Enhancing rock slope stability prediction using random forest machine learning: A case study

Abstract

The prediction of slope stability is a complex nonlinear problem. This paper proposes a new method based on the random forest (RF) algorithm to study the rocky slopes stability. Taking the Bukit Merah, Perak and Twin Peak (Kuala Lumpur) as the study area, the slope characteristics of geometrical parameters are obtained from a multidisciplinary approach (consisting of geological, geotechnical, and remote sensing analyses). 18 factors, including rock strength, rock quality designation (RQD), joint spacing, continuity, openness, roughness, filling, weathering, water seepage, temperature, vegetation index, water index, and orientation, are selected to construct model input variables while the factor of safety (FOS) functions as an output. The area under the curve (AUC) value of the receiver operating characteristic (ROC) curve is obtained with precision and accuracy and used to analyse the predictive model ability. With a large training set and predicted parameters, an area under the ROC curve (the AUC) of 0.95 is achieved. A precision score of 0.88 is obtained, indicating that the model has a low false positive rate and correctly identifies a substantial number of true positives. The findings emphasise the importance of using a variety of terrain characteristics and different approaches to characterise the rock slope.

Keywords

Slope stability prediction / Random Forest Algorithm / Remote sensing in Geology / Factor of Safety (FOS) / Geometrical parameters / Rock quality designation (RQD) / Multilayer perceptron (MLP)

CRediT authorship contribution statement

Afiqah Ismail, Ahmad Safuan A. Rashida, Ali Dehghanbanadaki, Rafiuddin Hakim Roslan, Mohd Firdaus Md Danazlan, Abd Wahid Rasi, Radzuan Saari, Mushairry Mustaffar, Azman Kassim, Rini Asnida Abdullah, Khairul Hazman Padil, Norbazlan Mohd Yusof, and Norisam Abd Rahman conceived the presented idea. Afiqah Ismail, Ahmad Safuan A. Rashida, and Ali Dehghanbanadaki designed and performed the experiments. Rafiuddin Hakim Roslan, Mohd Firdaus Md@Danazlan, and Abd Wahid Rasi contributed to data analysis and interpretation. Radzuan Saari, Mushairry Mustaffar, and Azman Kassim contributed to methodology and validation. Rini Asnida Abdullah, Khairul Hazman Padil, Norbazlan Mohd Yusof, and Norisam Abd Rahman contributed to manuscript preparation and review. All authors discussed the results and contributed to the final manuscript.

Declaration of competing interest

The authors declare no conflicts of interest.

Acknowledgement

The authors would like to thank PLUS Sdn Bhd for their support in providing the data and the Universiti Teknologi Malaysia supported this work under UTM Flagship CoE/RGCoe/RG 5.2: Evaluating Surface PGA with Global Ground Motion Site Response Analyses for the highest seismic activity location in Peninsular Malaysia (Q.J130000.5022. 10G47) and Universiti Teknologi Malaysia - Earthquake Hazard Assessment in Peninsular Malaysia Using Probabilistic Seismic Hazard Analysis (PSHA) Method (Q.J130000.21A2.06E9).

References

[1] Abdul Rahim AF, Simon N, Mohamed TR, Md Rafek AG, Serasa AS, Chen YL, Zhang MW, Lee KE, Goh TL. 2019. Probabilistic analysis of potential planar-type rock slope failure of selected Malaysian rock slopes. Bulletin of the Geological Society of Malaysia, 67, 83-90. doi: 10.7186/bgsm67201910.

[2] Ali M. 2020. Pycaret: An open source, low-code machine learning library in Python. Pycaret version, 2.

[3] Amiri ST, Dehghanbanadaki A, Nazir R, Motamedi S. 2020. Unit composite friction coefficient of model pile floated in Kaolin clay reinforced by recycled crushed glass under uplift loading. Transportation Geotechnics, 22, 100313. doi: 10.1016/j.trgeo.2019.100313.

[4] Bhandari AK, Kumar A, Singh GK. 2012. Feature extraction using normalized difference vegetation index (NDVI): A case study of Jabalpur City. Procedia Technology, 6, 612-621. doi: 10.1016/j.protcy.2012.10.074.

