Abstract
Rock breaking is one of the most important stages in mining operations. The commonly used method of rock breaking is blasting which aims to provide rocks according to the size of fragmentation that has been planned. The effectiveness blasting activities is one of the measurements success of loading activities which greatly affect the achievement of mining productivity targets. To produce good fragmentation, many things affect the blasting geometry, one of the things that can be controlled. To determine the level of fragmentation resulting from blasting, split desktop software can be used. This study aims to obtain a fragmentation modeling by predicting the fragmentation of blasting products at PT Dempo Bangun Mitra. The prediction of fragmentation in this study uses the Kuz-Ram model method and the Kuznetzov-Cunningham-Ouchterlony (KCO) model. To validate the method used, the correlation coefficient (R²) and Root Mean Square Error (RMSE) will be analyzed between actual measurements using split desktop software and the prediction method used. This study also evaluates the blasting geometry using R.L.Ash theory. Based on the data analysis carried out, the percentage of the calculation of actual blasting fragmentation results is 74.95% with a boulder percentage of 25.05%. Meanwhile, the results of the analysis of the fragmentation prediction of the Kuz-Ram model were 96.94% and the results of the analysis from the KCO model were 72.71%. To reduce the amount of boulder percentage, it is necessary to evaluate the blasting geometry based on the R.L.Ash theory by using the KCO model so that the boulder percentage is 0.05%, thus the number of boulders is less.