Mine operations are supported by a short-term production schedule, which defines where and when mining activities are performed. However, deviations can be observed in this short-term production schedule because of several sources of uncertainty and their inherent complexity. Therefore, schedules that are more likely to be reproduced in reality should be generated so that they will have a high adherence when executed. Unfortunately, prior estimation of the schedule adherence is difficult. To overcome this problem, we propose a generic simulation–optimization framework to generate short-term production schedules for improving the schedule adherence using an iterative approach. In each iteration of this framework, a short-term schedule is generated using a mixed-integer linear programming model that is simulated later using a discrete-event simulation model. As a case study, we apply this approach to a real Bench and Fill mine, wherein we measure the discrepancies among the level of movement of material with respect to the schedule obtained from the optimization model and the average of the simulated schedule using the mine schedule material’s adherence index. The values of this index decreased with the iterations, from 13.1% in the first iteration to 4.8% in the last iteration. This improvement is explained because the effects of the operational uncertainty within the optimization model can be considered by integrating the simulation. As a conclusion, the proposed framework increases the adherence of the short-term schedules generated over iterations. Moreover, these increases in the adherence of schedules are not obtained at the expense of the Net Present Value.

中文翻译：

---

短期地下矿山生产调度的模拟优化框架

矿山运营受到短期生产计划的支持，该计划定义了在何处以及何时进行矿山活动。但是，由于不确定性的多种来源及其固有的复杂性，在此短期生产计划中可能会出现偏差。因此，应该生成更可能在现实中重现的日程表，以使它们在执行时具有很高的依从性。不幸的是，很难预先估计时间表的遵守情况。为了克服这个问题，我们提出了一个通用的仿真-优化框架来生成短期生产计划，以使用迭代方法来提高计划的遵守率。在此框架的每次迭代中，使用混合整数线性规划模型生成短期计划，随后使用离散事件模拟模型对其进行模拟。作为案例研究，我们将此方法应用于实际的Bench and Fill矿山，其中我们测量相对于从优化模型获得的进度表的物料移动水平与使用矿井进度表模拟的平均值之间的差异。材料的附着指数。该索引的值随迭代而降低，从第一次迭代的13.1％降低到最后一次迭代的4.8％。之所以说明此改进，是因为可以通过集成仿真来考虑优化模型中操作不确定性的影响。结论是，提出的框架增加了迭代产生的短期计划的遵守性。