1071
Can A.I. Save the Mining Industry?
Dr Rajiv Chandramohan, Matt Pyle
Ausenco
ABSTRACT: The mining industry is at the cusp of re-innovating itself and transforming into a data-driven
business driven by efficiency and maximising value. This vision of the future is no different from any mining
business’s current objectives–maximising the extraction process’s value while minimising the environmental
impact footprint. Will the age of Artificial Intelligence (AI) transform the industry into a fully sustainable
business? This paper explores the importance of AI in mining, from exploration to design and operation. Key
focus areas include transforming current operational practices, introducing technology for effective operation
and a guide to leveraging data to drive efficiency.
Keywords: Artificial Intelligence, Sensors and Instrumentation, Data Management, Operational Optimisation
THE BEGINNING
Alan Turing, who is described as the father of ‘modern com-
puting’, defined Artificial Intelligence (AI) as machines that
think (Turing, 1950). In his seminal work, ‘Computing
Machinery and Intelligence’, Turing postulated that a
computing machine can think like a person based on sim-
plified reasoning and logic. To be a credible postulation,
the machine’s solution is inferred from a set of non-direct
questions and answers, where the objective is to deduce
the outcome from historical interpretation. The premise of
Turing’s work is the backbone of most modern computing
and AI.
Since the invention of the first transistor by John
Bardeen, Walter Brattain and William Shockley from
Bell Labs in 1947, computing power and efficiency have
increased steadily in the last 50 years (Gaudin, 2007).
Gordan Moore, founder of Intel, generalized that comput-
ing power (number of calculations undertaken per cycle
time) roughly doubles every two years, Figure 1. On the
other hand, AI systems operating in high-performing com-
puters have significantly increased computational perfor-
mance, outperforming Moore’s Law by a factor of three,
primarily driven by parallel computational processing,
advanced algorithm techniques, and access to large data sets
for training the software (The Physics arXiv, 2022).
In his influential book, The Meaning of the 21st
Century: A Vital Blueprint for Ensuring Our Future, James
Martin postulated that machines will eventually surpass
human intelligence in the 21st century (Martin, 2006). This
moment in time was termed the singularity and was coined
‘tertiary evolution’, stemming from biology to describe a
species’ final goal in creating automated evolution (or arti-
ficial beings), bypassing natural evolutionary progression
as famously described by Sir Charles Darwin. Primary and
secondary evolutionary terms describe a species’ natural
and evolved progressions the latter aims to augment natu-
ral evolution by manipulating the DNA sequence to extend
life.
With recent advancements in the large language mod-
els (LLMs), such as OpenAI’s ChatGPT ® (OpenAI, 2024)
and Google’s Bard ® (Bard, 2024) and the cautious views
on the development of AGI (Artificial General–Purpose
Intelligence) and the future of humanity by prominent sci-
entists, such as Geoffrey Hinton (Brown, 2023) and Stuart
Can A.I. Save the Mining Industry?
Dr Rajiv Chandramohan, Matt Pyle
Ausenco
ABSTRACT: The mining industry is at the cusp of re-innovating itself and transforming into a data-driven
business driven by efficiency and maximising value. This vision of the future is no different from any mining
business’s current objectives–maximising the extraction process’s value while minimising the environmental
impact footprint. Will the age of Artificial Intelligence (AI) transform the industry into a fully sustainable
business? This paper explores the importance of AI in mining, from exploration to design and operation. Key
focus areas include transforming current operational practices, introducing technology for effective operation
and a guide to leveraging data to drive efficiency.
Keywords: Artificial Intelligence, Sensors and Instrumentation, Data Management, Operational Optimisation
THE BEGINNING
Alan Turing, who is described as the father of ‘modern com-
puting’, defined Artificial Intelligence (AI) as machines that
think (Turing, 1950). In his seminal work, ‘Computing
Machinery and Intelligence’, Turing postulated that a
computing machine can think like a person based on sim-
plified reasoning and logic. To be a credible postulation,
the machine’s solution is inferred from a set of non-direct
questions and answers, where the objective is to deduce
the outcome from historical interpretation. The premise of
Turing’s work is the backbone of most modern computing
and AI.
Since the invention of the first transistor by John
Bardeen, Walter Brattain and William Shockley from
Bell Labs in 1947, computing power and efficiency have
increased steadily in the last 50 years (Gaudin, 2007).
Gordan Moore, founder of Intel, generalized that comput-
ing power (number of calculations undertaken per cycle
time) roughly doubles every two years, Figure 1. On the
other hand, AI systems operating in high-performing com-
puters have significantly increased computational perfor-
mance, outperforming Moore’s Law by a factor of three,
primarily driven by parallel computational processing,
advanced algorithm techniques, and access to large data sets
for training the software (The Physics arXiv, 2022).
In his influential book, The Meaning of the 21st
Century: A Vital Blueprint for Ensuring Our Future, James
Martin postulated that machines will eventually surpass
human intelligence in the 21st century (Martin, 2006). This
moment in time was termed the singularity and was coined
‘tertiary evolution’, stemming from biology to describe a
species’ final goal in creating automated evolution (or arti-
ficial beings), bypassing natural evolutionary progression
as famously described by Sir Charles Darwin. Primary and
secondary evolutionary terms describe a species’ natural
and evolved progressions the latter aims to augment natu-
ral evolution by manipulating the DNA sequence to extend
life.
With recent advancements in the large language mod-
els (LLMs), such as OpenAI’s ChatGPT ® (OpenAI, 2024)
and Google’s Bard ® (Bard, 2024) and the cautious views
on the development of AGI (Artificial General–Purpose
Intelligence) and the future of humanity by prominent sci-
entists, such as Geoffrey Hinton (Brown, 2023) and Stuart