MIK (Multiple Indicator Kriging)

The workflow to produce an MIK estimate consists of three broad steps:

• MIK Pre-processing: From a chosen input variable, you create a macro variable containing the indicators at selected cutoffs; the cutoff grades will be the indices of the macro variable.

• Estimation of the indicators: Isatis.neo gives the user complete freedom in how to estimate the indicators. A typical workflow would be to model the variogram of each indicator individually in the Exploratory Data Analysis tool, and then estimate by Kriging.

  Note: In the EDA, check that all the indicator variograms are using the same stationarity option (simple or ordinary kriging). This is a requirement for the Multiple kriging.

  Using the dedicated Multiple kriging special option in the Kriging page, the estimates should be stored in a grid as one macro variable. The kriged indicators will be the indices of the macro variable.

  To ensure that the kriged indicators follow a standard naming convention if not using the Multiple kriging special option, such a consistent convention can be achieved by using "%var" in the pattern for the output variables.

  Note: Be careful, when using Multiple kriging the same neighborhood is applied to all the indicators.

• MIK Post-processing: The estimates of the indicators are used to derive quantities such as grade-tonnage values (quantity of metal Q, tonnage T, and mean grade M above chosen cutoffs), or an e-type estimate.