Abstract

In industrial branches related to geology (such as oil, gas and groundwater management; mining, risk management in transport infrastructure or soil pollution) we are always confronted with the need to take design or operative decisions to increase our profit and decrease the risk of losses, human and economic. The general rule is to maximize the expected value of some profit function, which requires encapsulating our uncertainties on the nature of the environment into a probability distribution, conditional on the available information. In this talk we will see how this general idea is used to establish an evidence-based routing system for materials from an iron ore mine based on simple beneficiation models and a geostatistical description of the ore body, obtained from borehole assay data. The conclusions will have general implications in mining and in the other fields mentioned above.

Biography

Raimon Tolosana-Delgado is an engineering geologist with 20 years of experience at the interface between geosciences, mathematics and engineering. Coauthor of more than 100 papers and three books, he is a holder of the Felix Chayes Prize for excellence in mathematical petrology (2013) of the International Association for Mathematical Geosciences. His areas of expertise cover geostatistics, compositional data analysis, stochastic modelling and machine learning, in application fields such as sedimentology, ocean sciences and geometallurgy.