Efficacy of Information in Mineral Exploration Drilling

In mineral exploration, targets need to be confirmed by means of drilling. Such drilling is based on anomalies often detected in geochemical and/or geophysical exploration. Because drilling is costly and because mineral exploration has low success rate, the need to know quickly, with minimal drilling, whether a target is mineable or not, is an important problem. Value of information (VOI) is a decision-theoretic concept that quantifies by how much information reduces risk about a project dollar-value; hence, optimizing its value provides a means for optimizing the location of drilling. However, VOI depends on actual dollar values of success vs. failure, which may be uncertain themselves, or not yet relevant. In this paper, we introduce a new concept, termed efficacy of information (EOI), which is like VOI but without monetary rewards or costs. It quantifies by how much, on average, future information reduces uncertainty on some (discrete) property of interest. We present an efficient method for calculating EOI based on approximately solving a Bayesian inverse problem. We determine the location of a single, or set of boreholes sequentially, by optimizing their EO1. We also apply this method on an actual mineral deposit.