Prediction on compression indicators of clay soils using XGBoost with Bayesian optimization

Abstract

The determination of the compressibility of clay soils is a major concern during the design and construction of geotechnical engineering projects. Directly acquiring precise values of compression indicators from consolidation tests are cumbersome and time-consuming. Based on experimental results from a series of index tests, this study presents a hybrid method that combines the XGBoost model with the Bayesian optimization strategy to show the potential for achieving higher accuracy in predicting the compressibility indicators of clay soils. The results show that the proposed XGBoost model selected by Bayesian optimization can predict compression indicators more accurately and reliably than the artificial neural network (ANN) and support vector machine (SVM) models. In addition to the lowest prediction error, the proposed XGBoost-based method enhances the interpretability by feature importance analysis, which indicates that the void ratio is the most important factor when predicting the compressibility of clay soils. This paper highlights the promising prospect of the XGBoost model with Bayesian optimization for predicting unknown property parameters of clay soils and its capability to benefit the entire life cycle of engineering projects.

Keywords

machine learning / clay soils / compression indicators / XGBoost / Bayesian optimization

References

[1] JiaoC-J, YouS, JiH-G. Numerical analysis of building structures response under effect of continuous and stepped non-uniform settlement [J]. Journal of Central South University, 2023, 30(12): 4066-4081

[2] GongC-J, XieC-R, LinZ-Q, et al. . Ground deformation prediction induced by shield tunnelling considering existing multi-story buildings [J]. Journal of Central South University, 2023, 30(4): 1373-1387

[3] TERZAGHI K, PECK R B, MESRI G. Soil mechanics in engineering practice [M]. John Wiley & Sons, 1956.

[4] AzzouzA S, KrizekR J, CorotisR B. Regression analysis of soil compressibility [J]. Soils and Foundations, 1976, 16(2): 19-29

[5] YoonG L, KimB T, JeonS S. Empirical correlations of compression index for marine clay from regression analysis [J]. Canadian Geotechnical Journal, 2004, 41(6): 1213-1221

[6] ShimobeS, SpagnoliG. A general overview on the correlation of compression Index of clays with some geotechnical index properties [J]. Geotechnical and Geological Engineering, 2022, 40(1): 311-324

[7] LecunY, BengioY, HintonG. Deep learning [J]. Nature, 2015, 521(7553): 436-444

[8] ZhangW-G, ZhangR-H, WuC-Z, et al. . State-of-the-art review of soft computing applications in underground excavations [J]. Geoscience Frontiers, 2020, 11(4): 1095-1106

[9] ZHANG Wen-gang, LI Hong-rui, LI Yong-qin, et al. Application of deep learning algorithms in geotechnical engineering: A short critical review [J]. Artificial Intelligence Review, 2021: 1–41. DOI: https://doi.org/10.1007/s10462-021-09967-1.

[10] ZHANG Zi-long, PAN Qiu-jing, YANG Zi-han, et al. Physics-informed deep learning method for predicting tunnelling-induced ground deformations [J]. Acta Geotechnica, 2023: 1–16. DOI: https://doi.org/10.1007/s11440-023-01874-9.

[11] ParkH I, LeeS R. Evaluation of the compression index of soils using an artificial neural network [J]. Computers and Geotechnics, 2011, 38(4): 472-481

[12] AladagC H, KayabasiA, GokceogluC. Estimation of pressuremeter modulus and limit pressure of clayey soils by various artificial neural network models [J]. Neural Computing and Applications, 2013, 23: 333-339

[13] BenbourasM A, KettabM R, ZediraH, et al. . A new approach to predict the compression index using artificial intelligence methods [J]. Marine Georesources & Geotechnology, 2019, 37(6): 704-720

[14] PhamB T, NguyenM D, vanD D, et al. . Development of artificial intelligence models for the prediction of compression coefficient of soil: An application of Monte Carlo sensitivity analysis [J]. Science of the Total Environment, 2019, 679: 172-184

[15] ZhangD-M, ZhangJ-Z, HuangH-W, et al. . Machine learning-based prediction of soil compression modulus with application of 1D settlement [J]. Journal of Zhejiang University-Science A (Applied Physics & Engineering), 2020, 21(6): 430-444

[16] BuiD T, NhuV H, HoangN D. Prediction of soil compression coefficient for urban housing project using novel integration machine learning approach of swarm intelligence and multi-layer perceptron neural network [J]. Advanced Engineering Informatics, 2018, 38: 593-604

[17] AsterisP G, MamouA, FerentinouM, et al. . Predicting clay compressibility using a novel Manta ray foraging optimization-based extreme learning machine model [J]. Transportation Geotechnics, 2022, 37: 100861

[18] ZhangZ-L, ZhangT-T, LiX-Z, et al. . Bayesian ensemble methods for predicting ground deformation due to tunnelling with sparse monitoring data [J]. Underground Space, 2024, 1679-93

