A hybrid metaheuristic approach using random forest and particle swarm optimization to study and evaluate backbreak in open-pit blasting

Abstract

Backbreak is a rock fracture problem that exceeds the limits of the last row of holes in an explosion operation. Excessive backbreak increases operational cost and also poses a threat to mine safety. In this regard, a new hybrid intelligence approach based on random forest (RF) and particle swarm optimization (PSO) are proposed for predicting backbreak with high accuracy to reduce the unsolicited phenomenon induced by backbreak in open-pit blasting. A data set of 234 samples with 6 input parameters including special drilling (SD), spacing (S), burden (B), hole length (L), stemming (T) and powder factor (PF)) and one output parameter backbreak (BB) is set up in this study. Seven input combinations (one with six parameters, six with five parameters) are built to generate the optimal prediction model. The PSO algorithm was integrated with the RF algorithm to find the optimal hyper-parameters ( [[EQUATION]] and [[EQUATION]] ) of each model and the fitness function, which is the MAE of 10 cross-validations. The performance capacities of the optimal models are assessed using mean absolute error, root mean square error, Pearson correlation coefficient and mean absolute percentage error. Findings demonstrated that the PSO-RF model combining L-S-B-T-PF with MAE of (0.0132, and 0.0568), RMSE of (0.0811, 0.1686), R 2 of (0.9990, 0.9961), and MAPE of (0.0027, and 0.0116) in training and testing phases, respectively, has optimal prediction performance. The optimal PSO-RF models are compared with the classical Artificial Neural Network (ANN), RF, Genetic Programming (GP), and Support Vector Machine (SVM) models that show that the PSO-RF model has superiority in predicting backbreak. The Gini index of each input variable has also been calculated in the RF model, which were 31.2(L), 23.1(S), 27.4(B), 36.6(T), 23.4(PF), and 16.9(SD), respectively.

Four Severity Levels for Grading the Tortuosity of a Retinal Fundus Image

“…A powerful bagging-based ensemble algorithm is the distributed random forest (DRF), which improves learning by addressing the problem of local optima and covering the full search space [ 48 ]. It uses a combination of decision trees to maximize the model classification efficiency rather than using just one as a weak learner.…”

Smart Energy Management: A Comparative Study of Energy Consumption Forecasting Algorithms for an Experimental Open-Pit Mine

“…Each tree is made up of column variables and row observations that are chosen at random. A single DT is difficult to predict accurately, but all DTs form a forest, allowing the aggregated findings to incorporate the outcomes of all DTs, resulting in a more accurate overall prediction [59].…”

Application of Artificial Intelligence Techniques for Predicting the Back-break in Blasting Operation

“…Dai Yong et al (2022) modeled BB based on the parameters of hole length (L), spacing (B, S), stemming (T), specific drilling (SD), specific charge (PF), and an output parameter (BB) using random forest (RF) and particle optimization swarm (PSO) algorithms. The results showed that parameters L and T have the highest Gini index in the RF model, indicating their effect on the output of the model (Dai et al, 2022). In 2021 and 2022, Sohrabi et al used Whale algorithm and neural network to optimize crude oil and coal price estimation.…”