Abstract

This work presents a novel approach to solve the mining cut definition problem with geometallurgical interactions in short-term mine planning for open-pit operations. Mining cut definition deals with the aggregation of blocks into clusters which are extracted and processed as a single unit. The aggregation must fulfill operational considerations given by the loading equipment selectivity and should maximize the objectives given by the mine operation. The proposed approach utilizes mixed integer programming and a model inspired by column generation that, contrary to previous works, has its decision variables defined directly on the set of all feasible cuts. The advantage of this is that the model does not require linear approximations of the geometallurgical behavior of the cuts based on the blocks it contains and, therefore, can utilize any nonlinear function. An industry-sized dataset is used to show that the model can be solved in reasonable time. Also, the results show that nonlinear recovery functions influence the destination policy and expected profit. Therefore, following the traditional free selection policy (based on cutoff grade) is not the best strategy when geometallurgical interactions are considered.