Machine Learning-Based Prediction of the Seismic Bearing Capacity of a Shallow Strip Footing over a Void in Heterogeneous Soils

Abstract

The seismic bearing capacity of a shallow strip footing above a void displays a complex dependence on several characteristics, linked to geometric problems and to the soil properties. Hence, setting analytical models to estimate such bearing capacity is extremely challenging. In this work, machine learning (ML) techniques have been employed to predict the seismic bearing capacity of a shallow strip footing located over a single unsupported rectangular void in heterogeneous soil. A dataset consisting of 38,000 finite element limit analysis simulations has been created, and the mean value between the upper and lower bounds of the bearing capacity has been computed at the varying undrained shear strength and internal friction angle of the soil, horizontal earthquake accelerations, and position, shape, and size of the void. Three machine learning techniques have been adopted to learn the link between the aforementioned parameters and the bearing capacity: multilayer perceptron neural networks; a group method of data handling; and a combined adaptive-network-based fuzzy inference system and particle swarm optimization. The performances of these ML techniques have been compared with each other, in terms of the following statistical performance indices: coefficient of determination (); root mean square error (); mean absolute percentage error; scatter index; and standard bias. Results have shown that all the ML techniques perform well, though the multilayer perceptron has a slightly superior accuracy featuring noteworthy results (0.9955 and 0.0158).

Special Issue “2021 Selected Papers from Algorithms’ Editorial Board Members”

“…The paper [10] deals with a prediction of the seismic bearing capacity of a shallow strip footing above a void based on machine learning. Three machine learning techniques have been adopted to learn the connection between the parameters under consideration and the bearing capacity, which have been compared with each other.…”

Satin bowerbird optimizer-neural network for approximating the capacity of CFST columns under compression

“…Due to the highly non-linear relationship between the mechanical parameters of construction material and influential characteristics, recent scientific efforts advice employing machine learning like artificial neural network (ANN) 32 , gradient tree boosting algorithm 33 , support vector regression (SVR) 34 , and adaptive neuro-fuzzy inference system (ANFIS) 35 models for such purposes. These models are able to map and reproduce the intrinsic dependency of any output parameter on its corresponding inputs 36 – 38 . For example, Ghasemi and Naser 39 could successfully use two explainable artificial intelligence techniques called XGBoost and random forest to predict the compressive strength of 3D concrete mixtures.…”

Machine learning approaches to estimation of the compressibility of soft soils

“…Over the past decades, the applicability of machine learning (ML) approaches, such as artificial neural networks (ANNs), random forest (RF) methods, and support vector machines (SVMs), among others, has been well-proven in terms of their ability to efficiently and accurately map highly non-linear problems in a wide variety of areas of engineering (Arditi and Pulket, 2010;Chen et al, 2021), including geotechnical engineering. Successful examples of applications include analyses of slope stability (Kardani et al, 2021;Meng et al, 2021) and deformation (Zhang et al, 2019;Zhang et al, 2020a;; pile designs (Makasis et al, 2018;Zhang et al, 2020e); prediction of the bearing capacity of strip footings (Acharyya, 2019;Sadegh et al, 2021); lateral wall deformation and basal heave stability for braced excavations (Goh et al, 1995;Zhang et al, 2020); soil constitutive relations (Najjar and Huang, 2007); liquefaction resistance of sands (Kim and Kim, 2006); lining response for tunnels (Zhang et al, 2020g); calibration of resistance factors for reliability-based load and resistance factor design (Hu and Lin, 2019); prediction of soil transparency (Wang et al, 2021); analysis of ground settlement induced by shield tunneling (Zhang et al, 2020c); reliability analysis by SVM (Pan and Dias, 2017); and mapping of groundwater potential using SVM, RF, and GA models (Naghibi et al, 2017), among others. In addition to solving geotechnical analysis problems, these ML approaches have also achieved success in mapping from the physical parameters of soil to the mechanical parameters.…”