CFSG 2022 – Module 5 – Domaining| Training course
Get introduced to a powerful machine-learning-based technique for geological domaining.
The Specialized Training Cycle in Geostatistics, known as CFSG, is a high-level training program in mining geostatistics delivered by the Geostatistics Team from MINES ParisTech and Geovariances. Its objective is to give you a comprehensive and deep knowledge of geostatistics for mineral resource estimation so that, when you return to work, you will be able to build the block models that your company needs for confident mine planning. By attending the program, you learn the theory behind the techniques presented to you and practice them through numerous exercises and a real-life project.
The 2022 edition of CFSG is delivered online in 6 modules over 10 weeks starting from February 2022. This module is the fifth one of the series. At the end of it, you will be able to derive domains from surfaces and drill hole information.
- Half of the training program is devoted to methodological presentations, the other half to practical exercises to deepen the understanding of concepts.
– The methodological courses are given by MINES ParisTech professors and are scheduled in order to match the time zone of the audience. For more convenience, these courses are recorded and made available to participants.
– The practical sessions will be driven by Geovariances’ consultants, either from our French office (for those whose time zone is compatible with the French one) or from our Australian and Brazilian offices for the other participants.
- A typical training week would then be:
– Monday, Tuesday, and Wednesday: theoretical course (on a half-day) and hands-on practice with Isatis.neo Mining Edition (on the half-day following the theoretical course).
– Thursday or Friday: Homework using Isatis.neo Mining Edition with compulsory rendering at the end of the day.
– Friday: Live correction and comments from the teaching team. Validation of prior learning.
- CFSG 2022 is a certification training. The knowledge acquired in each module is validated through an examination. At the end of each module, you will get a training certificate with a distinction giving official recognition to the full completion of the module.
- Course material and a temporary software license are provided.
- A minimum number of 5 participants is required for a module to actually take place.
Who should attend
CFSG training program is meant for mining geologists and engineers who are willing to achieve a high level in geostatistics.
A 3D domain is modeled by its boundaries, either considered as a pair of 2D surfaces or closed manifold in 3D.
- Data classification: statistical multivariate geographical classification using Geostatistical Hierarchical Clustering (GHC)
– Classification into rock types from multivariate attributes
– Integrating spatial dependency through an adjacency graph
– Updating the classification with Support Vector Machine learning methods (SVM)
- The multi-layer approach
– Joint modeling of top and bottom 2D surfaces
– Extension to the multi-layer case
– Dealing with vertical or deviated drill hole data
– Missing intercepts: how to deal with pinch-outs or drill holes not reaching the deepest surfaces
– What to do in presence of faults
- The Potential method
– Data: intercepts of drill holes with boundaries, normal and tangent to the surface
– The implicit model
– Estimating the latent potential using cokriging
– Dealing with several surfaces: conformal layers, layers with pinch-outs or faulted
– Uncertainty quantification (probability to belong to the domain) with simulations of the potential field
Several exercises to learn how to run sample clustering for automatic domaining and domain modeling with Isatis.neo Mining Edition.
Participants are requested to have attended CFSG 2022 first module.
We recall that CFSG 2022 is made of 6 modules:
– Module 1: Geostatistics Fundamentals
– Module 2: Advanced in-situ resources
– Module 3: Recoverable resources
– Module 4: Facies simulations
– Module 5: Domaining
– Module 6: Handling big data sets
Please note that Module 1 is mandatory. Modules 2 to 6 can be attended à la carte, according to your needs, independently from each other.
We offer a 25% discount on the purchase of all 6 modules.