Of the various seismic risks, the liquefaction of the soil is one of the main causes of loss of life and damage to infrastructures and life-saving systems. Soil liquefaction phenomena have been observed in many historical earthquakes following the first large-scale observations of liquefaction damage in the 1964 earthquakes in Niigata, Japan and 1964 in Alaska, USA. Because of the difficulty in obtaining high quality undisturbed samples and the cost involved in them, geotechnical engineers prefer in situ tests, standard penetration test (SPT) and cone penetration test (CPT) to evaluate the potential of Liquefaction with the limited use of other init tests such as shear wave velocity tests and Baker penetration tests. Liquefaction evaluation in the deterministic setting is preferred by geotechnical engineering professionals because of its simple mathematical approach with a minimum of data, time and effort. However, for important lifeline structures, there is a need for probabilistic and reliability methods to make risk-based design decisions. In recent years, soft computing techniques, such as the artificial neural network (RNA), the supporting vector machine (SVM), and the vector machine of relevance (RVM) have been successfully implemented to assess the liquefaction potential with a higher accuracy compared to the statistical methods available. In the recent past, evolutionary genetic programming of the soft computing (GP) technique based on the Darwinian theory of natural selection is being used as an alternative soft computing technique.
The shear wave velocity (VS) provides a means to determine soil seismic resistance to liquefaction by a fundamental soil property. The method of Iwasaki (1982) is used to measure the rate of liquefaction potential for both. It follows the general format of the simplified Seed-Idriss (1685) procedure based on the standard penetration test blow rate and the shear wave velocity (VS) based on Andrus et al. (2004), using data from 43 wells in soils ranging from fine sand, silty sand, grave sand to profiles including silty clay layers and the mean shear wave velocity (VS30) in southern Tehran . The liquefaction resistance curves were established by applying a modified ratio between the shear wave velocity and the cyclic stress ratio for the constant mean cyclic shear strain. The study area is the southeast of Tehran and the route of line 7 of the Tehran metro. It is observed that there is no perfect agreement between the results of the two methods based on five empirical relationships assuming cemented and non cemented conditions for soils. In addition, the value of the liquefaction potential index (PL) in the standard penetration test method (SPT) is more than that of the VS method. The values of the liquefaction potential index (PL) based on the shear wave velocity (VS) using five empirical relationships in two cementless and cemented soils show that the relationships used are overestimated and most of them have shown a no liquefaction for soils in the studied area.