Abstract
Evaluation of slope stability using conventional limit equilibrium methods is very time consuming and repetitive, while the use of simplified approaches like regression analysis does not provide accurate estimation due to complexity and nonlinearities involved in the process. In such cases Artificial Neural Networks (ANNs) provide a better alternative. By proper training, an ANN with desirable transfer function and suitable number of hidden layers is able to well predict the nonlinearities and can provide accurate estimation of slope stability. However, performance of ANN in the past studies on slope stability prediction is found to be poor, while the prediction of relative importance of various slope contributing factors is not reliable. This is primarily due to the use of limited number of real field data cases and/or synthetic data covering limited parametric variations, in the training process. In the present study, an ANN has been trained using extensive synthetic dataset consisting of 15,000 cases covering wide range of soil properties & slope geometry, and then applied to the real field slopes to test its accuracy. The ANN presented here is showing significant improvement in assessing the Factor of Stability of slopes as compared to the ANN used in previous studies. The present ANN is also able to provide accurate estimation of Factor of Safety of real slopes comparable to any conventional limit equilibrium methods. Thus, ANN can be used for the estimation of Factor of Safety of real slopes, especially where it is required to estimate stability conditions rapidly such as landslide early warning, post-earthquake landslide activity, etc. Further, reliable estimates of relative importance of various contributing factors to slope stability have also been obtained, which have several applications.
Similar content being viewed by others
References
Alkasawneh W, Malkawi AIH, Nusairat JH, Albataineh N (2008) A comparative study of various commercially available programs in slope stability analysis. Comput Geotech 35:428–435
Araei AA (2013) Artificial neural networks for modeling drained monotonic behavior of rockfill materials. Int J Geomech 14(3):04014005
Arora MK, Das Gupta AS, Gupta RP (2004) An artificial neural network approach for landslide hazard zonation in the Bhagirathi (Ganga) Valley. Himal Int J Remote Sens 25(3):559–572
Bishop AW (1955) The use of the slip circle in the stability analysis of slopes. Geotechnique. 5(1):7–17
Cetina T (2014) The prediction of the critical factor of safety of homogeneous finite slopes subjected to earthquake forces using neural networks and multiple regressions. Geomech Eng 6(1):1–15
Chauhan S, Sharma M, Arora MK (2010) Landslide susceptibility zonation of the Chamoli region, Garhwal Himalayas, using logistic regression model. Landslides 7(4):411–423
Chugh AK (2003) On the boundary conditions in slope stability analysis. Int J Numer Anal Methods Geomech 27(11):905–926
Das SK (2013) 10 Artificial neural networks in geotechnical engineering: modeling and application issues. Metaheuristics Water Geotech Transp Eng 45:231–267
Das SK, Biswal RK, Sivakugan N, Das B (2011) Classification of slopes and prediction of factor of safety using differential evolution neural networks. Environ Earth Sci 64:201–210
Erzin Y, Cetin T (2012) The use of neural networks for the prediction of the critical factor of safety of an artificial slope subjected to earthquake forces. Scientia Iranica 19:188–194
Erzin Y, Cetin T (2013) The prediction of the critical factor of safety of homogeneous finite slopes using neural networks and multiple regressions. Comput Geosci 51:305–313
Fellenius, W. 1936 Calculation of the stability of earth dams. In Proceedings of the 2nd nternational Congress on Large Dams. Washington, D.C, p. 445
Funahashi K-I (1989) On the approximate realization of continuous mappings by neural networks. Neural Netw 2:183–192
Garson GD (1991) Interpreting neural-networks connection weights. AI Expert 6:47–51
Guide MUS (2002) Neural network toolbox. The MathWorks.
GeoStudio (2012) User’s manual. Geo-slope international Ltd., Calgary, Alberta, Canada.