[5] Bhattacharya S, Mishra S. 2018. Applications of machine learning for facies and fracture prediction using Bayesian Network Theory and Random Forest: Case studies from the Appalachian basin, USA. Journal of Petroleum Science and Engineering, 170, 1005-1017. doi: 10.1016/j.petrol.2018.06.075.

[6] Bisong E. 2019. Building Machine Learning and Deep Learning Models on Google Cloud Platform: A Comprehensive Guide for Beginners. Berkeley, CA, Apress. doi: 10.1007/978-1-4842-4470-8.

[7] Breiman L. 2001. Random forests. Machine Learning, 45(1), 5-32. doi: 10.1023/A:1010933404324.

[8] Bui XN, Nguyen H, Choi Y, Nguyen-Thoi T, Zhou J, Dou J. 2020. Prediction of slope failure in open-pit mines using a novel hybrid artificial intelligence model based on decision tree and evolution algorithm. Scientific Reports, 10(1), 9939. doi: 10.1038/s41598-020-66904-y.

[9] Burton CK. 1970. The palaeotectonic status of the Malay Peninsula. Palaeogeography, Palaeoclimatology, Palaeoecology, 7(1), 51-60. doi: 10.1016/0031-0182(70)90026-X.

[10] Corcoran J, Knight J, Gallant A. 2013. Influence of multi-source and multi-temporal remotely sensed and ancillary data on the accuracy of random forest classification of wetlands in northern Minnesota. Remote Sensing, 5(7), 3212-3238. doi: 10.3390/rs5073212.

[11] Dai C, Li WL, Wang D, Lu HY, Xu Q, Jian J. 2021. Active landslide detection based on sentinel-1 data and InSAR technology in Zhouqu County, Gansu Province, northwest China. Journal of Earth Science, 32(5), 1092-1103. doi: 10.1007/s12583-020-1380-0.

[12] Dehghanbanadaki A. 2021. Intelligent modelling and design of soft soil improved with floating column-like elements as a road subgrade. Transportation Geotechnics, 26, 100428. doi: 10.1016/j.trgeo.2020.100428.

[13] Department of Mineral and Geoscience 2012. Geological Map of Peninsular Malaysia: Modified based on the 8th edition, 1985. Director-General of Minerals and Geoscience Malaysia.

[14] Du PJ, Samat A, Waske B, Liu SC, Li ZH. 2015. Random Forest and Rotation Forest for fully polarized SAR image classification using polarimetric and spatial features. ISPRS Journal of Photogrammetry and Remote Sensing, 105, 38-53. doi: 10.1016/j.isprsjprs.2015.03.002.

[15] Feng RH, Grana D, Balling N. 2021. Imputation of missing well log data by random forest and its uncertainty analysis. Computers & Geosciences, 152, 104763. doi: 10.1016/j.cageo.2021.104763.

[16] Fouedjio F. 2020. Exact conditioning of regression random forest for spatial prediction. Artificial Intelligence in Geosciences, 1, 11-23. doi: 10.1016/j.aiig.2021.01.001.

[17] Ghani A A. 2009. Plutonism, In: Hutchison CS, Tan DNK (Eds.),Geology of Peninsular Malaysia. University of Malaya/Geological Society of Malaysia, Kuala Lumpur, 211-231.

[18] Ghani AA, Searle M, Robb L, Chung SL. 2013. Transitional I S type characteristic in the main range granite, peninsular Malaysia. Journal of Asian Earth Sciences, 76, 225-240. doi: 10.1016/j.jseaes.2013.05.013.

[19] Gómez H, Kavzoglu T. 2005. Assessment of shallow landslide susceptibility using artificial neural networks in Jabonosa River Basin, Venezuela. Engineering Geology, 78(1-2), 11-27. doi: 10.1016/j.enggeo.2004.10.004.