[19] ZhangP, YinZ-Y, JinY-F, et al. . Intelligent modelling of clay compressibility using hybrid meta-heuristic and machine learning algorithms [J]. Geoscience Frontiers, 2021, 12(1): 441-452

[20] ZhuX, ChuJ, WangK-D, et al. . Prediction of rockhead using a hybrid N-XGBoost machine learning framework [J]. Journal of Rock Mechanics and Geotechnical Engineering, 2021, 13(6): 1231-1245

[21] ZhangB-Y, LiM-Y, LiW-X, et al. . Machine learning strategies for lithostratigraphic classification based on geochemical sampling data: A case study in area of Chahanwusu River, Qinghai Province, China [J]. Journal of Central South University, 2021, 28(5): 1422-1447

[22] ZhangW-G, WuC-Z, ZhongH-Y, et al. . Prediction of undrained shear strength using extreme gradient boosting and random forest based on Bayesian optimization [J]. Geoscience Frontiers, 2021, 12(1): 469-477

[23] ShahaniN M, ZhengX-G, LiuC-C, et al. . Developing an XGBoost regression model for predicting young’s modulus of intact sedimentary rocks for the stability of surface and subsurface structures [J]. Frontiers in Earth Science, 2021, 9: 761990

[24] MaK, ShenQ-Q, SunX-Y, et al. . Rockburst prediction model using machine learning based on microseismic parameters of Qinling water conveyance tunnel [J]. Journal of Central South University, 2023, 30(1): 289-305

[25] PhoonK K, KulhawyF H. Characterization of geotechnical variability [J]. Canadian Geotechnical Journal, 1999, 36(4): 612-624

[26] PanQ-J, WuH-T, SuD, et al. . Multivariate probability distributions for index and mechanical clay parameters in Shenzhen [J]. Computers and Geotechnics, 2024, 165105934

[27] LinP-Y, ChenX-Y, JiangM-J, et al. . Mapping shear strength and compressibility of soft soils with artificial neural networks [J]. Engineering Geology, 2022, 300106585

[28] LiuH-F, LinP-Y, WangJ-Q. Machine learning approaches to estimation of the compressibility of soft soils [J]. Frontiers in Earth Science, 2023, 111147825

[29] SmitiA. A critical overview of outlier detection methods [J]. Computer Science Review, 2020, 38100306

[30] KaastraI, BoydM. Designing a neural network for forecasting financial and economic time series [J]. Neurocomputing, 1996, 10(3): 215-236

[31] CHEN Tian-qi, GUESTRIN C. Xgboost: A scalable tree boosting system [C]// Proceedings of the 22nd ACM Sigkdd International Conference on Knowledge Discovery and Data Mining. 2016: 785–794. DOI: https://doi.org/10.1145/2939672.2939785.

[32] ZhangW-G, LiH-R, WuC-Z, et al. . Soft computing approach for prediction of surface settlement induced by earth pressure balance shield tunneling [J]. Underground Space, 2021, 6(4): 353-363

[33] TURNER R, ERIKSSON D, MCCOURT M, et al. Bayesian optimization is superior to random search for machine learning hyperparameter tuning: Analysis of the black-box optimization challenge 2020 [C]// NeurIPS 2020 Competition and Demonstration Track. PMLR, 2021: 3–26. DOI: https://doi.org/10.48550/arXiv.2104.10201.

[34] SuJ, WangY-Z, NiuX-K, et al. . Prediction of ground surface settlement by shield tunneling using XGBoost and Bayesian Optimization [J]. Engineering Applications of Artificial Intelligence, 2022, 114105020

[35] HuangC-F, LiQ, WuS-C, et al. . Assessment of empirical equations of the compression index of muddy clay: sensitivity to geographic locality [J]. Arabian Journal of Geosciences, 2019, 121-13

[36] AlzabeebeeS, AlshkaneY M, RashedK A. Evolutionary computing of the compression index of finegrained soils [J]. Arabian Journal of Geosciences, 2021, 14(19): 2040

[37] ZhaoY-M, JiangH-H, ZhangH-M. Deformation parameters of Shenzhen soft clay [J]. China Railway Science, 2004, 25340-43

[38] ZhangZ-R. Correlation analysis of soil parameters in the western coastal railway of Guangdong [J]. Urban Geology, 2012, 7(1): 31-34

[39] HerreroO R. Closure to “universal compression index equation” by Oswald Rendon-Herrero (November, 1980) [J]. Journal of Geotechnical Engineering, 1983, 109(5): 755-761

[40] KalantaryF, KordnaeijA. Prediction of compression index using artificial neural network [J]. Scientific Research and Essays, 2012, 7(31): 2835-2848

[41] MayneP W. Cam-clay predications of undrained strength [J]. Journal of the Geotechnical Engineering Division, 1980, 106(11): 1219-1242

[42] ObajiN O, KomolafeO O, OkeJ A. Prediction of initial void ratio from the natural moisture content of cohesive soils [J]. Uniabuja J. Eng. Technol, 2020, 1: 1-13