Bayesian adaptive LASSO seismic liquefaction discrimination model incorporating soil classification and model uncertainty

Abstract

Soil classification and model uncertainty significantly affect the accuracy of seismic liquefaction discrimination models. In this study, we developed a Bayesian Logistic Regression (BLR) model for soil classification and a Bayesian Adaptive Least Absolute Shrinkage and Selection Operator Logistic Regression (BALASSO-LR) model for liquefaction discrimination based on a cone penetration test with pore pressure measurement data. We aimed to evaluate the effects of two classification strategies, i.e., one-versus-one (OvO) and one-versus-rest (OvR), on the performance of a BLR model. Additionally, we evaluated the influence of liquefaction-related factors, including the soil behavior type index, I_c, and performed model uncertainty analysis to enhance the predictive reliability of the BALASSO-LR model. These models were then compared with the conventional simplified soil behavior classification method and the existing logistic regression (LR) models for liquefaction prediction. The results indicated that the BLR soil classification model using the OvR scheme achieved a prediction accuracy of 81.2%, representing improvements of 16.2% and 1.9% over the conventional Soil Behavior Type (SBT) method and the OvO scheme-based BLR model, respectively. The BALASSO-LR model for liquefaction discrimination achieved an accuracy of 84.1%. Omitting the soil classification index I_c decreased accuracy by 2.3%. When compared with the three simplified methods and existing LR models, BALASSO-LR exhibited an improvement of 5.7%-11.4% in the accuracy. In the uncertainty analysis, the No-U-Turn sampler (NUTS) algorithm with a prior distribution of N∼ (0,100) achieved the highest accuracy (84.1%), surpassing that of the Metropolis algorithm by 6%.

Keywords

Soil classification / Liquefaction assessment / Bayesian logistic regression / Adaptive LASSO / Model uncertainty

References

[1] Y. Sonmezer, A. Akyuz, K. Kayabali, Investigation of the effect of grain size on liquefaction potential of sands, Geomech. Geoeng. 20 (2020) 243-254, doi: 10.12989/gae.2020.20.3.243.

[2] L. Guo, Y. Xu, H. Li, L. Lu, W. Xu, P. Liang, Using keyhole images to map soil liq- uefaction induced by the 1966 Xingtai Ms 6.8 and 7.2 earthquakes, North China, Remote Sens 15 (2023) 5777, 10.3390/rs15245777.

[3] C. Li, Z. Li, X. Yuan, Simplified prediction method for regional seismic soil liquefac- tion, Chin. J. Geophys. 63 (2020) 2084-2095, doi: 10.6038/cjg2020N0052.

[4] H. Yang, C. Chang, Suggested earthquake insurance claim evaluation criteria for building damage caused by soil liquefaction in Taiwan, Earthq. Spectra 38 (2022) 2866-2885, doi: 10.1177/87552930221106234.

[5] D. Lin, Q. Zhang, K. Xia, H. Peng, Q. Chen, Construction of CPT-based liquefaction database from 2003 Bachu Earthquake, Technol. Earthq. Disaster Prev. 18 (2023) 833-845, doi: 10.11899/zzfy20230417.

[6] Z. Liu, P. Wang, Z. Zhang, Z. Li, Z. Cao, J. Zhang, X. Yuan, W. Wei, X. Xing, Liq- uefaction in western Sichuan basin during the 2008 Mw 7.9 Wenchuan earthquake, Tectonophysics 694 (2017) 214-238, doi: 10.1016/j.tecto.2016.11.001.

[7] L. Paolella, R. Spacagna, G. Chiaro, G. Modoni, A simplified vulnerability model for the extensive liquefaction risk assessment of buildings, Bull. Earthq. Eng. 19 (2021) 3933-3961, doi: 10.1007/s10518-020-00911-2.

[8] Z. Li, X. Yuan, Seismic damage summarization of site effect and soil liquefac- tion in 2016 Kaohsiung earthquake, Earthq. Eng. Eng. Dyn. 36 (3) (2016) 1-7, doi: 10.13197/j.eeev.2016.03.1.lizy.001.

[9] R.S. Olsen, Cyclic liquefaction based on the cone penetrometer test, NCEER 32 (1997) 110-123 https://escholarship.org/uc/item/8xw0p63c.

[10] P.K. Robertson, C.E. Wride, Evaluating cyclic liquefaction potential using the cone penetration test, Can. Geotech. J. 35 (1998) 442-459, doi: 10.1139/t98-017.

[11] Z. Zhang, M.T. Tumay, Simplification of soil classification charts derived from the cone penetration test, Geotech. Test. J. 19 (1996) 203-216, doi: 10.1520/GTJ10342J.

[12] P.K. Robertson, Soil classification using the cone penetration test, Can. Geotech. J. 27 (1990) 151-158, doi: 10.1139/t90-014.

[13] J.A. Schneider, M.F. Randolph, P.W. Mayne, N.R. Ramsey, Analysis of fac- tors influencing soil classification using normalized piezocone tip resistance and pore pressure parameters, J. Geotech. Geoenviron. Eng. 134 (2008) 1569-1586, doi: 10.1061/(ASCE)1090-0241(2008)134:11(1569).