[20] He XL, Xu C, Qi WW, Huang YD, Cheng J, Xu XW, Yao Q, Lu YK, Dai BY. 2021. Landslides triggered by the 2020 Qiaojia Mw5.1 earthquake, Yunnan, China: Distribution, influence factors and tectonic significance. Journal of Earth Science, 32(5), 1056-1068. doi: 10.1007/s12583-021-1492-1.

[21] Hutchison CS, Tan DNK (Eds.). 2009. Geology of peninsular Malaysia. Published jointly by the University of Malaya and the Geological Society of Malaysia.

[22] Iranzad R, Liu X. 2024. A review of random forest-based feature selection methods for data science education and applications. International Journal of Data Science and Analytics, 1-15. doi:.10.1007/s41060-024-00509-w.

[23] ISRM. 1978. Suggested methods for the quantitative description of discontinuities in rock masses. International Journal of Rock Mechanics and Mining Sciences & Geomechanics Abstracts, 15(6), 319-368. https://doi.org/10.1016/0148-9062(78)91472-9.

[24] Jasin B. 1997. Permo-Triassic radiolaria from the Semanggolformation, northwest Peninsular Malaysia. Journal of Asian EarthSciences, 15(1), 43-53. doi: 10.1016/s1367-9120(96)00079-x.

[25] Ji L, Zhang L, Wylie B. 2009. Analysis of dynamic thresholds for the normalized difference water index. Photogrammetric Engineering & Remote Sensing, 75(11), 1307-1317. doi: 10.14358/pers.75.11.1307.

[26] Kang F, Xu B, Li JJ, Zhao SZ. 2017. Slope stability evaluation using Gaussian processes with various covariance functions. Applied Soft Computing, 60, 387-396. doi: 10.1016/j.asoc.2017.07.011.

[27] Khan AF.Rahman MT. 2022. Deep learning approach for landslide susceptibility mapping using remote sensing and GIS: A case study of northern Pakistan. Geomatics, Natural Hazards and Risk, 13(1), 1537-1560. doi: 10.1080/19475705.2022.2094789

[28] Lin S, Liang ZL, Zhao SX, Dong M, Guo HW, Zheng H. 2024. A comprehensive evaluation of ensemble machine learning in geotechnical stability analysis and explainability. International Journal of Mechanics and Materials in Design, 20(2), 331-352. doi: 10.1007/s10999-023-09679-0.

[29] Lu P, Rosenbaum MS. 2003. Artificial neural networks and grey systems for the prediction of slope stability. Natural Hazards, 30(3), 383-398. doi: 10.1023/B:NHAZ.0000007168.00673.27.

[30] Lu SL, Wu BF, Yan NN, Wang H. 2011. Water body mapping method with HJ-1A/B satellite imagery. International Journal of Applied Earth Observation and Geoinformation, 13(3), 428-434. doi: 10.1016/j.jag.2010.09.006.

[31] McFeeters S. 2013. Using the normalized difference water index (NDWI) within a geographic information system to detect swimming pools for mosquito abatement: A practical approach. Remote Sensing, 5(7), 3544-3561. doi: 10.3390/rs5073544.

[32] McKay G, Harris JR. 2016. Comparison of the data-driven random forests model and a knowledge-driven method for mineral prospectivity mapping: A case study for gold deposits around the huritz group and nueltin suite, Nunavut, Canada. Natural Resources Research, 25(2), 125-143. doi: 10.1007/s11053-015-9274-z.

[33] Mohamed T, Kasa A, Taha MR. 2012. Fuzzy logic system for slope stability prediction. International Journal on Advanced Science, Engineering and Information Technology, 2(2), 151. doi: 10.18517/ijaseit.2.2.174.

[34] Nanehkaran YA, Zhu LC, Jin CY, Chen JD, Anwar S, Azarafza M, Derakhshani R. 2023. Comparative analysis for slope stability by using machine learning methods. Applied Sciences, 13(3), 1555. doi: 10.3390/app13031555.

[35] Paudel U, Oguchi T, Hayakawa Y. 2016. Multi-resolution landslide susceptibility analysis using a DEM and random forest. International Journal of Geosciences, 7(5), 726-743. doi: 10.4236/ijg.2016.75056.

