Estimation of Blast-Induced Peak Particle Velocity through the Improved Weighted Random Forest Technique

Blasting is one of the primary aspects mining operations, and its environmental effects interfere with safety lives property. Therefore, it essential to accurately estimate impact blasting, i.e., peak particle velocity (PPV). In this study, a regular random forest (RF) model was developed using 102 blasting samples that were collected from an open granite mine. The inputs included six parameters, while output PPV. Then, improve performance RF model, five techniques, refined weights based on accuracy decision trees optimization three metaheuristic algorithms, proposed enhance predictive capability model. results showed all weighted models have better than particular, whale algorithm (WOA) best performance. Moreover, sensitivity analysis revealed powder factor (PF) has most significant prediction PPV in project case, which means magnitude can be managed by controlling size PF.