Abstract
The foundation of classifying mineral reserves in open-pit mining is the definition of a mine plan. This plan starts with a pit limit encompassing various critical factors like mining, processing, metallurgical, economic, marketing, legal, environmental, infrastructure, social, and governmental aspects. Traditional approaches to ultimate pit limit, such as Lerchs and Grossmann and Pseudoflow algorithms, often fall short in adequately incorporating the time-value of money. This limitation becomes even more pronounced in mining operations constrained by time-sensitive and spatially-defined extraction limitations, such as infrastructure reallocation and operational permits, as their timing can significantly impact project feasibility. Hence, it has long been recognized that an optimal mine plan should emerge from long-term mine production scheduling that directly maximizes the net present value (NPV). In addition, traditional ultimate pit limit methods rely on estimated orebody models, which are a smooth representation of the mineral deposit, misrepresenting the spatial distribution and variability of grades, material types, and geometallurgical attributes. Therefore, they are unable to manage and quantify the risk associated with the definition of the mineral reserves.
A new software technology now enables simultaneous stochastic optimization of mining complexes (SSOMC) with multiple open-pit mines, stockpiles, waste dumps, and processing facilities. This technology uses geostatistical simulations of mineral deposits to describe geological uncertainty and variability as inputs for the optimization process. The approach enhances the exploitation of synergies between time-dependent extraction restrictions and production scheduling, while effectively managing the risks associated with various production targets.
This case study showcases the application of SSOMC in mineral reserves evaluation for a real-world gold project whose resources are situated underneath a village. The project's viability requires assessing the costs associated with relocating a portion of the village's houses and infrastructure, along with operational permit restrictions. The results revealed enhanced recoverable gold derived from geostatistical simulations. Compared to traditional methods, this approach yielded reserves that differed significantly in terms of higher recovered gold and higher ore tonnage, providing a more comprehensive consideration of reallocation costs and constraints. Finally, the approach was able to manage the risks related to the fresh-to-oxide ratio in the processing plant feed, thereby expanding the project's feasibility potential.