Abstarct: Blasting is the one of the most important operation in open pit mining, which is associated with undesirable and unwanted effects. One of the consequences of the blasting operation is the flyrock phenomenon, which can have unpleasant consequences such as damage to equipment and people in the mine. The first step to reducing the harmful effects of the flryock is to predict it. There are several methods, including experimental methods, mathematical methods and artificial intelligence to predict this phenomenon. Artificial intelligence methods have more accuracy, efficiency and speed than other methods. The factors affecting the flyrock can be divided into two main categories which includes controllable and uncontrollable parameters. The aim of this thesis is to predict the flyrock of the blasting operation in the Anguran mine. For this purpose, at first the data of 91 blasting operation were collected from this mine. Then, two models were developed to predict flyrock using predictive adaptive network-baxsed fuzzy inference system (ANFIS) in MATLAB and support vector machine (SVM) in WEKA software. The coefficient of determination (R2) and the Root Mean Square Error (RMSE) for the ANFIS were obtained of 0.959 and 6.878 and for the SVM were found of 0.974 and 5.522 respectively, which indicates a better performance of SVM. Finally, with the use of the Cosine Amplitude Method (CAM), sensitivity analysis was performed on the parameters. The result show that the powder factor and subdrilling had the most and least effect on the flyrock respectively. Knowing the amount of flyrock, it is possible to use controlling methods to reduce the harmful effects of this phenomenon, which will ultimately improve the performance of blasting and mining operations.