Uniform Conditioning - LUC Model Reports

To access this dialog:

Activate the Estimate ribbon and click Uniform Conditioning. Select the SMU Block Model Reports panel on the left.

Resource model grades are traditionally estimated into blocks whose size is representative of the drillhole/sample spacing. However, recoverable reserves depend in part on the variability of grade at a Selective Mining Unit (SMU) size. The main idea is based on assigning to each block (“SMU”) a grade such that the distribution of block grades in the panel finds back the local grade tonnage curve estimated by Uniform Conditioning (UC).

Once a panel model has been generated (on the previous screen) as a result of Uniform Conditioning, you can post-process the model to estimate resources at a more granular level - the SMU using a process known as Localised Uniform Conditioning or LUC. This is a post-processing phase to UC which assigns within each Panel a grade value to each SMU (Selective Mining Unit) whilst preserving the local grade tonnage curve of SMU’s estimated by the preceding process. The goal of Localized Uniform Conditioning is to provide a more 'intuitive' presentation of uniform conditioning result, for example this could be a higher dispersion of grade obtained in LUC than direct block kriging.

The overall goal is provide a more 'intuitive' presentation of the results of Uniform Conditioning.

Find out more about Localized Uniform Conditioning...

Field Details:

The SMU Block Model Reports screen contains the following fields:

Create Prototype from SMU block sizes: if the selective mining unit dimensions are known, select this option to enter them manually, using the following field:

SMU Size: specify the dimensions of the mining units within each panel that will ultimately receive a locally (and uniformly) conditioned grade value.

Create Prototype from SMU model: to extract SMU dimensions from an existing SMU model, select this option and browse for the file. If you select this option, you can also use estimates within the SMU model during localized uniform conditioning if you wish.

Search volume parameter file: browse for a search volume parameter file or manually enter the name of a file residing in the current project directory.

Reference Number: the selected Search Volume Parameter file will contain an SREFNUM data column (effectively, the identifier for a search volume within the data file - there can be more than one). This drop-down list will display all available search volume definition IDs that relate to the selected parameter file. If only one value is contained, it will automatically be selected.

LUC block model: the automatically-generated name of the SMU model (the locally-conditioned model) that will be created on disk when Create SMU block model is selected.

Create LUC block model: perform localized uniform conditioning on the previous panel model and generate an SMU model on disk.

LUC block model summary: generate a scatter plot (as a new plot sheet) showing the distribution of locally-conditioned grades throughout the SMU model. Another sheet is also generated, showing 3 histograms:

The frequency of Kriged values (above a zero cut-off) as represented by the Panel model
The frequency of grades resulting from ordinary Kriging as stored in the SMU model - this is provided for comparison against the locally-conditioned output (below). Note that, as the locally conditioned output is bound by the kriged value at the panel support level, the mean value across the range should be very close or identical between top and bottom graphs.
The frequency of grades in the locally-conditioned model

View LUC block model: load the generated SMU model into memory and display it in Studio.