Abstract

It is vital to predict back breaks during blasting operations to avert the undesirable effect it might cause. For this reason, 40 blasting operations were witnessed in the Chinese Government Company (CGC) quarry to obtain the bench height, burden, hole diameter, and spacing. The pertinent equations were employed to compute the stiffness ratio and rock strength. Artificial Neural Network (ANN) methodology was adopted to analyze the impact of the different blasting parameters on the back break. The trained ANN is subsequently applied to predict back-break outcomes for new sets of blasting parameters. The results highlight a robust high correlation of (R2 = 0.9954) between the input blasting parameters and back break, underscoring their substantial influence on the phenomenon. Furthermore, the ANN model exhibits a high level of accuracy in predicting back breaks, suggesting its potential as a valuable tool for optimizing blasting parameters to mitigate back-break occurrences. It was observed that stiffness ratio has a higher effect on backbreak than rock strength. It was deduced that by using the predicted equation, there would be a reduction in back breaks as compared to the usual outcome after blasting.

Key words: Blasting, Backbreak, Correlation, Blast parameters, Artificial Neural Network