Abstract:

The binaIγregression method was used to derive two logarithmic models based on 278 cases of liquefac-tion or nonliquefaction occurrences. In the two proposed models , the liquefaction probability was regarded as a function of earthquake load and soil resistance parameters , and the liquefaction resistance was measured through the tested values of the corrected standard penetration. Comparisons between these two logarithmic models and deter-ministic criteria were then performed. Moreover , the two models were applied to the analysis of typical sites. It is found that the two logarithmic models can objectively evaluate the information contained in the data and can be ap-plied to the direct quantification of conditional liquefaction probability based on the data.

Key words: liquefaction, estimation, regression analysis, logistic model, deterministic criterion