Comparative performance evaluation of machine learning models for predicting the ultimate bearing capacity of shallow foundations on granular soils

In this paper, a probabilistic distribution for the bearing capacity and safety factor of shallow foundations is proposed to account for the variability and randomness of the soil strength properties and applied loads. A probabilistic-based model is developed to assess the bearing capacity of shallow foundations. A Monte Carlo simulation is performed to infer probabilistic descriptions of the bearing capacity of shallow foundations. The effects of the variation in strength and load random variables on the variation of the bearing capacity and the safety factor are studied. The generalized extreme value reduced to type II extreme distribution was proved to be best suited in describing the variability in both the bearing capacity of shallow foundations and the safety factor. The reliability index and the deterministic safety factor are compared. A risk-based safety factor for the ultimate bearing capacity of shallow foundations is proposed and assessed.

Ultimate bearing capacity is one of the most important parameters in designing shallow foundations. This study focused on developing a hybrid model using Random Search (RS) technique and Deep Neural Network (DNN) to predict the maximum bearing capacity of shallow foundations in sandy soil. The data included 97 load tests on the steps that were used to train and test the model. This data is divided into two parts of the training data set (7%) and the testing set (30%) to build and validate the corresponding models. The performance of the final DNN model is comprehensively assessed with a random hyper-parameters DNN model developed independently using the same data. The values of performance evaluation measures such as R-squared (R2), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE) and the Variance Accounted For (VAF) are used to determine to get the performance of the DNN model in predicting the ultimate bearing capacity of shallow foundations. In addition, a parallel coordinate plot is utilized to show and evaluate the effect of hyperparameters combination on the performance of DNN model. Besides, a global sensitivity analysis technique was deployed to detect the most important input variables in predicting the ultimate bearing capacity of shallow foundations. This study can provide an effective tool to identify the ultimate bearing capacity of shallow foundations.

Groundwater variation has a significant effect on the bearing capacity of sandy shallow foundations. Groundwater and capillary water in the shallow foundation would result in the various water distributions in the soil mass. Therefore, there are three types of water conditions in the shallow foundation. They are the total saturated, capillary-water-effect zone and dry soil. In this study, a physical mode experimental was developed to investigate the effect of groundwater variation on the deformation behavior under different loading conditions. The effect of water level and fluctuation times were examined by a novel setup with a water-pressure control system. A total of 10 group model tests were carried out. The results indicated that the relationship between water level height and foundation bearing capacity is negatively correlated. In addition, the numerical analysis was carried out to investigate the effect of water-level change on the bearing capacity of the foundation. The bearing capacity of the foundation decreases as the water-level cycles increase. The increase in the fluctuation range of the water level will decrease the bearing capacity of the foundation. The outcome of this study would be helpful to predict the bearing capacity of shallow foundations due to the change of the water level.

The bearing capacity of foundations is one of the interesting subjects in geotechnical engineering. In many cases, constructing foundations on natural or artificial soil slopes to develop the infrastructures is controversial. The construction of foundations on slopes can significantly affect the bearing capacity and slope stability. Soil stabilisation by polymer reinforcements is a modern method employed in various projects to prevent the failure of soil slopes and to improve the bearing capacity of foundations, subsequently. This paper aims to evaluate the bearing capacity of shallow strip foundations constructed on geosynthetic reinforced sand slope using a finite difference programme, FLAC. The effects of geometrical and resistivity parameters of reinforcements layers was investigated for determining the optimal values to achieve maximum bearing capacity. Furthermore, the effects of strength properties of sand embankment, foundation position and slope angle on the behaviour of strip foundation rested on reinforced soil slope were investigated. The results indicated that the bearing capacity of shallow foundations remarkably increased using geosynthetic reinforcement layers.

In the context of engineering practice, the problem of the seismic bearing capacity of shallow foundations has been solved indirectly, either due an increase of the static allowable soil pressures related to the probability of occurrence of the design earthquake or by adopting an equivalent pseudo-static approach. However, during last decades, a series of analytical methods that directly address the problem from the seismic point of view has been developed. This paper presents a parametric comparative analysis of different methods for estimating seismic bearing capacity of shallow strip foundations. Analytical methods, developed in the framework of both limit equilibrium and limit analysis theories, and also simplified design procedures typically used in practice were considered. The results obtained show an important decrease of the bearing foundation capacity with increasing of the maximum earthquake acceleration, which highlights the need to obtain a measure of the reliability associated with both calculation methods and safety factors commonly used for seismic design.

