Optimizing block caving drawpoints over multiple geostatistical models

ABSTRACT

Large-scale mining methods are commonplace in the mining industry, and caving has become the preferred underground mining technique. The use of block caving has increased worldwide and is expected to grow because of high production rates, low mining costs, and low waste production. One key engineering consideration in block caving is the drawpoint spacing layout, which must be selected prior to production based on widely spaced exploration data; uncertainty is highest during this stage of mining; and selecting an appropriate extraction level and layout is important for economic and safety considerations. In the proposed methodology, uncertainty in the grade block model is assessed with sequential Gaussian simulation and an optimisation procedure is developed to maximise net present value over many realisations while being constrained by mining considerations. This novel approach generates one layout that is optimal over all realisations rather than consider optimisation on a single kriged model which is standard practice; in this way risk can be assessed and mitigated by selecting the layout that maximises profit but also minimises the variance of profit over all realisations. The methodology is demonstrated using a case study, and the proposed technique outperforms traditional optimisation based on a kriged model.

KEYWORDS:

• Block caving
• realisation
• simulation
• uncertainty
• sequential Gaussian simulation
• drawpoint layout

Disclosure statement

No potential conflict of interest was reported by the authors.