With the recent advancement in drill core imaging systems, a large volume of high resolution drill core images can now be rapidly collected and mineralogically classified. This paper introduces a novel semi-automated method for rapid prediction of process engineering parameters from this type of image. The aim is to allow for a rapid and low-cost assessment of the ore processing variability within a mineral deposit with a reduced amount of metallurgical test work. This method will see its application primarily in the project evaluation phase, where proper evaluation of the processing variability of the orebody normally requires a prolonged and costly metallurgical test work program. Applying this image-based texture analysis method on the mineralogical classified drill core images, the ores are classified according to their textures. For each engineering parameter, a small number of physical test work is required to produce a data set for training and testing of the predictive model. The paper also presents a case study in which this method has been applied to predict the A*b parameter used in mineral process plant design._x000D_
CITATION: Nguyen, A, Jackson, J, Nguyen. K and Manlapig, E, 2016. A New Semi-automated Method to Rapidly Evaluate the Processing Variability of the Orebody, in Proceedings The Third AusIMM International Geometallurgy Conference (GeoMet) 2016, pp 145-152 (The Australasian Institute of Mining and Metallurgy: Melbourne).