Hierarchical Clustering Algorithm for Block Aggregation in Open Pit Mines

Open pit mine plans define the complex strategy of displacement of ore and waste over the mine life. Various mixed integer linear programming (MILP) formulations have been used for production scheduling of open pit mines. The main problem with these models is the inability to solve real size mining problems. We present a mixed integer linear programming formulation for the open pit long-term production scheduling problem. To reduce the number of binary integer variables in the formulation, we aggregate blocks into larger units referred to as mining-cuts. We present a semi-supervised clustering algorithm which aggregates smaller blocks based on rock types and grade ranges into mining-cuts without introducing an error in the grade estimated values. The cluster-based aggregation algorithm is fully integrated into an MILP mine production scheduling prototype software. We have implemented, verified, and validated our models. The efficiency of the new platform is evaluated through iron ore mine case study.