Stochastic Simulation of Settlement Prediction of Shallow Foundations Based on a Deterministic Artificial Neural Network Model

The problem of estimating the settlement of shallow foundations on granular soils is complex and not yet entirely understood. In the past, many empirical and theoretical methods have been developed for predicting the settlement of shallow foundations on granular soils; however, these methods are far from accurate and consistent. In recent times, artificial neural networks (ANNs) have been used for settlement prediction of shallow foundations on granular soils and have shown to outperform the most commonly used traditional methods. However, despite the relative advantage of the ANN based approach, it is like most traditional methods in the sense that it is based on a deterministic approach that does not take into account the considerable level of uncertainty that may affect the magnitude of the predicted settlement. Thus, it provides single values of settlement with no indication of the level of risk associated with these values. In this paper, an alternative stochastic approach that considers the uncertainty associated with the predicted settlement from a deterministic ANN model is provided. The proposed stochastic approach is based on combining Monte Carlo simulation with the deterministic ANN model from which a set of stochastic design charts for settlement prediction of shallow foundations on granular soils is developed. The charts will enable the designer to make informed decisions regarding the level of risk associated with predicted settlements and consequently provide a more realistic indication of what the actual settlement might be.