Reliability assessment of liquefaction potential using the jointly distributed random variables method

Abstract

Prediction of liquefaction potential is a complex engineering problem due to heterogeneous nature of soils and participation of a large number of factors involved. The inherent uncertainties of the parameters which affect liquefaction potential dictate that problem is of a probabilistic nature rather than being deterministic. Probabilistic analysis of liquefaction potential has received considerable attention in the literature and has been used as an effective tool to evaluate uncertainty so prevalent in variables. In this research the jointly distributed random variables method is used for probabilistic analysis and reliability assessment of liquefaction potential based on triaxial test results. The selected stochastic parameters are relative density which is modeled using a truncated normal probability distribution function and the earthquake acceleration ratio and earthquake magnitude which are considered to have exponential probability distribution function. The depth is regarded as a constant parameter. The results are compared with the Monte Carlo simulation. Comparison of the results indicates very good performance of the proposed approach for assessment of reliability. The results of sensitivity analysis show that relative density is the most effective parameter in factor of safety against liquefaction.