[36] Qu JK, Qi SC, Zhou JW. 2023. A three-dimensional limit equilibrium method for slope stability analysis integrating machine learning optimized intercolumn force function. Geological Journal, 58(6), 2438-2453. doi: 10.1002/gj.4770.

[37] Reading AM, Cracknell MJ, Bombardieri D, Chalke T. 2015. Combining machine learning and geophysical inversion for applied geophysics. In ASEG-PESA 24th International Geophysical Conference and Exhibition, Extended Abstracts (ASEG Extended Abstracts, Vol. 2015, p. ab070). Perth, Western Australia. doi: 10.1071/ASEG2015ab070

[38] Rodriguez-Galiano V, Sanchez-Castillo M, Chica-Olmo M, Chica-Rivas M. 2015. Machine learning predictive models for mineral prospectivity: An evaluation of neural networks, random forest, regression trees and support vector machines. Ore Geology Reviews, 71, 804-818. doi: 10.1016/j.oregeorev.2015.01.001.

[39] Sahin E. 2020. A comparative study of machine learning algorithms for landslide susceptibility mapping: A case study from Turkey. Environmental Earth Sciences, 79(5), 1-17. doi: 10.1007/s12665-020-08847-1.

[40] Salman HA, Kalakech A, Steiti A. 2024. Random forest algorithm overview. Babylonian Journal of Machine Learning, 2024, 69-79. doi: 10.58496/bjml/2024/007.

[41] Singh P, Bardhan A, Han FC, Samui P, Zhang WG. 2023. A critical review of conventional and soft computing methods for slope stability analysis. Modeling Earth Systems and Environment, 9(1), 1-17. doi: 10.1007/s40808-022-01489-1.

[42] Shu YK, 1969. Some NW trending faults in Kuala Lumpur and other areas. Newsletter Geological Society of Malaysia, 17, 1-5.

[43] Stauffer P. H. 1968. The Kuala Lumpur fault zone: A proposed major strike-slip fault across Malaya. Newsletter Geological Society of Malaysia, 15, 2-4.

[44] Tsangaratos P, Benardos A. 2016. Applying artificial neural networks in slope stability related phenomena. Bulletin of the Geological Society of Greece, 47(4), 1901. doi: 10.12681/bgsg.10945.

[45] Wang GJ, Zhao B, Wu BS, Zhang C, Liu WL. 2023. Intelligent prediction of slope stability based on visual exploratory data analysis of 77 in situ cases. International Journal of Mining Science and Technology, 33(1), 47-59. doi: 10.1016/j.ijmst.2022.07.002.

[46] Wen GZ, Han LL, Lin W, Xing Z, Yan MZ. 2023. Machine Learning Algorithms. Application of Machine Learning in Slope Stability Assessment, 13-29.

[47] Wu HS, Chen YL, Lv HY, Xie QH, Chen YG, Gu J. 2022. Stability analysis of rib pillars in highwall mining under dynamic and static loads in open-pit coal mine. International Journal of Coal Science & Technology, 9(1), 38. doi: 10.1007/s40789-022-00504-1.

[48] Xie HY, Dong JJ, Deng Y, Dai YW. 2022. Prediction model of the slope angle of rocky slope stability based on random forest algorithm. Mathematical Problems in Engineering, 2022, 9441411. doi: 10.1155/2022/9441411.

[49] Ye J, Liu Z, Wang Y, Zhang L. 2022. Slope stability prediction using improved random forest and SHAP analysis: A case study in Southwest China. Engineering Geology, 302, 106618. doi: 10.1016/j.enggeo.2022.106618.

[50] Zhang WG, Gu X, Hong L, Han L, Wang L. 2023. Comprehensive review of machine learning in geotechnical reliability analysis: Algorithms, applications and further challenges. Applied Soft Computing, 136, 110066. doi: 10.1016/j.asoc.2023.110066.

[51] Zhu JF, Pierskalla WP. 2016. Applying a weighted random forests method to extract Karst sinkholes from LiDAR data. Journal of Hydrology, 533, 343-352. doi: 10.1016/j.jhydrol.2015.12.012.