[14] D.N.K. Pandiri, R. Murugan, T. Goel, Smart soil image classification system using lightweight convolutional neural network, Expert Syst. Appl. 238 (2024) 122185, doi: 10.1016/j.eswa.2023.122185.

[15] S.A. Al-Haddad, L.A. Al-Haddad, A.A. Jaber, Environmental engineering solu- tions for efficient soil classification in southern Syria: a clustering-correlation ex- treme learning approach, Int. J. Environ. Sci. Technol. 22 (2025) 2177-2190, doi: 10.1007/s13762-024-05784-5.

[16] H.B. Seed, I.M. Idriss, Simplified procedure for evaluating soil liquefaction potential, J. Soil Mech. Found. Div. 97 (1971) 1249-1273, doi: 10.1061/JSFEAQ.0001662.

[17] J. Wang, J. Hwang, Y. Deng, C. Lu, Model uncertainties of SPT, CPT, and vs-based simplified methods for soil liquefaction assessment, Bull. Eng. Geol. Environ. 82 (2023) 260, 10.1007/s10064-023-03300-6.

[18] D. Dey, M.S. Haque, M.M. Islam, U.I. Aishi, S.S. Shammy, M.S.A. Mayen, S.T.A. Noor, M.J. Uddin, The proper application of logistic regression model in complex survey data: a systematic review, BMC Med. Res. Methodol 25 (2025) 15, 10.1186/s12874-024-02454-5.

[19] S. Liao, D. Veneziano, R.V. Whitman, Regression models for evalu- ating liquefaction probability, J. Geotech. Eng.114 (1988) 389-411, doi: 10.1061/(ASCE)0733-9410(1988)114:4(389).

[20] Y. Yang, Y. Wei, A new shear wave velocity-based liquefaction probability model us- ing logistic regression: emphasizing fines content optimization, Appl. Sci. 14 (2024) 6793, 10.3390/app14156793.

[21] X. Yao, L. Liu, Z. Wang, A Vs-based logistic regression method for liquefaction eval- uation, Adv. Civ. Eng. 2021 (2021) 5535387, 10.1155/2021/5535387.

[22] N. Pirhadi, J. Hu, Y. Fang J. Idriss, X. Wan, J. Lu, Seismic gravelly soil liquefaction assessment based on dynamic penetration test using expanded case history dataset, Bull. Eng. Geol. Environ. 80 (2021) 8159-8170, doi: 10.1007/s10064-021-02423-y.

[23] G. Saulière, J. Dedecker, L.A. Marquet, P. Rochcongar, J.F. Toussaint, G. Berthelot, Z-scores-based methods and their application to biological monitoring: an example in professional soccer players, Biostatistics 20 (2019) 48-64, doi: 10.1093/biostatis-tics/kxx044.

[24] M.B. Pereira, M. Wallroth, V. Jonsson, E. Kristiansson, Comparison of normalization methods for the analysis of metagenomic gene abundance data, BMC Genomics 19 (2018) 274, 10.1186/s12864-018-4637-6.

[25] J. Zhou, G. Claeskens, A tradeoff between false discovery and true positive propor- tions for sparse high-dimensional logistic regression, Electron. J. Stat. 18 (2024) 395-428, doi: 10.1214/23-EJS2204.

[26] J. Xia, Z. Yang, S. Li, S. Zhang, Y. Fu, D.N. Gunduz, X. Li, Blind super-resolution via meta-learning and Markov chain Monte Carlo simulation, IEEE Trans. Pattern Anal. Mach. Intell. 46 (2024) 8139-8156, doi: 10.1109/TPAMI.2024.3400041.

[27] Y. Zhao, X. Huo, A survey of numerical algorithms that can solve the Lasso problems, Wiley Interdiscip. Rev. Comput. Stat. 15 (2023) e1602, doi: 10.1002/wics.1602.

[28] S. Pirenne, G. Claeskens, Parametric programming-based approximate selective in- ference for adaptive lasso, adaptive elastic net and group lasso, J. Stat. Comput. Simul. 94 (2024) 2412-2435, doi: 10.1080/00949655.2024.2337342.

[29] A. Gunasekaran, Z. Irani, K.L. Choy, L. Filippi, T. Papadopoulos, Performance mea- sures and metrics in outsourcing decisions: A review for research and applications, Int. J. Prod. Econ. 161 (2015) 153-166, doi: 10.1016/j.ijpe.2014.12.021.

[30] P. Zuo, Study on uncertainty of soil layers and pile bearing effect based on SCPTu M. S. Thesis, Jiangsu University, 2020, doi: 10.27170/d.cnki.gjsuu.2020.002116.

[31] G. BAI, Study on engineering application of soil classification and pile bearing char- acteristics based on CPTu M. S. Thesis, Jiangsu University, 2017.

