maximum lateral displacement prediction of bored pile wall in granular soil using gene expression programming

 bored pile walls are one of the most common structures in soil-retaining geotechnical projects. they must be designed for lateral loads due to earth pressure, earthquake. the maximum horizontal displacement at the top of these piles is one of the most important parameters in design because of workability of itself and adjacent structures. in this paper, gene expression programming is employed as an artificial intelligence method for developing a model to predict the maximum horizontal displacement of piles in granular soils. the input parameter consists of pile spacing, excavation depth, excavation length, soil unit weight, elastic module and friction angle of granular soil. the output consists of the maximum lateral displacement of pile. a database obtained from 3d modeling of a soil-bored pile wall modeled by abaqus is employed to develop the model by gene expression programing (gep).the comparison of the model prediction with abaqus data indicates that the proposed approach is accurate and robust in representing the maximum lateral displacement of pile in granular soils.