• The varied roles of mineral resource estimation - does the end use affect the estimate?
• Who are the end-users and what do they expect from the estimate?
• What are valid and invalid uses for resource estimates?
• What end-use considerations should influence the estimation approach?

• The practical application of reconciliation data in resource estimation.
• Understanding Met Accounting and why production data is not always as it seems.
• What does ‘good reconciliation’ prove?
• Can you have ‘good reconciliation’ and a poor estimate?
• Is reconciling to the resource valid?
• Does reconciliation add value commensurate with the effort? If not, why not and what should be done differently?

• The assumptions and limitations of available estimation methods.

Finding out the reasons for a block model that is failing (Where is the problem, how do you determine an error in the resource estimate when there are multiple post-estimation factors in play).

• Sampling issues, trusting the data
• Assessing expected error (via simulation)
• Is the problem coming from sampling, from the geological interpretation or from the estimation?
• Understanding the limits of models: how do you know there is a flaw in the resource? How do you pinpoint the error in complex interactive systems?

• Commercial software solutions can help break paradigms/inertia and develop new techniques. When mining is compared to the oil industry, the development and advance gap is clear.
• If software-specific techniques/methodologies are not democratised, the average user will not be able to access such innovation. Software innovation requires plenty flexibility and interoperability between different platforms, workflows, and data types, which is not always present in commercial software.
• What is the next frontier for geology software? How to implement ML in our workflows? Integration with programming languages? Interoperability between commercial software solutions? Automation of estimation and validation workflows?

Block histogram shape, block variability and grade tonnage curves – everything matters and needs to be examined holistically. How is this acheived or validated?

• Domain definition
• Composite size
• Block size
• Variogram model
• Search neighbourhood
• Number of samples selected
• Number of passes
• Thresholding
• Estimation method

• Things changed between evaluation and execution (e.g. market, prices, product).
• Learning from mistakes is a key pathway to improvement. How do we do this without making it personal?

• Does the current classification system need to be challenged?
• Differences between reporting jurisdictions.

• How do you know your estimate is the central case?
• Communicating confidence to end-users
• Going beyond JORC, classification for internal usage
• Confidence that the block means are accurate?
• That the global mean is accurate?
• That the range of variability is well or poorly understood?
• Does it make sense with the mining method?
• Is the problem a general lack of data or related to geometry or specific types of data?

• What’s on the horizon?
• Artificial intelligence and machine learning for domaining and estimation:
  • State of the industry
  • Domaining
  • The non-spatial estimation problem
  • Wireframe-free estimation

• What is ‘best practice’?
• Is best practice the same for every resource and every operation?
• Dealing with exceptional cases (and why are they exceptional?)
• Is ‘best practice’ a limiting paradigm?
• Does it exist?
• Who and how?
• Competence
• Extra guidance, governing bodies, etc.

.

• Why do we estimate mineral resources?
• Is a resource model a scientific model?
• Is it a good idea to assign ‘value’ to a resource estimate?
• The limitations and there implications of estimation (e.g. kriging)

Have a thought-provoking idea on mineral resource estimation? We want to hear it!

• Submit an abstract outlining a panel discussion topic and why you think it would worth discussing.
• Or, submit a question that you’ve always wanted answering and we will consider including into one of the panel discussions.