Analysis of sandy soil liquefaction by reliability method and genetic algorithm

Document Type : Original Article


Department of Civil Engineering, Faculty of Engineering, Islamic Azad University of Arak, Arak, Iran



Probability of liquefaction (PL) of sandy soil is based on deterministic analysis to investigate changes in soil resistance and loads caused by dynamic loads such as earthquakes. In addition, there is no possibility to establish a relationship between PL and the factor of safety for the use of these methods. To solve this problem, the statistical analysis and reliability analysis were used. In this article, the reliability analysis was based on the analysis of PL by Seed et al. (1985) and the recommendations of the National Center for Research on Earthquake Engineering (NCEER, 1998). Since the use of this method requires the quantification of uncertainty values in calculations, it is possible to calculate the most probable values related to their uncertainty using statistical moments related to non-deterministic parameters such as mean, variance, standard deviation and probability density function. Then, using the advance first order second moment (AFOSM), the reliability index was calculated by the new method of genetic algorithm and the relationship between PL, the reliability index and the factor of safety was established. For the performance of the proposed method, PL in line 7 of Tehran Metro has been studied based on the results of standard penetration test (SPT) by MATLAB. The results showed the efficiency of the proposed method.