Enhancing rock fragmentation prediction in mining operations: A Hybrid GWO-RF model with SHAP interpretability

Abstract

In the mining industry, precise forecasting of rock fragmentation is critical for optimising blasting processes. In this study, we address the challenge of enhancing rock fragmentation assessment by developing a novel hybrid predictive model named GWO-RF. This model combines the Grey Wolf Optimization (GWO) algorithm with the Random Forest (RF) technique to predict the D₈₀ value, a critical parameter in evaluating rock fragmentation quality. The study is conducted using a dataset from Sarcheshmeh copper mine, employing six different swarm sizes for the GWO-RF hybrid model construction. The GWO-RF model’s hyperparameters are systematically optimized within established bounds, and its performance is rigorously evaluated using multiple evaluation metrics. The results show that the GWO-RF hybrid model has higher predictive skills, exceeding traditional models in terms of accuracy. Furthermore, the interpretability of the GWO-RF model is enhanced through the utilization of SHapley Additive exPlanations (SHAP) values. The insights gained from this research contribute to optimizing blasting operations and rock fragmentation outcomes in the mining industry.

Keywords

blasting / rock fragmentation / random forest / grey wolf optimization / hybrid tree-based technique

References

[1] RoyM P, PaswanR K, SarimM D. Rock fragmentation by blasting-A review[J]. Journal of mines, metals and fuels, 2016, 64(9): 424-431

[2] FaramarziF, MansouriH, FarsangiM A E. A rock engineering systems based model to predict rock fragmentation by blasting[J]. International Journal of Rock Mechanics and Mining Sciences, 2013, 60: 82-94

[3] Fourney W L. Mechanisms of rock fragmentation by blasting [J]. Compr Rock Eng, 2016: 39–68.

[4] ChoS H, KanekoK. Rock fragmentation control in blasting [J]. Materials transactions, 2004, 45(5): 1722-1730

[5] HasanipanahM, Jahed ArmaghaniD, MonjeziM. Risk assessment and prediction of rock fragmentation produced by blasting operation: a rock engineering system[J]. Environmental Earth Sciences, 2016, 75: 1-12

[6] ZhangY, HeH, KhandelwalM, DuK, ZhouJ. Knowledge mapping of research progress in blast-induced ground vibration from 1990 to 2022 using CiteSpace-based scientometric analysis. Environmental Science and Pollution Research, 2023, 30(47): 103534-103555

[7] MorinM A, FicarazzoF. Monte Carlo simulation as a tool to predict blasting fragmentation based on the Kuz – Ram model [J]. Computers & geosciences, 2006, 32(3): 352-359

[8] LiE, YangF, RenM, et al. . Prediction of blasting mean fragment size using support vector regression combined with five optimization algorithms[J]. Journal of Rock Mechanics and Geotechnical Engineering, 2021, 13(6): 1380-1397

[9] DongL, WangJ, WangJ, et al. . Safe and intelligent mining: Some explorations and challenges in the era of big data[J]. Journal of Central South University, 2023, 30(6): 1900-1914

[10] AslamM N, RiazA, ShaukatN, et al. . Analysis of incompressible viscous fluid flow in convergent and divergent channels with a hybrid meta-heuristic optimization techniques in ANN: An intelligent approach[J]. Journal of Central South University, 2023, 30(12): 4149-4167

[11] GheibieS, AghababaeiH, HoseinieS H, et al. . Modified Kuz—Ram fragmentation model and its use at the Sungun Copper Mine[J]. International Journal of Rock Mechanics and Mining Sciences, 2009, 46(6): 967-973

[12] HjelmbergH. Some ideas on how to improve calculations of the fragment size distribution in bench blasting[C]. 1st international symposium on rock fragmentation by blasting, 1983, Lulea Sweden, Lulea University Technology: 469494

[13] Stagg M S, Ottemess R E, Siskind D E. Effects of blasting practices on fragmentation[C]//ARMA US Rock Mechanics/Geomechanics Symposium. ARMA, 1992: ARMA-92-0313.

[14] GhiasiM, AskarnejadN, DindarlooS R, et al. . Prediction of blast boulders in open pit mines via multiple regression and artificial neural networks[J]. International journal of mining science and technology, 2016, 26(2): 183-186

[15] BakhtavarE, KhoshrouH, BadroddinM. Using dimensional-regression analysis to predict the mean particle size of fragmentation by blasting at the Sungun copper mine[J]. Arabian Journal of Geosciences, 2015, 8: 2111-2120

[16] DhekneP, PradhanM, JadeR K. Assessment of the effect of blast hole diameter on the number of oversize boulders using ANN model[J]. Journal of the Institution of Engineers (India): Series D, 2016, 97: 21-31

[17] MonjeziM, AmiriH, FarrokhiA. Prediction of rock fragmentation due to blasting in Sarcheshmeh copper mine using artificial neural networks[J]. Geotechnical and Geological Engineering, 2010, 28: 423-430

[18] MonjeziM, RezaeiM, VarjaniA Y. Prediction of rock fragmentation due to blasting in Gol-E-Gohar iron mine using fuzzy logic[J]. International Journal of Rock Mechanics and Mining Sciences, 2009, 46(8): 1273-1280

[19] RigattiS J. Random forest[J]. Journal of Insurance Medicine, 2017, 47(1): 31-39

[20] PetersonL E. K-nearest neighbor[J]. Scholarpedia, 2009, 4(2): 1883

[21] Awad M, Khanna R, Awad M, et al. Support vector regression [J]. Efficient learning machines: Theories, concepts, and applications for engineers and system designers, 2015: 67–80.

