Effects of mineral price models on mineral project evaluation

The authors present a new mineral price model, compare it with time series and naive models, and analyze the effects of the forecasting models on mineral project evaluation. Mineral commodity prices are volatile, which means that the results of evaluation tools that do not treat the stochasticity of metal prices rigorously may be misleading. But, in mine valuation, commodity price forecasts are required to assess the economic viability of a project. In this study, a new mineral forecasting method, called the MNDRVG-MFNN-RM model, which incorporates randomness, neural networks and regression models, is introduced. The MNDRVG-MFNN-RM model, a naive method and time series model were used to forecast gold prices for two successive years for the evaluation of a proposed open-pit mine. The MNDRVG-MFNN-RM model yielded the best results among the three methods. It produced the true optimum pit limits and an optimum pit value slightly less than the true optimum value. The main novelty of the methodology is the simulation and rigorous analysis of the randomness property associated with mineral prices to reduce the estimation and forecast errors, an important contribution to mineral venture evaluation, mine planning and design.