[32] R.E. Moss, R.B. Seed, R.E. Kayen, J.P. Stewart, A.D. Kiureghian, K.O. Cetin, CPT-based probabilistic and deterministic assessment of in situ seismic soil liquefaction potential, J. Geotech. Geoenviron. Eng. 132 (2006) 1032-1051, doi: 10.1061/(ASCE)1090-0241(2006)132:8(1032).

[33] R.E. Moss, R.E. Kayen, L. Tong, S. Liu, G. Cai, J. Wu, Reinvestigation of liquefaction and nonliquefaction case histories from the 1976 Tangshan earthquake, Civ. Envi- ron. Eng. 137 (2009) 334-343 https://digitalcommons.calpoly.edu/cenv_fac/174.

[34] C.H. Juang, C. Chen, P.W. Mayne, CPTU simplified stress-based model for evaluating soil liquefaction potential, Soil. Found 48 (2008) 755-770, doi: 10.3208/sandf.48.755.

[35] P. Njock, S.L. Shen, A. Zhou, H.M. Lyu, Data on a coupled ENN /t-SNE model for soil liquefaction evaluation, Data Brief. 29 (2020) 105-125, doi: 10.1016/j.dib.2020.105125.

[36] C.H. Juang, T. Jiang, R.D. Andrus, Assessing probability-based methods for lique- faction potential evaluation, J. Geotech. Geoenviron. Eng. 128 (2002) 580-589, doi: 10.1061/(ASCE)1090-0241(2002)128:7(580).

[37] M.G. Jefferies, M.P. Davies, Use of CPTu to estimate equivalent SPT N60, Geotech. Test. J. 16 (1993) 458-468, doi: 10.1520/GTJ10286J.

[38] J. Hu, J. Wang, Z. Zhang, H. Liu, Continuous-discrete hybrid bayesian network mod- els for predicting earthquake-induced liquefaction based on the Vs database, Com- put. Geosci. 169 (2022) 105231, doi: 10.1016/j.cageo.2022.105231.

[39] G. Cai, S. Liu, Identification of soil strata based on general regression neural network model from CPTu data, Mar. Georesour. Geotechnol. 33 (2014) 2-5, doi: 10.1080/1064119X.2013.843046.

[40] W. Duan, G. Cai, S. Liu, Z. Zhao, S. Pu, S. Liu, X. Dong, M. Wu, R. Chen, Unified evaluation method for soil liquefaction potential based on modern in- situ piezocone penetration tests, Chin. J. Geotech. Eng. 44 (2022) 435-443, doi: 10.11779/CJGE202203005.

[41] H.B. Seed, I.M. Idriss, Simplified procedure for evaluating soil liq- uefaction potential, J. Soil Mech. Found. Div. 97 (1971) 1249-1273 https://scholarsmine.mst.edu/icrageesd/04icrageesd/session14/2.

[42] N. Pirhadi, X. Wan, J. Lu, Y. Fang, I. Jasiri, J. Hu, DPT-based seismic liquefaction triggering assessment in gravelly soils based on expanded case history dataset, Eng. Geol. 311 (2022) 106894, doi: 10.1016/j.enggeo.2022.106894.

[43] A. Brown, GL. Jones, Convergence rates of metropolis-hastings algorithms, Wiley Interdiscip. Rev. Comput. Stat. 16 (5) (2024) e70002, doi: 10.1002/wics.70002.

[44] L. Devlin, M. Carter, P. Horridge, P.L. Green, S. Maskell, The no-U-turn sam- pler as a proposal distribution in a sequential Monte Carlo sampler without ac- cept/reject, IEEE Signal Process. Lett. 31 (2024) 1089-1093, doi: 10.1109/LSP.2024.3386494.

[45] T. Yu, G. Yu, P.Y. Li, L. Wang, Citation impact prediction for scientific pa- pers using stepwise regression analysis, Scientometrics 101 (2014) 1233-1252, doi: 10.1007/s11192-014-1279-6.

[46] Z. Zhang, G. Wang, E.J.M. Carranza, J. Fan, X. Liu, X. Zhang, Y. Dong, X. Chang, D. Sha, An integrated framework for data-driven mineral prospec- tivity mapping using bagging-based positive-unlabeled learning and Bayesian cost-sensitive logistic regression, Nat. Resour. Res. 31 (2022) 3041-3060, doi: 10.1007/s11053-022-10120-0.

[47] O.A. Montesinos López, A.Montesinos López, J. Crossa, Overfitting, model tuning, and evaluation of prediction performance, in: Multivariate statistical machine learn- ing methods for genomic prediction, Springer International Publishing, Cham, 2022, pp. 109-139, doi: 10.1007/978-3-030-89010-0_4.

[48] P. Abbasimaedeh, Soil liquefaction in seismic events: pioneering predictive models using machine learning and advanced regression techniques, Environ. Earth Sci. 83 (2024) 189, doi: 10.1007/s12665-024-11480-x.

[49] M. Al-Arafat, M.Enamul Kabir, A. Dasgupta, O.F. Nahid, Designing earthquake- resistant foundations: a geotechnical perspective on seismic load distri- bution and soil-structure interaction, Avijit Nahid, Omar Faruq (2024) https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5113708.