Turkish Journal of Earth Sciences




Grade value is a crucial parameter for the mineral industry. Investigation of grade value of mineral resources provides the optimum benefit. In this study, an adaptive neuro-fuzzy inference system (ANFIS) and artificial neural network (ANN) model were applied for the prediction of grade values. The spatial coordinates X, Y, and Z of the study area along with bore hole geochemical data were used as input variables in the model. In order to illustrate the applicability and capability of these methods, the western part of Turkey, between the latitudes 38°01 45 N and 38°09 52 N and between the longitudes 31°23 20 E and 31°32 52 E was chosen as the case study area. Measured grades of barite samples were obtained from 47 boreholes using the chemical analysis method. The performance of these models in training and testing sets were evaluated and compared with the observations. The results indicate that the ANFIS model is better than the ANN model and can successfully provide high accuracy and reliability for grade estimation.


Sedimentary barite, adaptive neuro-fuzzy inference system, artificial neural network, grade estimation, uncertainty, membership function, rule base

First Page


Last Page