Data-Driven Optimized Artificial Neural Network Technique for Prediction of Flyrock Induced by Boulder Blasting

One of the most undesirable consequences induced by blasting in open-pit mines and civil activities is flyrock. Furthermore, production oversize boulders creates many problems for continuation work usually imposes additional costs on project. In this way, breakage associated with throwing small fragments particles at high speed, which can lead to serious risks human resources infrastructures. Hence, accurate prediction flyrock boulder crucial avoid possible its’ environmental side effects. This study attempts develop an optimized artificial neural network (ANN) particle swarm optimization (PSO) jellyfish search algorithm (JSA) construct hybrid models anticipating distance resulting a quarry mine. The PSO JSA algorithms were used determine optimum values neurons’ weight biases connected neurons. regard, database involving 65 monitored recording was collected that comprises six influential parameters distance, i.e., hole depth, burden, angle, charge weight, stemming, powder factor one target parameter, distance. ten various ANN, PSO–ANN, JSA–ANN established estimating their results investigated applying three evaluation indices coefficient determination (R2), root mean square error (RMSE) value accounted (VAF). calculation indicators revealed R2, (0.957, 0.972 0.995) (0.945, 0.954 0.989) determined train test proposed predictive models, respectively. yielded denoted although ANN model capable PSO–ANN anticipate more accuracy. performance accuracy level estimate better compared models. Therefore, identified as superior from blasting. final, sensitivity analysis conducted showed factor, angle have impact changes.