Application of Soft Computing Technique based on ANN Model Prediction in Diverse Area of Mining Blasting Operations

The mining industry is a big business globally, blasting is carried out with high explosive energy for massive rock disintegration. South Africa as a country makes a significant contribution to mining in terms of GDP. For a competitive mining industry to be achieved in South Africa and globally, there is a need for increased efficiency in blasting operations, optimization of process parameters, and others. Creative algorithm is required as a computing technique that plays a quintessential role in environmental analysis, exploration, and exploitation process. It has been observed in recent times that one of the major challenges of this industry is the problem of uncertainty which exist because of the nature of the sub-surface during blasting operations. This study makes contribution in that direction. Firstly, a detailed study on relevant areas of application of ANN in mining operation was exposed, focusing on rock fragmentation; blast-induced ground vibration and peak-particle velocity; blast-induced air-blast; cost analysis, and equipment selection during blasting. Secondly, the ANN model was applied to achieve the optimal results using dataset obtained during blasting operations. 70% of the dataset was sectioned for training and 30% for testing and validation. The statistical indicators produced highly satisfactory results. It was concluded that the ANN model is a suitable technique for unraveling the problem of uncertainties in diverse areas of mining operations.