Decision support system for bioleaching processes

Abstract The use of information available in the organizations to understand what good performance looks like has been proposed for improving the decreased productivity in mine sector. Detailed monitoring has been performed at the heap bioleaching process in Minera Escondida since the start of the industrial operation in 2006. The huge industrial data recorded represents an opportunity to raise knowledge about complex bioleaching processes for improving the technology. A systematic approach using machine learning tools for the analysis of High Dimensional Feature Space is now being developed to deliver experience-based learning with the aim to serve as the foundation for optimal production planning and operational decision making, in the presence of inherent process variations. The construction of a Decision Support System (DSS) is reported, which considers a Real Time PCR array, a database for data logging and storage, the application of suitable statistical and computational tools for knowledge acquiring and finally the creation of a system of knowledge translation to transform it into action by applying recommendations that come to terms with operational limitations. The user can accurately retrieve data and design similar matches to the historic operation to get, for instance the expected metallurgical performance (such as copper recovery, acid consumption and bacterial activity) and recommendations. The process followed to construct the base of knowledge of the DSS is discussed.