Probability-Based Method for Assessing Liquefaction Potential of Soil Using Genetic Programming

Abstract

Liquefaction of soil due to earthquake is one of the major causes for the significant damages to the buildings, lifeline systems, and harbor facilities. Probabilistic method is now being preferred over deterministic method due to uncertainty in soil and seismic parameter. At present, artificial intelligence techniques such as artificial neural network (ANN) and support vector machine (SVM) based models are found to be more efficient compared to statistical methods. In the present study, an attempt has been made to develop a limit state function for assessing the cyclic resistance ratio (CRR) of soil based on cone penetration test (CPT) data obtained after Chi-Chi earthquake, Taiwan, 1999, using evolutionary artificial intelligence technique, genetic programming (GP), and to evaluate the liquefaction potential of soil in a probabilistic approach through a Bayesian mapping function. A comparative evaluation of the present study is made with three existing CPT-based statistical methods for prediction of liquefied and non-liquefied cases in terms of percentage success rate with respect to the field manifestations.