An integrated 3D simulation and MCDM for estimation of ore deposits

Abstract

The success of a mining project depends largely on accurate design and production planning as well as the optimization of mines, which involves a precise simulation and estimation of ore reserves. In this research, given the great importance of simulating and estimating reserves, a new hybrid approach is presented for 3-D simulations based on kriging and multi-criteria decision-making (MCDM) with a focus on determining the most accurate volumetric model of ore deposits. To illustrate this approach, the Gharehaghaj titanium ore deposit was considered as a case study. This ore deposit has different challenges, such as the lack of a regular exploration drilling network within the ore extension area. A 3-D crude model of the titanium ore deposit was first prepared based on exploration data and variograms. Then, the parameters affecting the shape and in-depth extension of the reserve were determined with regard to volume, amount, and distribution of exploration data. Accordingly, seven volumetric sub-models were developed with different characteristics that most closely corresponded with the main volumetric model derived from the original crude model. Finally, using the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) method, the best volumetric model with the greatest fitness compared with the actual ore deposit was determined, which included 92 million t of ore at a grade of 5.64%. The reserve determined from this model was the highest compliance to the normal curve based on grade distribution. It was also concluded that the grade results obtained by the approach are 87% in accordance with the real grade obtained from the on-site sampling and assaying.