A novel classification approach was proposed for prediction of flyrock throw distance based on the site measurements performed in a sandstone quarry. The classification approach was created by using multiple discriminant analysis (MDA) technique. The input parameters of discriminant analysis are blast design parameters and a rock mass parameter. The grouping stage of classification was performed considering a well-known blast regulation for flyrock. Additionally, multiple regression analysis technique applied to blast data to create a flyrock prediction equation. By this way, the capability and differences of the classification approach were investigated. Model validation was performed on the test blasts. MDA model successfully estimated the flyrock throw distance. Unlike the classical prediction models, the MDA model predicts severity of flyrock throws instead of a numerical value. MDA model can be easily practiced by the created territorial map. The model does not require any specific software or training for usage and suitable for practical applications at mines.