The application of median indicator kriging and neural network in modeling mixed population in an iron ore deposit

This paper presents the results of a study comparing median indicator kriging and an artificial neural network in the estimation of iron grades in the Jalal-Abad Zarrand iron ore deposit located in the southern Iran. The data used in this study is from 2017 composite samples with 2 m length from 32 exploration boreholes. The iron grade data is sparse, irregularly spaced and has mixed distribution, which can be problematic for the stationarity assumptions of the widely used ordinary kriging estimation method. The two estimation techniques applied in this study make no assumptions about the distribution of the sample data, and accommodate moderately mixed sample populations.