2087 projects, the correlation between variables, for example, grades, recoveries, costs, and their concurrent variability can significantly alter the Net Present Value (NPV) distribution compared to what a single reading would suggest. Therefore, a framework that integrates probability distributions and correlations is required to derive comparable operational and financial risk metrics. Conventional industry practice typically relies on univariate sensitivity analysis and deterministic scenarios (optimistic, base case, and pessimistic). While these tools are useful for a first approximation, they are insufficient to represent the stochastic and simultaneous nature of the main drivers of value. In gold projects, the correlation between variables for example, grades, recoveries, costs, and their concurrent variability can significantly alter the Net Present Value (NPV) distribution compared to what a single reading would suggest. Therefore, a framework that integrates probability distributions and correlations is required to derive comparable operational and financial risk metrics. Methodologically, the study fits each input variable with an appropriate empirical or theoretical distribution (e.g., Normal, Lognormal, or Triangular) using parameters obtained from historical series and internal projections and incorporates correlation structures consistent with the business logic. This probabilistic construction avoids deterministic extrapolations and improves risk traceability throughout the chain of assumptions, offering a comprehensive view of the outcome space and the residual uncertainty of the base case. The work makes a twofold contribution. First, it operationalizes financial risk assessment in gold mining projects using a reproducible framework that can be implemented in widely used tools, integrating random number generation, sensitivity analysis, and the reading of economic percentiles. Second, it shifts risk management from a qualitative exercise to a quantitative and predictive platform that strengthens investment decision-making, planning, and hedging, aligned with long-term financial sustainability criteria in the gold industry. Taken together, the proposed approach allows management to accurately estimate the probability of achieving NPV/IRR targets and prioritize value drivers under uncertainty. Figure 1 – Financial risk analysis flow with Monte Carlo Simulation. 2. SCOPE This work falls within the thematic area of AI and data-driven decision-making, contributing a quantitative and predictive approach to the financial evaluation of gold mining projects. The paper's scope encompasses the development, implementation, and validation of a Monte Carlo simulation model that transforms historical data and operational parameters into probabilistic metrics of value, thereby improving the quality and robustness of investment decisions under conditions of high uncertainty. The study focuses on modeling key variables in the mining business gold price, metallurgical recovery, and mining and processing costs from which empirically or theoretically adjusted probability distributions are derived. These distributions, combined with correlation structures representative of operational reality, feed into an iterative simulation model that generates a Net Present Value (NPV) distribution. This allows the analysis to move away from traditional deterministic scenarios and adopt a datadriven approach that reflects the stochastic nature of the business.
RkJQdWJsaXNoZXIy MTM0Mzk2