Ep 146: How Machine Learning Can Close the Gap Between Grade Control and Resources. Part 2
Fresh Thinking by Snowden Optiro
Release Date: 02/11/2026
Fresh Thinking by Snowden Optiro
If your resource model is statistically robust, why doesn't it always reconcile with production? In this episode of Fresh Thinking by Snowden Optiro, Melanie Bulley and Senzeni Mandava Matondi explore why good resource estimates can still lead to disappointing reconciliation results. They discuss the impact of grade smoothing, explain the concept of conditional bias, and explore why reproducing the average grade is only part of the story. The conversation also covers practical techniques used by resource geologists to assess estimation performance and improve local predictions that support...
info_outlineFresh Thinking by Snowden Optiro
How do you model mineralisation when geological boundaries aren't clearly defined? In this episode of Fresh Thinking by , Senior Consultant, Melanie Bulley is joined by Principal Consultant, Senzeni Matondi to explore how Categorical Indicator Kriging (CIK) can be used to model diffuse mineralisation more objectively and consistently. The discussion covers: Why traditional wireframing can be challenging in diffuse deposits How probability-based domaining improves consistency The role of CIK in reducing subjectivity Building more robust and defensible resource models Whether you're a...
info_outlineFresh Thinking by Snowden Optiro
What do consultants see that mine sites often miss? In this episode of Fresh Thinking bySnowden Optiro, Rayleen Hargreaves is joined by Ian Glacken to discuss the most common mine-to-mill reconciliation challenges, from stockpile management and siloed teams to understanding variance and improving decision-making. Learn why reconciliation is more than a monthly report—it's a powerful tool for improving mine performance.
info_outlineFresh Thinking by Snowden Optiro
Which commodities matter most in the global race for critical and strategic minerals? In this episode of Fresh Thinking podcast, Senzeni Mandava Matondi is joined by Julian Aldridge and Laurie Hassall to explore the commodities that consistently appear on critical minerals lists around the world and the factors driving their importance. The discussion examines key battery minerals including lithium, cobalt, nickel, manganese and graphite, alongside rare earth elements, tungsten and aluminium. The team also explores how geopolitical considerations, supply chain security and processing...
info_outlineFresh Thinking by Snowden Optiro
What makes a mineral critical? What makes it strategic? And why are governments around the world racing to secure supplies of certain commodities? In this episode of Fresh Thinking by Snowden Optiro , Senzeni Mandava Matondi, Julian Aldridge, and Laurie Hassell explore the growing importance of critical and strategic minerals in today's economy. From copper, lithium and rare earth elements to gallium and germanium minerals, the discussion examines how countries define these resources and why those definitions matter. The team explores: The difference between critical and strategic minerals...
info_outlineFresh Thinking by Snowden Optiro
Will AI replace geologists? In this episode of Fresh Thinking by , we explore how artificial intelligence is being used across the mining industry, from exploration targeting and core logging to resource estimation and technical reporting. Our team discusses the opportunities AI presents, its current limitations, and why geological expertise, judgement and accountability remain critical in mining workflows.
info_outlineFresh Thinking by Snowden Optiro
What are the key differences between private mining companies and publicly listed mining companies when it comes to project development, risk, and reporting? In this episode of 's Fresh Thinking podcast, Senior Consultant, Laurie Hassall speaks with Managing Consultant, Julian Aldridge, about how private companies approach mining project economics compared with companies reporting under the CRIRSCO framework. The discussion covers: CRIRSCO Code compliance and disclosure requirements Investor protection and technical defensibility Resource and reserve classification Mining project economics...
info_outlineFresh Thinking by Snowden Optiro
In this episode of Fresh Thinking by Snowden Optiro, Ian Glacken and Susan Havlin explore localised conditional simulation (LCS) and its role in modern resource estimation workflows. Building on earlier discussions around estimation methods, recoverable resources, and conditional simulation, this episode brings everything together to explain how LCS converts multiple simulation realisations into a single, practical block model—while preserving realistic grade variability. Key topics include: What localized conditional simulation is and why it matters How localization transforms...
info_outlineFresh Thinking by Snowden Optiro
In this episode of Fresh Thinking by , Susan Havlin and Ian Glacken explore the fundamentals of conditional simulation and its role in modelling grade variability and uncertainty in mineral resource estimation. They discuss how simulation differs from traditional methods like kriging and uniform conditioning, why capturing uncertainty is critical for modern mining decisions, and how multiple realizations can provide valuable insight into upside and downside risk. The conversation also touches on practical challenges, including reporting to the market and integrating uncertainty into mine...
info_outlineFresh Thinking by Snowden Optiro
In this episode of Fresh Thinking by @SnowdenOptiro, Susan Havlin speaks with Tarrant Elkington about the mine planning implications of simulation. The conversation explores: Why resource models are fundamental to mine planning and project evaluation The impact of grade variability and smoothing on mine planning outcomes How LCS compares with conventional estimation methods such as Ordinary Kriging The operational implications of dilution, ore loss and mining selectivity How simulation-based approaches can support risk-aware and stochastic mine planning Why LCS can integrate into existing...
info_outlineIan Glacken (Executive Consultant geology) and Dr Gregory Zhang (Senior Consultant Geology) explore how machine learning and convolutional neural networks (CNNs) can be used to bridge the gap between grade control data and resource estimation, and why treating resource models as static can hold operations back.
Key discussion points:
⏱ 00:00 Introduction and context for using CNNs with grade control data
⏱ 00:55 Why machine learning must be an ongoing, iterative process
⏱ 03:18 Handling multivariate data and complex geological relationships
⏱ 04:00 Practical considerations: data quality, alignment, and validation
⏱ 06:02 Trust, interpretability, and keeping geologists in the loop
⏱ 07:19 Using lithology and categorical data in CNN models
⏱ 08:58 How an operation can get started with these techniques
⏱ 11:58 Final thoughts on machine learning as a decision-support tool
If you enjoyed this episode, please Subscribe for more mining-focused technical discussions across the mine value chain.
If you would like to contact Ian or Gregory: contact@snowdenoptiro.com
Listen on the go: Fresh Thinking by Snowden Optiro is rapidly becoming the best mining podcast globally, and is available on all major podcast platforms including the video format on the Snowden Optiro YouTube channel: https://www.youtube.com/playlist?list=PLZm0zjSNmpo27fX_tfI79Yzhxy3VXjvMt
👍 Like, comment, and subscribe for more technical mining insights from our global consulting team.
Snowden Optiro:
Snowden Optiro is a resources consulting and advisory group that provides independent advice, consulting and training to mining and exploration companies, their advisors and investors.
We help mine developers to advance their projects, mining companies to improve their operations and their professionals, and investors to de-risk their investments by the provision of quality advice, training and software in the field of Mineral Resources and Mineral/Ore Reserves.
Explore more: https://snowdenoptiro.com/