Updating Mining Resources with Uncertain Data

Abstract

In mining operations, the time delay between grade estimations and decision-making based on those estimations can be substantial. This may lead to the scheduling of stopes mining that is based on information which is seriously out of date and, consequently, results in a substantial mined resources and reserves bias. To mitigate this gap between the grade estimation of an orebody and its exploitation, this paper proposes a method to quickly update resources and reserves that are integrated into the concept of real-time mining. The current standard for grade data collection in underground mines relies on a conventional chemical lab analysis of sparse drill hole or chip/face samples. The proposed methodology for the continuous and swift updating of mine resources and reserves requires a constant and rapid stream of measurements at the stopes. Consequently, this work proposes using portable X-ray fluorescence (XRF) devices to carry out the fast and abundant monitoring of ore grades. However, these fast data are highly uncertain; hence, the objective of this proposed method is to use the total data measurements and integrate their uncertainty into the resources modeling. The first step in the proposed methodology consists of creating a joint distribution function between “uncertain” XRF and the corresponding “hard” measurements based on empirical historical data. Then, the uncertainty of the XRF measurements is derived from these joint distributions through the conditional distribution of the real values applied to the known XRF measurement. The second step involves updating the reserves by integrating these uncertain XRF data, which are quantified by conditional distributions, into the grade characterization models. To achieve this, a stochastic simulation with point distributions is applied. An actual case study of a copper sulfide deposit illustrates the proposed methodology.