Mineral reserve classification is an industry requirement complying with reporting codes such as CRIRSCO. However, deterministic methods rely on geometric criteria, such as drill hole spacing or kriging estimation variances. These metrics quantify data configuration and spatial position rather than grade uncertainty within the deposit. Addressing this limitation requires an approach that integrates geological uncertainty into reserve classification. Studies demonstrate that stochastic mine planning maximizes Net Present Value (NPV) by managing geological risk, expanding optimal pit limits, increasing total tonnages, and extending the life-of-mine (LOM). By evaluating grade uncertainty to define the production schedule, stochastic mine planning generates reserves that can be classified. To implement this approach, the following framework is applied to a gold deposit. First, high-order simulations capture the complex geometry and high-grade connectivity of the deposit, overcoming limitations of conventional methods. Subsequently, stochastic optimization is applied to these models to generate the production schedule. This integration ensures the mining schedule is informed by geological risk. Therefore, classification is evaluated based on grade uncertainty rather than spatial parameters. Scheduled volumes are classified as Proved and Probable reserves when the probability of the grade falling within a ±15% interval around the predicted mean exceeds 90% for quarterly and annual production periods, respectively. By deriving classification from the production schedule, the framework ensures declared reserves are coherent volumes with operational feasibility. Finally, reserves defined by this framework are compared against those defined using geometric criteria. This comparison demonstrates the operational value of deriving reserve classification from stochastic schedules rather than deterministic constraints.
Risk Uncertainty and Opportunity Identification
Integrating High-Order Simulation into Mineral Reserve Classification via Stochastic Mine Planning
2 views
1 Downloads