Amplification factor is defined as the ratio of the spectral acceleration at the soil surface to the spectral acceleration at bedrock in various periods. The effects of site conditions at the time of the earthquake are often estimated by amplification factor. Site amplification models are empirical relations to predict the amplification factor based on site characteristics. These models are used in ground motion prediction equations (GMPEs). In most models, the average shear wave velocity to the depth of 30 m (VS30) and the depth of the bedrock (Z1.0) are used to predict the amplification factor. However, these parameters are not sufficient in describing site effects especially at high strain level. In this paper, soil plasticity index (PI) has been studied as a soil physical characteristic for the development of these models using site response analysis. PI is the parameter that plays an important role in the formability of soft soils. Analyzing site responses showed that the average plasticity index in the first 30 m (PI30m) as a characteristic of soft soil plasticity is a parameter affecting the amplification factor in the nonlinear zone in short periods (T ≤ 1 s). So, higher PI30m leads to increasing of the amplification factor. The analysis of all responses also showed that the effects of PI30m on the amplification factor increase with a decrease in VS30 and an increase in input acceleration in the bedrock. Former amplification models using genetic algorithms have been developed to study the effects of PI30m. This model has been presented in terms of VS30, PI30m, and input acceleration in bedrock.
