ABSTRACT
A heuristic algorithm, referred to as the “Maximum Value Neighbourhood” (MVN) algorithm, has been developed to optimise stope geometries of underground mines. The method uses economic model of a given deposit to search for the best combination of mining blocks that results in the maximum profit while imposing geo-technical and practical mining constraints. The algorithm has been implemented using Excel Visual Basic modules. The search algorithm is executed row by row, column by column and section by section to provide a 3D result. The MVN of each block provides an island of blocks, which grows gradually as the optimisation progresses over the deposit. The optimised stope is displayed on Excel worksheets, with the cells corresponding to the optimum blocks are identified by distinct colours and borders. A step by step explanation of how the MVN algorithm was implemented is presented with an example block model.
