New explicit formulation for ultimate bearing capacity of shallow foundations on granular soil using M5' model tree. (15th October 2020)

• Record Type: Journal Article
• Authors: Khorrami, Rouhollah; Derakhshani, Ali; Moayedi, Hossein
• Is Part Of: Measurement. Volume 163(2020)
• Publication Date: 2020-10-15
• Subjects: Bearing capacity -- Shallow foundation -- Computational intelligence -- Probabilistic design; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2020.108032
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: M5' Model Tree was used to estimate ultimate bearing capacity of shallow foundations. Ultimate limit states were derived from high-precision power-law functions framework. Explicit non-dimensional expressions were also developed to be easily used in practice. Proposed model outperforms conventional theories & existing soft computing approaches. It is demonstrated that the new model has lower uncertainty compared to other models. Abstract: Proper estimation of the ultimate bearing capacity of the shallow foundations is of vital significance as a necessary part of foundation design for different structures. This study aims at developing explicit and high-precision expressions for predicting the ultimate limit states of shallow foundations under vertical loading over a practical range of predictors including foundation width ( B ), foundation length ( L ), embedment depth ( D ), internal friction angle ( φ ) and specific gravity ( γ ). A more comprehensive database of the ultimate bearing capacity test results, in contrast to previous studies, were considered for modeling using the M5' Model Tree algorithm. Predictions are compared with measurements, conventional bearing capacity theories and available soft computing approaches. Moreover, it is shown that the new model has lower uncertainty compared with other models. It is concluded that the ultimate bearing capacity of a shallow footing can be derived from power-law functions framework and also non-dimensional equations in practice.