Solving a large SIP model for production scheduling at a gold mine with multiple processing streams and uncertain geology

Abstract

In this paper, an efficient heuristic solution approach is applied and tested to the stochastic mine production scheduling of a relatively large gold deposit containing about 120 thousand blocks and considering a set of 15 geological scenarios generated stochastically. The case study addresses multiple processing streams and a ‘grade’ stockpile, which adds flexibility to the specific operation by advancing the processing of highly valuable material. The solution approach first generates an initial feasible solution by sequentially solving the stochastic open-pit mine production scheduling (OPMPS) period by period, and then a network-flow algorithm is used to sequentially search for improvements. In this network graph, the nodes identify candidate blocks that might have their extraction postponed or advanced, aiming for new schedules with higher value and lower risks. The results show that production schedules with low deviations from production expectations can be generated in a reasonable time for an actual mining environment.