Abstract
A review of 175 technical reports (NI 43-101, SEC, and JORC) published between 2020-2023 examines current practices in mineral resource classification and estimation methodologies in the mining industry, highlighting trends and differences between reporting jurisdictions and deposit types. As expected, ordinary kriging (OK) is the predominant estimation technique, although inverse distance (IDW) is still widely used. Capping after compositing is the most common strategy for handling extreme values. Drill hole spacing (DHS) and the use of search neighborhoods (SN) are the most common classification methods, while statistical metrics like kriging variance, slope of regression, or confidence intervals applied to simulated realizations are rarely used. JORC reports often rely on DHS for classification, whereas NI 43-101 reports contain more diversity. The choice of classification method is influenced by deposit type, variogram range, nugget effect, commodity type, drilling pattern, and reporting standards. We also propose a data-driven approach using a decision tree (DT) classifier to prioritise decision-making factors for classification and estimation methods, achieving accuracies of 81% for estimation methods, 83% for classification methods, and 88% for capping strategies. This model aids practitioners by providing insights into industry practices for similar deposits but is not intended to provide an ‘optimal’ strategy for a given deposit nor does it attempt to replace expert judgment. This work highlights industry practices rather than recommending or judging the best workflows for classification, estimation and capping.