Prediction of Residual Strength After Liquefaction Using Artificial Intelligence Model

This research aims to develop a hybrid artificial intelligence model to predict the residual strength required to resist soil movement after post-liquefaction. The model is trained using available case history and experimental data, with a focus on soil parameters such as standard penetration test, cone penetration test resistance, percentage fine, void ratio, relative density, and pore water pressure. Detailed statistical analysis of the model is conducted using previous case histories to assess its accuracy. The practical implications of this research lie in the challenge of having to extrapolate beyond available data for flow failures and lateral spreading after liquefaction. By providing a reliable prediction model for residual strength, this paper offers a valuable tool for geotechnical engineers and practitioners to assess the stability of soil and mitigate risks associated with soil movement after post-liquefaction.