Prediction of Safe Bearing Capacity of Noncohesive Soil in Arid Zone Using Artificial Neural Networks

Abstract

Estimation of safe bearing capacity (SBC) of noncohesive soil based on Indian Standard Code requires a lot of field work, viz, conducting direct shear tests to determine cohesion and angle of internal friction, performing the standard penetration test to determine the N-value of soil, and finding the relative density and dry density of soil. The present study does away with these soil parameters except for the design value of density and uses the results of sieve analysis to determine the SBC of soil. This research proposes the use of artificial neural network (ANN) to predict the SBC of noncohesive soil as a function of coefficient of curvature, coefficient of uniformity, and design value of soil density along with footing dimensions such as depth, width and diameter (in case of circular footing), and the desired settlement of the footing. The results show that ANN is a useful technique in estimating SBC of noncohesive soil using parameters derived from sieve analysis results and match closely from the results derived from the traditional methods based on Terzaghi’s theories.