Saliba and Dimitrakopoulos 2020 Maximize DCF, penalize deviations from production targets, minimize tailings and processing rehandling cost Longterm SMIP Metaheuristics (simulated annealing, particle swarm optimization) Levinson and Dimitrakopoulos 2020 Maximize NPV, penalize deviations from waste and stockpile capacity, penalize deviations from production targets Longterm SMIP Metaheuristics (multineighbourhood simulated annealing with adaptive neighbourhood search) Levinson and Dimitrakopoulos 2024 Maximize DCF, minimize reclamation costs, minimize deviations from waste blend target, penalize deviations from production targets, encourage progressive reclamation Longterm SMINLP CPLEX, metaheuristics with ML (simulated annealing with contextual bandits) Despite significant contributions of the aforementioned studies, further development of waste and AMD techniques is required. One of the disadvantages of the presented methodologies is their simplification of the waste dump/TSF evolution over time. Apart from the inherently dynamic nature of waste material behaviour, climate change introduces additional risk by changing rainfall patterns and intensity, among other environmental factors. Some novel models aim to address this issue. For example, Vaziri et al. (2025) present a SMIP waste scheduling formulation that incorporates simulated AMD evolution within the waste pile, along with multiple rainfall and waste dump permeability scenarios. However, the study in question does not integrate geological uncertainty into the objective function, nor addresses the extraction sequence, thus failing to capitalize on the synergies with upstream processes. A natural extension of the models presented in this work is a more holistic approach, integrating the production schedule with proactive waste management and AMD mitigation efforts.
RkJQdWJsaXNoZXIy MTM0Mzk2