Goh AT (1994) Seismic liquefaction potential assessed by neural networks. J Geotech Eng 120:1467–1480
Goh AT (1996) Neural-network modeling of CPT seismic liquefaction data. J Geotech Eng 122(1):70–73
Gupta R, Goyal K, Yadav N (2015) Prediction of safe bearing capacity of noncohesive soil in arid zone using artificial neural networks. Int J Geomech. https://doi.org/10.1061/(ASCE)GM.1943-5622.0000514
Janbu N (1975) Slope stability computations. In: RC Hirschfeld and SJ Poulos (Eds). Embankment-dam Engineering Textbook, John Wiley and sons inc pub, NY. International Journal of Rock Mechanics and Mining Sciences and Geomechanics Abstracts, Pergamon, pp 67
Kanungo DP, Arora MK, Sarkar S, Gupta RP (2006) A comparative study of conventional, ANN black box, fuzzy and combined neural and fuzzy weighting procedures for landslide susceptibility zonation in Darjeeling Himalayas. Eng Geol 85(3):347–366
Lu P, Rosenbaum, MS (2003) Artificial neural networks and grey systems for the prediction of slope stability. Nat Hazards, 30(3):383–398
Manual GUS (2012) Geo-slope international Ltd. Calgary, Alberta, Canada T2P 2Y5
Mansour ZS, Kalantari B (2011) Traditional methods vs. finite difference method for computing safety factors of slope stability. Electron J Geotech Eng 16:1119–1130
Mayoraz F, Vulliet L (2002) Neural Networks for Slope Movement Prediction. Int J Geomech 2:153–173
Michalowski RL (2002) Stability charts for uniform slopes. J Geotech Geoenviron Eng 128:351–355
Morgenstern NR, Price VE (1965) The analysis of the stability of general slip surfaces. Géotechnique 15:79–93
Neaupane KM, Piantanakulchai M (2006) Analytic network process model for landslide hazard zonation. Eng Geol 85(3):281–294
Padmini D, Ilamparuthi K, Sudheer K (2008) Ultimate bearing capacity prediction of shallow foundations on cohesionless soils using neurofuzzy models. Comput Geotech 35:33–46
Paola JD, Schowengerdt RA (1995) A review and analysis of backpropagation neural networks for classification of remotely-sensed multi-spectral imagery. Int J Remote Sens 16(16):3033–3058
Rashidian V, Hassanlourad M (2013) Application of an artificial neural network for modeling the mechanical behavior of carbonate soils. Int J Geomech 14(1):142–150
Sah N, Sheorey P, Upadhyaya L (1994) Maximum likelihood estimation of slope stability. Int j rock mech min sci geomech abstr 31(5):47–53
Sah NK, Sheorey PR, Upadhyaya LN (1994) Maximum likelihood estimation of slope stability. In Int J Rock Mech Mining Sci Geomech Abstr (Vol. 31, No. 1, pp. 47–53), Pergamon.
Sakellariou M, Ferentinou M (2005) A study of slope stability prediction using neural networks. Geotech Geol Eng 23:419–445
Spencer E (1967) A method of analysis of the stability of embankments assuming parallel interslice forces. Géotechnique 17:11–26
Steward T, Sivakugan N, Shukla S, Das B (2010) Taylor’s slope stability charts revisited. Int J Geomech 11:348–352
Suman S, Khan SZ, Das SK, Chand SK (2016) Slope stability analysis using artificial intelligence techniques. Nat Hazards 84(2):727–748
Tiwari RC (2014) Simplified numerical implementation in slope stability modeling. Int J Geomech 15(3):04014051
Acknowledgements
Authors are thankful to Mrs Anjana Kukunuri for providing valuable inputs & suggestions in the work carried out as well as in the preparation of the manuscript.
Author information
Authors and Affiliations
Corresponding authors
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix: Calculation of Relative Importance Factors
Appendix: Calculation of Relative Importance Factors
The residual interconnection weights between the neurons obtained after successful training of ANN with the huge data (Sect. 5) is provided in the following Table
9. This information is used further to calculate importance factors of various parameters contributing slope stability based on the method suggested by Garson (1991).
The absolute value of interconnection weights between each hidden neuron and input neurons is multiplied by the corresponding absolute value of interconnection weight between the hidden and output neuron. These connection weights products are given in the Table
10.
Each connection weight product of hidden neuron is normalized with respect to sum of the all connection weight products of the corresponding hidden neuron. The normalized connection weight products are given in Table
11.
The relative importance of each input parameter is obtained by adding the normalized connection weight products of that particular input neuron to all the hidden neurons (See Table
12). These importance factors are further expressed in percentages.
Rights and permissions
About this article
Cite this article
Marrapu, B.M., Kukunuri, A. & Jakka, R.S. Improvement in Prediction of Slope Stability & Relative Importance Factors Using ANN. Geotech Geol Eng 39, 5879–5894 (2021). https://doi.org/10.1007/s10706-021-01872-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10706-021-01872-2