[22] FanJ, MaX, WuL, et al. . Light Gradient Boosting Machine: An efficient soft computing model for estimating daily reference evapotranspiration with local and external meteorological data[J]. Agricultural water management, 2019, 225: 105758

[23] Wilson A G, Knowles D A, Ghahramani Z. Gaussian process regression networks[J]. arXiv preprint arXiv: 1110.4411, 2011.

[24] MirjaliliS, MirjaliliS M, LewisA. Grey wolf optimizer[J]. Advances in engineering software, 2014, 6946-61

[25] Dai Y, Khandelwal M, Qiu Y, et al. A hybrid metaheuristic approach using random forest and particle swarm optimization to study and evaluate backbreak in open-pit blasting[J]. Neural Computing and Applications, 2022: 1–16.

[26] ZhouJ, ZhangY, LiC, et al. . Enhancing the performance of tunnel water inflow prediction using Random Forest optimized by Grey Wolf Optimizer[J]. Earth Science Informatics, 2023, 16(3): 2405-2420

[27] ZhouJ, DaiY, TaoM. Estimating the mean cutting force of conical picks using random forest with salp swarm algorithm [J]. Results in Engineering, 2023, 17100892

[28] ZhouJ, HuangS, QiuY. Optimization of random forest through the use of MVO, GWO and MFO in evaluating the stability of underground entry-type excavations[J]. Tunnelling and Underground Space Technology, 2022, 124104494

[29] ZhouJ, DaiY, DuK, et al. . COSMA-RF: New intelligent model based on chaos optimized slime mould algorithm and random forest for estimating the peak cutting force of conical picks[J]. Transportation Geotechnics, 2022, 36100806

[30] ZhouJ, DaiY, KhandelwalM, et al. . Performance of hybrid SCA-RF and HHO-RF models for predicting backbreak in open-pit mine blasting operations[J]. Natural Resources Research, 2021, 304753-4771

[31] ZhouJ, ZhangY, QiuY. State-of-the-art review of machine learning and optimization algorithms applications in environmental effects of blasting[J]. Artificial Intelligence Review, 2024, 57(1): 1-54

[32] KhorasanipourM, JafariZ. Environmental geochemistry of rare earth elements in Cu-porphyry mine tailings in the semiarid climate conditions of Sarcheshmeh mine in southeastern Iran. Chemical Geology, 2018, 477: 58-72

[33] KhorasanipourM, RashidiS. Geochemical fractionation pattern and environmental behaviour of rare earth elements (REEs) in mine wastes and mining contaminated sediments; Sarcheshmeh mine, SE of Iran[J]. Journal of Geochemical Exploration, 2020, 210: 106450

[34] BenestyJ, ChenJ, HuangY. On the importance of the Pearson correlation coefficient in noise reduction[J]. IEEE Transactions on Audio, Speech, and Language Processing, 2008, 16(4): 757-765

[35] ZhouJ, ChenY, LiC, et al. . Machine learning models to predict the tunnel wall convergence[J]. Transportation Geotechnics, 2023, 41101022

[36] HoangN D, BuiD T. Predicting earthquake-induced soil liquefaction based on a hybridization of kernel Fisher discriminant analysis and a least squares support vector machine: a multi-dataset study[J]. Bulletin of Engineering Geology and the Environment, 2018, 77191-204

[37] PaulA, MukherjeeD P, DasP. Improved random forest for classification[J]. IEEE Transactions on Image Processing, 2018, 27(8): 4012-4024

[38] Nohara Y, Matsumoto K, Soejima H, et al. Explanation of machine learning models using improved shapley additive explanation[C]//Proceedings of the 10th ACM international conference on bioinformatics, computational biology and health informatics. 2019: 546–546.

[39] WuY, ZhouY. Hybrid machine learning model and Shapley additive explanations for compressive strength of sustainable concrete[J]. Construction and Building Materials, 2022, 330: 127298

[40] FutagamiK, FukazawaY, KapoorN, et al. . Pairwise acquisition prediction with SHAP value interpretation[J]. The Journal of Finance and Data Science, 2021, 722-44

[41] CaoA, LiuY, YangX, et al. . FDNet: Knowledge and data fusion-driven deep neural network for coal burst prediction [J]. Sensors, 2022, 22(8): 3088

[42] WilcoxonF. Individual comparisons by ranking methods[M]. Breakthroughs in statistics: Methodology and distribution, 1992, New York, NY, Springer New York: 196-202

[43] WilcoxonF. Probability tables for individual comparisons by ranking methods[J]. Biometrics, 1947, 3(3): 119-122