Improvementof small scale dolomite mine blast fragmentation efficiency using hybrid artificial intelligence and soft computing approaches-A case study

Abstract Blast fragmentation efficiency prediction can improve blast productivity, but it is capital intensive when done with available software. Soft computing models based on artificial intelligence approaches and multivariate regression techniques were proposed in this study to predict small diameter hole blast fragmentation efficiency at the Akoko Edo dolomite quarry. WipFrag software was used to analyze and evaluate blast fragmentation performance based on images of blast muck piles. The proposed model's performance was evaluated using five model performance indices, comprising the average bias error, root mean square error, correlation coefficient, and average absolute error. The obtained results indicate that the proposed model of the Adaptive Neuro-fuzzy inference system (ANFIS) is proficient of accurate prediction of blast fragmentation efficiency when compared to artificial neural networks and multivariate regression models.