SPT-based liquefaction assessment with a novel ensemble model based on GMDH-type neural network

Liquefaction is one of the most complex problems in geotechnical earthquake engineering. This paper proposes a novel ensemble group method of data handling (EGMDH) model based on classification for the prediction of liquefaction potential of soils. The database used in this study consists of 451 standard penetration test (SPT)-based case records from two major earthquakes. The input parameters are selected as SPT blow numbers, percent finest content less than 75 mu m, depth of groundwater table, total and effective overburden stresses, maximum peak ground acceleration, and magnitude of earthquake for the prediction models. The proposed EGMDH model results were also compared with other classifier models, particularly the results of the GMDH model. The results of this study indicated that the proposed EGMDH model has achieved more successful results on predicting the liquefaction potential of soils compared with the other classifier models by improving the prediction performance of GMDH model.