TL;DR:
- SRK Consulting is using machine learning to revolutionize mineral exploration.
- Machine learning accelerates data analysis and reduces human bias in the decision-making process.
- SRK has been an early adopter of machine learning in mineral exploration.
- Crowdsourcing exploration competitions have fueled the proliferation of machine learning in the field.
- SRK combines data-driven machine learning with knowledge-driven geological reinterpretation.
- Knowledge-driven prospectivity modeling allows assessments in greenfield environments with limited data.
- SRK collaborated with BHP and DeepIQ for machine learning pilot studies in the Chilean Andes.
- Machine learning helps mining companies meet growing demand forecasts by guiding exploration.
- The integration of machine learning and domain expertise leads to enhanced accuracy and efficiency in exploration programs.
Main AI News:
SRK Consulting, a renowned international mining consultancy firm, is revolutionizing mineral exploration by leveraging the power of machine learning. In the ever-evolving landscape of artificial intelligence (AI), machine learning has emerged as a crucial methodology within the mining and industrial sectors. It is not only providing a competitive edge to major mining companies but also bolstering greenfields and brownfields projects by enhancing prospectivity modeling in mineral exploration.
The mineral exploration industry, characterized by its diverse and voluminous data, presents immense opportunities for machine learning. By employing this technology, not only can the analysis of assay results be accelerated, but it also mitigates the influence of human bias in the decision-making process.
Through the ingestion of data into a robust computing platform, machine learning algorithms discern patterns and gain valuable insights, enabling accurate predictions that support future drill programs and exploration initiatives. The remarkable aspect is that machine learning accomplishes this feat in a fraction of the time it would take a geologist to manually analyze the data and formulate estimations.
SRK has emerged as an early adopter of machine learning within the realm of mineral exploration. The firm’s interest in this emerging technology was sparked by Antoine Caté, a senior structural geologist at SRK. Caté actively participated in several crowdsourcing exploration competitions, including the Integra Gold Rush Challenge held in 2016. This unique event invited geologists and data scientists to unlock the potential of decades’ worth of exploration data and generate targets for Integra’s Lamaque project in Val-d’Or, Québec. Remarkably, the competition boasted a staggering prize pool of $C1 million.
Following suit, OZ Minerals launched its own Explorer Challenge in 2019, offering a $1 million prize pool to individuals capable of developing innovative exploration approaches near the Prominent Hill copper-gold mine in South Australia. The pivotal commonality across these competitions was the abundance of data at the participants’ disposal. The participating companies possessed extensive catalogs of valuable data, which provided fertile ground for the exploration of new approaches and the introduction of cutting-edge innovations.
This influx of talented geologists and data scientists from around the globe led to the rapid proliferation of machine learning in the field of exploration. Mark Rieuwers, a principal geologist at SRK, highlighted how the company’s Canada and Australia offices collaborated during OZ Minerals’ Explorer Challenge. Rieuwers and his team supplemented Caté’s expertise in machine learning with their profound knowledge of local geology.
Rieuwers shared, “OZ Minerals made a significant portion of their data publicly available. Participants were required to register and develop new data-driven targets using various approaches. Antoine focused on the machine learning aspect, and we realized that this specific exercise involved an iron oxide copper-gold (IOCG) style of exploration. Some of us in the Australia offices had considerable experience in this area, particularly with South Australian geology.”
Drawing from this experience, SRK developed an integrated workflow that amalgamates data-driven, machine learning-based methodologies with knowledge-driven geological reinterpretation. This cohesive framework has been successfully applied to numerous projects, establishing a new paradigm for mineral exploration.
SRK Consulting’s approach to prospectivity modeling sets it apart by creating feedback loops that leverage both machine learning intelligence and the domain expertise of its geologists. This unique philosophy, known as knowledge-driven prospectivity modeling, offers distinct advantages in the field. Unlike traditional machine learning algorithms that rely on known deposits or training sites, knowledge-driven interpretations require only concepts and the combination of traditional algorithms and formulas.
Mark Rieuwers emphasized the benefits of knowledge-driven interpretations, stating, “You can be in a greenfields environment where there aren’t many known deposit types of the ones you’re looking for and still be able to make an assessment.” While machine learning thrives in data-rich environments with numerous training sites, the integration of knowledge-driven methods and machine learning creates a powerful synergy, bridging the gap between limited data availability and accurate assessments.
In collaboration with BHP and DeepIQ, SRK embarked on a machine learning pilot study between 2020 and 2021 to explore novel methods of delineating large porphyry copper deposits in the northern Chilean Andes. Building upon the insights and workflow developments from the pilot study, a subsequent study was conducted in 2022 in a greenfield area located in another part of the Andes. Although the companies initially believed the terrain to be well-understood and extensively explored, the pilot study unearthed new exploration ideas that challenged existing notions.
The mining industry is facing mounting demand forecasts in the years ahead, underscoring the need for more accurate and precise exploration programs to unlock additional deposits. Machine learning plays a pivotal role in expediting this process by providing valuable insights and guiding mining and exploration companies toward their next discovery.
As SRK Consulting continues to embrace the potential of machine learning and its integration with domain expertise, the industry can look forward to a future where exploration programs are characterized by enhanced accuracy, efficiency, and the successful identification of untapped mineral resources.
Conlcusion:
the integration of machine learning in mineral exploration, as exemplified by SRK Consulting’s pioneering approach, holds significant implications for the market. The utilization of machine learning algorithms to accelerate data analysis, mitigate human bias, and uncover new exploration ideas enhances the efficiency and accuracy of exploration programs. This technological advancement empowers mining and exploration companies to make data-informed decisions, optimize resource allocation, and unlock untapped mineral resources.
As a result, the market can anticipate improved operational outcomes, increased productivity, and the potential for significant discoveries in the mining industry. The successful integration of machine learning and domain expertise sets a new standard for exploration practices and reinforces the importance of embracing innovative technologies to stay competitive in the evolving market landscape.