New design equations for estimation of ultimate bearing capacity of shallow foundations resting on rock masses

Determining ultimate bearing capacity of shallow foundations by using multi expression programming (MEP)

Seismic bearing capacity of shallow foundations resting on rock masses subjected to seismic loads

Several experimental and theoretical investigations have been carried out to predict the bearing capacity of shallow foundations on reinforced cohesionless soils. It has been demonstrated that placing layers of reinforcement within the foundation soil increases the bearing capacity of shallow foundations remarkably. A limited number of relations has been suggested for predicting the bearing capacity of shallow foundations on reinforced cohesionless soils. In this paper two common types of artificial neural network (ANN), feedforward backpropagation (BP) and radial basis function (RBF), are used to predict the bearing capacity of shallow foundations on reinforced cohesionless soils based on laboratory and field measurements. The results are then compared with the previous traditional methods, showing a much greater degree of accuracy.

A major concern in design of structures is to provide precise estimations of ultimate bearing capacity of soil beneath their foundations. Direct determination of the bearing capacity of foundations requires performing expensive and time consuming laboratory tests. To cope with this issue, several numerical models have been presented by researchers. This paper presents the development of a new design equation for the prediction of the ultimate bearing capacity of shallow foundations on granular soils using linear genetic programming (LGP) methodology. The ultimate bearing capacity is formulated in terms of width of footing, footing geometry, depth of footing, unit weight of sand, and angle of shearing resistance. The LGP-based design equation is established using the results of several load tests on real sized foundations presented in the literature. Validity of the model is verified using a part of laboratory data that are not involved in the calibration process. The statistical measures of coefficient of determination, root mean squared error and mean absolute error are used to evaluate the performance of the model. Sensitivity and parametric analyses are conducted and discussed. The proposed model accurately characterizes the ultimate bearing capacity resulting in a very good prediction performance. The LGP model reaches a better prediction performance than the well-known prediction equations for the bearing capacity of shallow foundations.

Bearing capacity of shallow foundation under cyclic load on cohesive soil

There are a number of factors affecting the bearing capacity of shallow foundations. Among them, the scale effect can be mentioned as one of the most important factors. Unlike the theoretical equations, experiments show that the bearing capacity of foundations does not increase without limit when the foundation size increases. The effect of stress level on soil shear strength parameters has been known as the main reason for this observation. The method of the zero extension lines (ZEL) for the solution of plasticity problems in soil mechanics has been utilized to take this effect into account by incorporating the stress level − dependent soil friction angle. The bearing capacity of shallow foundations is then computed with the aid of this method, showing a decreasing tendency in the third factor, N γ, which is the main contributor in shallow foundations. Comparisons have been made with experimental data, showing good consistency between experiments and theoretical predictions with the ZEL method.

Discrete element method for bearing capacity analysis

The purpose of this paper is to propose the geosynthetic soil reinforcing technique as a simple and cost-effective alternative of improving the bearing capacity of shallow foundations. It is also expected that the geosynthetic soil reinforcing technique presented herein can help prevent those buildings built on shallow foundations from excessive settlements. Over the last few decades, many pilot and full-scale tests have been conducted, and it has been confirmed that the geosynthetic soil reinforcing technique can improve the bearing capacity of shallow foundations. This paper summarizes the processes of the theoretical development, experimental work, and numerical simulation on the bearing capacities of shallow foundations built on soils reinforced utilizing the geosynthetic reinforcing technique.

The standard penetration tests are considered one of the most important tests to determine the bearing capacity of the soil. In this research, a set of objective maps and equations were produced to estimate the bearing capacity of shallow foundations based on the results of SPTs conducted in Baghdad Governorate. The work includes drilling 213 wells 10 meters deep below the surface of the earth, and conducting three standard penetration tests (SPT) on each well at depths of 1.5, 6 and 9 m. The bearing capacity of the shallow foundations in the city of Baghdad was determined using the MATLAB program after the SPT values were corrected. Then several polynomial equations with multi-order interpolation are proposed to estimate the bearing capacity of the foundation, but the first-order polynomial equation is considered simpler and straightforward. Moreover, the root mean square error (RMSE) of all the proposed polynomial equations is almost the same. Thematic maps show the difference in the carrying capacity of shallow foundations across the total ground of Baghdad Governorate with respect to different depths. Then, a comparison was made between the calculated and estimated values of the foundation’s bearing capacity, where the results showed a discrepancy of 30% at a confidence level of 95%.