Moving from an analytical model to a decision support system (DSS). This means that the result of the simulation is not a PDF → but a decision flow embedded in the management: • Dynamic CAPEX/OPEX policies based on risk. • Activation of hedging strategies according to VaR. • Mine planning adjustments based on critical drivers. • This aligns the work with the global trend towards smart mining. The latest literature integrates historical data, updated data, risk dashboards, and distribution fitting with recent information. Now, the incorporation of machine learning allows us to detect patterns in metal prices, costs, and all critical variables used in the analysis. It also enables us to detect nonlinear correlations and update the distribution in real time. On the other hand, it "generates a risk management policy" based on simulation, which includes NPV, IRR and risk quartile indicators, thus becoming a management model and not just a simple analysis. CONTRIBUTION OF THE RESEARCH This paper presents a methodological and applied innovation in the financial evaluation of gold mining projects by integrating a Monte Carlo Simulation framework with advanced risk metrics (probabilistic NPV, VaR, contribution to variance), transforming a traditionally deterministic process into a data-driven decision-making system. It´s main contribution is articulated in four dimensions: 1. Transition from deterministic approaches to a robust probabilistic model. - This work demonstrates that conventional methods, univariate sensitivity and optimistic, baseline and pessimistic scenarios, underestimate uncertainty and fail to capture the key interdependence of the mining business, as recognized in the specialized literature on project evaluation and risk analysis in mining. 2. Explicit incorporation of correlations between critical variables. - Many previous studies model input variables as independent, leading to biased risk estimates. Recent Monte Carlo research in mining and infrastructure highlights the need to incorporate correlation matrices to represent real dependences. This paper contributes to bridging this gap by modeling correlations between price, recovery, and costs, improving the accuracy of NPV forecasting and executive risk perception.
RkJQdWJsaXNoZXIy MTM0Mzk2