AI-Integrated Comparative Seismic Foundation Design Using Indian Standards and Eurocodes: Ensemble Learning-Based Prediction of Bearing Capacity and Settlement

The present paper concludes that an AI-integrated comparative framework can improve the reliability, efficiency, and technical clarity of seismic foundation design. The study successfully combines Indian Standards, Eurocode provisions, and ensemble machine learning models to evaluate bearing capacity and settlement under seismic conditions. The comparative results indicate that bearing capacity variation with respect to Indian Standard results remains moderate, with a mean variation of 8.09% and maximum variation of 24.94%, showing acceptable agreement among IS, Eurocode, and AI-based outputs. However, settlement variation is comparatively higher, with a mean value of 36.58% and maximum value of 64.57%, confirming that settlement prediction is more sensitive to soil stiffness, groundwater condition, deformation behavior, and serviceability assumptions. The AI model performs better in bearing capacity prediction because strength and geometry parameters have a more direct influence, while settlement involves complex nonlinear deformation mechanisms. The use of Random Forest and XGBoost provides a practical decision-support layer by validating codal results and identifying conservative design values. Overall, the proposed framework is useful for preliminary seismic foundation design, safety checking, and optimization, while maintaining engineering judgment and codal compliance as the primary basis of final design.

[1] Sadegh Es-haghi, M., Abbaspour, M., Abbasianjahromi, H., & Mariani, S. (2021). Machine learning-based prediction of the seismic bearing capacity of a shallow strip footing over a void in heterogeneous soils. Algorithms, 14(10), 288. https://doi.org/10.3390/a14100288 [2] Nazeer, S., & Dutta, R. K. (2021). Application of machine learning techniques in predicting the bearing capacity of E-shaped footing on layered sand. Journal of Soft Computing in Civil Engineering, 5(4), 74-89. https://doi.org/10.22115/SCCE.2021.303113.1360 [3] Pooya Nejad, F. (2009). Prediction of pile settlement using artificial neural networks based on standard penetration test data. Computers and Geotechnics, 36(7), 1125-1133. https://doi.org/10.1016/j.compgeo.2009.04.003 [4] Nikolaou, K. (2017). SoFA: A MATLAB-based educational software for the shallow foundation analysis and design. Computer Applications in Engineering Education, 25(2), 284-292. https://doi.org/10.1002/cae.21791 [5] Khan, S. R., Khan, A., & Khan, M. A. (2022). Optimization-based design of shallow foundations satisfying serviceability and ultimate limit states. Sustainability, 14(9), 5699. https://doi.org/10.3390/su14095699 [6] Sarfarazi, V. (2024). Bearing capacity and settlement behavior of surface foundations above underground voids. Scientific Reports, 14, 72220. https://doi.org/10.1038/s41598-024-72220-6 [7] Ghazavi, M., & Dehkordi, M. R. (2021). Interference effects between closely spaced shallow footings: A comprehensive review. Engineering Geology, 286, 106073. [8] Indian Standards Institution. IS 1893 (Part 1): Criteria for Earthquake Resistant Design of Structures. Bureau of Indian Standards, New Delhi. [9] Bureau of Indian Standards. IS 6403: Code of Practice for Determination of Bearing Capacity of Shallow Foundations. New Delhi. [10] Bureau of Indian Standards. IS 8009: Code of Practice for Calculation of Settlements of Foundations. New Delhi.