Blast designs can have productive and non-productive impacts on downstream stages, mine productivity and operating costs. On the other hand, ground vibration, fragmentation, and back break caused by blasting impose damage and financial penalties and which must be controlled by blast design. One of the significant variables in blast design is the burden. In this study, the potentials of artificial neural network are investigated in prediction of burden in the Sungun copper open-pit mine. Input data were assembled through 18 blasting blocks according to different levels and experimental geomechanical investigation. To construct the model blastability index, uniaxial compression strength, hole diameter, specific weight of rock, rock quality designation and cohesion strength are taken as input parameters, whereas, burden is considered as an output parameter. Mean square error was used as the performance function and back propagation algorithm as the training function, containing four hidden layers and 14 data sets. Four sets of data were used to make sure that correct training had been carried out. This produced the coefficient correlation of 0.662.