Thottoth, S.R.Khatri, V.N.Kolathayar, S.Keawsawasvong, S.Lai, V.Q.2026-02-042024Geotechnical and Geological Engineering, 2024, 42, 5, pp. 3307-33299603182https://doi.org/10.1007/s10706-023-02731-yhttps://idr.nitk.ac.in/handle/123456789/21048This study comprehensively analyzes seismic active earth pressure estimation for hunched retaining walls. The analysis utilizes the horizontal slices method within the modified pseudo-dynamic framework and incorporates depth-dependent dynamic parameters for the backfill soil. The friction angle of the backfill soil varied between 30° and 45°, while the hunch angle of the retaining wall increased from 0° to 20°. The findings of this study demonstrate that the use of hunched retaining walls results in a significant reduction in active earth pressure. In both static and dynamic cases, reductions of up to 23% and 18%, respectively, compared to vertical walls, were observed. Notably, this reduction is more pronounced for smooth walls under static conditions than for rough walls under dynamic conditions. The estimated active earth pressures for both vertical and hunched walls in static and dynamic scenarios closely align with those reported in the literature. Additionally, an empirical equation based on an artificial neural network model, utilizing the numerical analysis result, is proposed to establish a relationship between the investigated design parameters and the active earth pressure coefficient. The proposed equation demonstrated a high coefficient of determination (R2) value of 99.78% when compared to the numerical results. This study’s outcomes provide valuable insights and a tool for practicing engineers in the field. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2024.Neural networksNumerical methodsPressure distributionSeismology% reductionsActive earth pressureANNHorizontal slicesHunchbacked retaining wallMethod of horizontal sliceModified pseudo dynamic methodPseudo-dynamic methodSeismic earth pressureStatics and dynamicsRetaining wallsartificial neural networkearth pressureestimation methodparameter estimationretaining wallseismic dataseismic source