A Data-Driven Approach for the Simultaneous Stochastic Optimization of Mining Complexes

Abstract

The simultaneous stochastic optimization of mining complexes (SSOMC) is a large-scale stochastic combinatorial optimization problem that simultaneously manages the extraction of materials from multiple mines and their processing using interconnected facilities to generate a set of finished products while taking into account geological uncertainty to manage the associated risk. To address the need for self-managed solution approaches that are able to tackle large-scale instances without resorting to aggregation, this work proposes a data-driven framework for heuristic scheduling in a fully self-managed hyper-heuristic to solve the SSOMC. The proposed hyper-heuristic selects the heuristic (perturbation) to be applied in a self-adaptive manner using state-of-the-art reinforcement learning methods to efficiently explore which local search is best suited for a particular search point. Experiments on real-world mining complexes show a significant reduction in the computational time by 30–45%.