Assessing Ground Vibration Caused by Rock Blasting in Surface Mines Using Machine-Learning Approaches: A Comparison of CART, SVR and MARS

Ground vibration induced by rock blasting is an unavoidable effect that may generate severe damages to structures and living communities. Peak particle velocity (PPV) the key predictor for ground vibration. This study aims develop a model predict PPV in opencast mines. Two machine-learning techniques, including multivariate adaptive regression splines (MARS) classification tree (CART), which are easy implement field engineers, were investigated. The models developed using record of 1001 real blast-induced vibrations, with ten (10) corresponding parameters from 34 mines/quarries India Benin. suitability one technique over other was tested comparing outcomes support vector (SVR) algorithm, multiple linear regression, different empirical predictors Taylor diagram. results showed MARS outperformed this lower error (RMSE = 0.227) R2 0.951, followed SVR (R2 0.87), CART 0.74) predictors. Based on large-scale cases input variables involved, should lead better representative high generalization ability. proposed can easily be implemented engineers prediction reasonable accuracy.