A geostatistical method applied to the spatial distribution of Rock Mass Rating values in a mineralization zone

A precise understanding of the spatial distribution of rock mass properties is essential for the safe and economical design of rock structures. This paper adapts geostatistical methodologies, traditionally employed for estimating block ore grades and tonnage, to forecast rock properties crucial for structural modeling. The Rock Mass Rating (RMR) classification system, extensively utilized for evaluating rock mass quality, serves as a framework to inform excavation techniques and ensure slope stability in open-pit mining and rock support systems for tunnel construction. The study introduces a geostatistical simulation method to create three-dimensional (3D) models of rock mass quality distribution based on RMR. Geotechnical data from 37 drillholes, encompassing a total of 11,278 meters, were collected from the Miduk open pit mine in Iran. Two block models for RMR were constructed using the turning bands simulation method (TBM) with 100 realizations. The research utilized both direct and indirect approaches. In the direct method, the RMR value was considered a singular variable for simulation, whereas the indirect method involved simulating individual RMR parameters and subsequently summing them to derive the final RMR for each block. Cross-validation indicated strong consistency between the two approaches, reinforced by the 3D model of the faults and the contribution of joints, which were derived from scan-line mapping data collected from 24,160 surface stations. Although both methods yielded similar results, the block model developed via the indirect approach proved to be more comprehensive regarding geomechanical parameters and has thus been established as the final model.