Simulation provides personnel tracking systems that support mining companies to optimize training interventions that best match individual worker needs. One example technology can be seen in the video below.
Autonomous mining to remotely identify copper grades with the lowest possible error rate:
It is possible, with the use of machine learning, to vastly enhance image recognition. With the support of vision technology, rock samples and drilling data can automatically determine the type of discovered minerals with a high level of accuracy, as opposed to the time and effort that can be spent manually reviewing and labeling samples from different rocks. As Ionic Mechatronics mentions, “The ones who benefit with AI are the ones that make the investment with the vendors and have first access.”
Neural networks for efficient beneficiation:
Smart sorting is already applied in the mining industry for minerals and ores, and artificial intelligence algorithms powered by color sensors and X-ray data are improving the quality and quantity of, for example, the diamond recovery process. Deep neural networks, which use networks with massive amounts of data to learn, can provide even better grade ore and potential cost savings in the mining industry through significant enhancement of image and speech recognition.
Another important aspect is that the speed of data management that these neural networks can provide allows the development of real-time systems that can quickly recognize potential problems or even danger during the mining process. Data handling through neural networks can improve not only mining costs but also the safety level during the mining process.
The future of mining is smart:
The applications of artificial intelligence in the mining industry are uncountable. With the Green Revolution already in progress, the need for raw materials such as silicon for solar PV panels and lithium for batteries is significantly increased. Artificial intelligence shifts raw materials mining from a people-oriented operation to a process-oriented one, which is critical to ensure appropriate health and safety conditions for the mineworkers, a high level of accuracy, error elimination, and a faster decision-making process.
With the mining sector requiring more and more effective and efficient operations such as autonomous mining, the industry requires investments in various artificial intelligence technologies. Some mining companies would like to invest while some others are not yet ready. But either way, the need for smart mining is already here.
If you have any questions or would like to know if we can help your business with its innovation challenges, please contact us here or email our Natural Resources Lead, Eric Joyce, directly at email@example.com.