 This paper presents a novel approach for predicting peak particle velocity, PPV, in surface mines using an ensemble system consisting of artificial neural networks, ANS, and extreme gradient boosting, XG-Boosts. The results show that the ensemble system out performs the best individual models in terms of accuracy and can be used to accurately predict PPV in surface mines. Furthermore, the sensitivity analysis indicates that the spacing between blasts and the number of blast holes have the greatest and least effect on PPV, respectively.