Predicting ultimate bearing capacity of shallow foundations on reinforced cohesionless soils using artificial neural networks A. Soleimanbeigi 1 and N. Hataf 2 1 Graduate Student, Civil Engineering Department, Shiraz University, School of Engineering, Shiraz, Iran, Telephone: +98 711 6286531 5, Telefax: +98 711 6287294, E-mail: a_soleimanbeigi@yahoo.com 2 Associate Professor, Civil Engineering Department, Shiraz University, School of Engineering, Shiraz, Iran, Telephone: +98 711 6286531 5, Telefax: +98 711 6287294, E-mail: nhataf@shirazu.ac.ir Received 29 December 2004, revised 9 July 2005, accepted 20 August 2005 ABSTRACT: 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. KEYWORDS: Geosynthetics, Shallow foundations, Bearing capacity, Reinforcement, Neural network REFERENCE: Soleimanbeigi, A. & and Hataf, N. (2005). Predicting ultimate bearing capacity of shallow foundations on reinforced cohesionless soils using artificial neural networks. Geosynthetics International, 12, No. 6, 321–332 1. INTRODUCTION Determination of the bearing capacity of foundations is one of the most important tasks in geotechnical engineer- ing. In certain conditions, when the foundation does not have sufficient bearing capacity to resist the applied loading, some modifications can increase soil bearing capacity. Most granular soils have large compressive and shearing strength, with little or no tensile strength. The performance of these soils can be improved remarkably by placing reinforcing elements to supplement their poor tensile strength (Zhao et al. 1998; Hataf and Baziar 2000). Several experimental and theoretical studies have been carried out to investigate the bearing capacity of shallow foundations on reinforced cohesionless soils. Binquet and Lee (1975a, 1975b) proposed two failure criteria for reinforced foundations on sand as the result of laboratory tests with aluminium reinforcing strips. Dixit and Mandal (1993) used the variational method to estimate the bearing capacity of reinforced foundations. This included simulta- neously solving several complex differential equations. Boushehrian and Hataf (2003) investigated the bearing capacity of circular and ring footings on reinforced sand. Huang and Tatsuoka (1990) carried out several labora- tory tests on the bearing capacity of reinforced founda- tions using various reinforcing materials. From the test results, they classified two failure modes in reinforced sand, as follows. • Failure mode 1. In this mode compressive failure occurs in the unreinforced zone beneath the reinforced zone, as a rigid deep footing (Figure 1a). The bearing capacity increase in this mode of failure is obtained as follows: ˜q C ¼ ˜q B þ ˜S (1) ˜q B ¼ K P ó cd  K P ó cs ¼ K 2 P ª d 2 D f þ h f þ 2s 2 ð Þ 2  K 2 P ª d c þ s 1 ð Þ 2 (2) ˜S ¼ 2 X n i¼1 T e, i tan ö N i B (3) where ˜q B is the increase in the bearing capacity by deepening a rigid footing in unreinforced sand (depicted in Figure 2); ˜S is the increase in upward Geosynthetics International, 2005, 12, No. 6 321 1072-6349 # 2005 Thomas Telford Ltd