Ore Grade Estimation Using Kriging and Machine Learning

How well can the grade of a metal be estimated between the sampled points - with classical geostatistics or with machine learning? This project estimates cobalt, nickel and five other metals on the real Swiss Jura geochemical data with a variogram and ordinary kriging, inverse-distance weighting, Gaussian-process regression, Random Forest and gradient boosting with rock type and land use, and regression kriging, validates every method on independent samples and with spatial cross-validation, and turns the estimates into probability maps above a cut-off grade, a block model and grade-tonnage curves. A web application estimates new sample sets. R gstat, GSLIB (GeostatsPy) and GNU Octave cross-check the results, and a complete MATLAB version is included.

Category: Mining Engineering

Mining Engineering Project MN01 from ProTechSC Academy, with graphical abstract, video demonstration and project guidance.