<![CDATA[Blasting activities in mining are intended to fragment and displace rock masses, but they also generate environmental impacts in the form of ground vibration. Ground vibration measurement is required to predict allowable vibration limits. This study aims to predict ground vibration levels caused by blasting activities using the Artificial Neural Network method at the open pit coal mine. The research employed an applied research method using primary data consisting of ground vibration levels, blasting distance, and explosive charge weight obtained through direct field measurements. Data processing and the determination of model architecture and parameters were conducted using Matlab software with the backpropagation learning method. Model training and testing were carried out using blasting data from two blasting location. The results indicate that the Artificial Neural Network model with three input variables, ten hidden neurons, one output neuron, and one output layer was able to predict ground vibration levels with root mean square error values of 0.18 and 0.63. The prediction accuracy for peak particle velocity reached 93.24 percent at Pit 1 and 89.55 percent at Pit TSBC. These results demonstrate that the Artificial Neural Network method is effective for predicting ground vibrations caused by blasting activities, allowing blasting geometry planning to be better controlled to minimize environmental impacts]]>
Copyrights © 2026
