An extreme learning machine model for geosynthetic-reinforced sandy soil foundations

In the past, several experimental and theoretical studies have been carried out to evaluate the ultimate bearing capacity (UBC) of geosynthetic-reinforced sandy soil foundations (GRSSFs). The experimental studies consist of model footing load tests which are expensive and time consuming whereas the results obtained by theoretical expressions often lack consistency. In the study reported in this paper, a cost-effective, extreme learning machine (ELM) model was used for the first time to obtain a more realistic prediction of the UBC of a GRSSF. A large dataset consisting of actual field and laboratory measurements of UBC was used to develop and validate the model. Its predictive performance was then compared against robust machine learning regression models and traditional theoretical methods. The study shows that the proposed model is useful and attains an adequate level of accuracy in predicting the UBC of GRSSFs when compared with other data-driven models and some traditional methods. The research also shows that the ELM technique is a realistic and reliable approach that could be employed in geotechnical engineering intelligent systems for the prediction of multivariate non-linear problems.