Geostatistical simulation of optimum mining elevations for nickel laterite deposits
CIM Bulletin, Vol. 1, No. 6, 2006
J.A. McLennan, J.M. Ortiz, C.V. Deutsch
The file is a zipped PDF document.
Uncertainty is an inherent aspect in mine planning. Geological architecture and heterogeneity between drillhole samples is impossible to predict exactly. Estimation provides a single biased prediction of geology; however, conditional simulation allows alternative geological realizations or truths with the right heterogeneity to be created. Each realization honours the conditioning data value and the realization fluctuations in the space between data is a measure of geological uncertainty. This geological uncertainty can be transferred into uncertainty in mine planning parameters, which can then be used to make improved and confident production decisions. This principle is applied to a small nickel laterite mining scenario.
Nickel laterite deposits are typically formed from tropically weathered mafic to ultramafic complexes and constitute a major reserve of nickel worldwide. Typically, small dozer blade units are used to mine the loose nickel laterite soil material. Selectivity is excellent. Selective mining unit (SMU) digging elevations and the dilution and lost ore costs associated to these digging elevations are important mine planning parameters. A conditional simulation approach is used to calculate optimum representative dozer region (RDR) digging elevations.
There are two main phases to the methodology: (1) simulation of the ore-waste contact surface conditional to orewaste contacts, and (2) post-processing these ore-waste contact surface realizations through an optimization transfer function for determination of optimum mining elevations and minimum dilution and lost ore costs at each SMU.
Using ore-waste contact elevations interpreted from exploratory drillholes (based on a per cent Ni/m cutoff), geostatistical simulation is used to create multiple ore-waste contact surface realizations. The ore-waste contact surface is simulated at a resolution higher than the RDR. At one RDR location, the ore-waste contact surface is used to calculate dilution and lost ore for a range of possible digging elevations (see figure). The optimum digging elevation is the elevation that simultaneously minimizes the lost ore and dilution costs. This is repeated for all ore-waste surface realizations within the RDR from which the expected optimum digging elevation is found. This is repeated for all RDRs in the mine plan so that the resulting optimum mining elevation and the associated minimum dilution and lost ore costs can be used for mine planning purposes. The minimum RDR dilution is the cost of mining the volume of waste located below the expected orewaste contact and above the optimum mining elevation. The minimum RDR lost ore is the cost of not mining the volume of ore above the expected ore-waste contact and below the optimum mining elevation.
Hand contouring for digging elevations is sub-optimal. Notional hand contoured digging elevations or retrospective dilution and lost ore costs are hard to trust for important mine planning decisions. Within the conditional simulation framework presented, forecasts of dilution and lost ore costs at each RDR location are simultaneously minimized over multiple possible true ore-waste surfaces. The associated digging elevations are optimum for production planning and mine economics.
The methodology is straightforward and can be automated on virtually any moderately powered personal computer. Additional grade control information beyond the information available at the time of the simulation study is easy to incorporate in order to update the optimum digging elevations and minimum dilution and lost ore forecasts.