A drill hole spacing study (DHSS) and a value of information study (VOIS) are different. A DHSS quantifies uncertainty in tonnage, grade and metal of production volumes of relevance and relates that uncertainty to drill hole spacing to support resource classification decisions. A VOIS brings in economic considerations to recommend the drill hole spacing that maximizes the net benefit of drilling, acknowledging that (1) drilling too much will incur costs that are not recovered by improved decisions, and (2) drilling too little will lead to suboptimal decisions that could have been improved by additional drilling. This blog explains this perspective and challenges us to move towards VOIS.
The figure shows the essence of a DHSS and a VOIS. A DHSS determines the drill hole spacing that achieves a target uncertainty – the vertical dashed line on the left is where the expected uncertainty is at the target (the gray dots represent the variation we might expect from different production volumes). A VOIS determines the drill hole spacing that achieves the maximum net value considering the cost of data and the value of information – the vertical dashed line on the right is where the net value is maximized.
DHSS have become routine. RMS staff have performed many and have established important details including (1) the relevance of geology uncertainty, (2) the significance of parameter uncertainty, (3) the importance of trend modeling, (4) the criticality of defining production volumes realistically, and (5) the utility of analytical scaling of uncertainty to consider longer time periods and multiple mining faces. A carefully crafted DHSS provides accurate and precise uncertainty quantification for production volume uncertainty. A previous blog discussed the probabilistic criteria that could be relevant in different circumstances.
A DHSS involves: (1) creating reference truth models, (2) sampling those reference truth models with different data spacing scenarios, (3) resimulating with the sampled data scenarios, (4) calculating production volume uncertainty, and (5) assessing the data spacing that achieves a target measure of uncertainty. The reference truth models and their synthetic drill hole data permit: (1) study of drill configurations that do not currently exist, such as denser or rectangular spacings, and (2) the uniform application of those configurations across the area of interest, so that uncertainty of a given spacing of interest is not confounded by other local factors, as is the case with the existing variable drill configuration and associated uncertainty. The target measure of uncertainty is subjective.
A VOIS targets a measure of value. We require estimates of prices, costs and an engineering design strategy. Reasonable estimates and a simple automatic proxy for detailed engineering design / operational performance are sufficient. Then, a VOIS involves: (1) creating reference truth models, (2) sampling those reference truth models with different data spacing scenarios, (3) predicting the best possible models given the available sample data, (4) designing an optimal plan on those models, and (5) assessing the optimal plan against the original reference truth model. We plan based on what we know, but we get the inaccessible truth; therefore, the reference truth models are used for more than collecting synthetic drill hole data.
A VOIS has a measure of optimality; a DHSS has a subjective target. A DHSS is typically performed early in project valuation where classification for resource disclosure is particularly important. The importance of disclosure persists through the lifecycle of a mine but diminishes as the resource is depleted. A VOIS is typically performed during mining when cost savings and operational efficiency are paramount.
This comparison and perspective on DHSS and VOIS are standard. My claim, that may be more provocative, is that we should be undertaking more VOIS even with rough estimates of costs, prices and engineering design proxies. A DHSS aimed at subjective criteria for measured/indicated/inferred classification should be supplemented by a VOIS that quantifies how much data is optimal. The calculation of value ahead of mining is challenging, but important to consider.
Drilling more will reduce uncertainty, but is it worth the cost? Drilling less will cost less, but are we missing out on improved decisions? The only way to get at these questions is with a VOIS. Ben Harding completed an excellent M.Sc. thesis on this topic and summarizes his view with "a VOIS identifies the drillhole spacing that balances the cost of uncertainty and the cost of data."
The Resource Modeling Solutions platform (RMSP) within our more interactive and automatic modeling platform (AMP) have the workflows for both DHSS and VOIS. These studies bring meaningful value at different stages in the mining value chain.
