Over the last few years or so, the use of artificial neural networks ( ANNs) has increased in many areas of engineering. In particular, ANNs have been applied to many geotechnical engineering problems and have demonstrated some degree of success. A review of the literature reveals that ANNs have been used successfully in pile capacity prediction, modelling soil behaviour, site characterisation, earth retaining structures, settlement of structures, slope stability, design of tunnels and underground openings, liquefaction, soil permeability and hydraulic conductivity, soil compaction, soil swelling and classification of soils. The objective of this paper is to provide a general view of some ANN applications for solving some types of geotechnical engineering problems. It is not intended to describe the ANNs modelling issues in geotechnical engineering. The paper also does not intend to cover every single application or scientific paper that found in the literature. For brevity, some works are selected to be described in some detail, while others are acknowledged for reference purposes. The paper then discusses the strengths and limitations of ANNs compared with the other modelling approaches. The engineering properties of soil and rock exhibit varied and uncertain behaviour due to the complex and imprecise physical processes associated with the formation of these materials (Jaksa 1995). This is in contrast to most other civil engineering materials, such as steel, concrete and timber, which exhibit far greater homogeneity and isotropy. In order to cope with the complexity of geotechnical behaviour, and the spatial variability of these materials, traditional forms of engineering design models are justifiably simplified. An alternative approach, which has been shown to have some degree of success, is based on the data alone to determine the structure and parameters of the model . The technique is known as artificial neural networks ( ANNs) and is well suited to model complex problems where the relationship between the model variables is unknown (Hubick 1992). This paper is intended to be for readers in the field of geotechnical engineering who are not familiar with artificial neural networks. The paper aims to detail some features associated with ANNs through a review for some of their applications to-date in geotechnical engineering. It is hoped that this review may attract more geotechnical engineers to pay better attention to this promising tool. The paper starts with a brief overview of the structure and operation of the ANNs and gives a general overview of most ANN applications that have appeared in the geotechncial engineering literature. Finally, the paper discusses the relative success of ANNs in predicting various geotechnical engineering properties and behaviour.

Artificial neural network applications in geotechnical engineering

Over the last few years, artificial neural networks (ANNs) have been used successfully for modeling almost all aspects of geotechnical engineering problems. Whilst ANNs provide a great deal of promise, they suffer from a number of shortcomings such as knowledge extraction, extrapolation and uncertainty. This paper presents a state-of-the-art examination of ANNs in geotechnical engineering and provides insights into the modeling issues of ANNs. The paper also discusses current research directions of ANNs that need further attention in the future.

State of the Art of Artificial Neural Networks in Geotechnical Engineering

Probabilistic Reliability: an Engineering Approach

A study has been carried out of the bearing capacity of sand layers overlying clay soils for the case where the thickness of the sand layer is comparable to the width of a rigid foundation placed on the soil surface. A review of previous work is given and a discussion is presented of the dimensionless groups that govern the behaviour of this type of foundation. A parametric study is carried out using both finite element and finite difference methods. This study is based on the use of soil parameters obtained from an assessment of the range of values that might be expected to be appropriate for full-scale structures. The results of the parametric study are used to illustrate the mechanics of the system and also to develop charts that may be used directly in design. In particular, the results illustrate that the shear strength of the clay has an important influence on the mechanisms of load spread within the fill. Resume : Lion a rOalisO une Otude sur la capacitO portante de couches de sable reposant sur des sols argileux pour le cas oo liOpaisseur de la couche de sable est comparable ‡ la largeur diune fondation rigide posOe sur la surface du sol. Lion prOsente une revue des travaux antOrieurs et une discussion sur les groupes sans dimension qui rOgissent le comportement de ce type de fondation. Une Otude paramOtrique est rOalisOe en utilisant les mOthodes tant des OlOments finis que des diffOrences finies. Cette Otude est fondOe sur liutilisation des paramtres de sols obtenus en partant de liOvaluation de la plage des valeurs que lion pourrait siattendre ‡ trouver dans des structures ‡ pleine Ochelle. Les rOsultats de liOtude paramOtrique ont OtO utilisOs pour illustrer le mOcanisme du systme et aussi pour dOvelopper des abaques qui peuvent Œtre utilisOes directement dans la conception. En particulier, les rOsultats illustrent que la rOsistance au cisaillement de liargile a une influence importante sur les mOcanismes de rOpartition de la charge ‡ liintOrieur du remblai. Mots clOs : capacitO portante, sols multicouches, analyse numOrique. (Traduit par la rOdaction)

Bearing capacity of plane-strain footings on layered soils

Comparison of different probabilistic methods for predicting stability of a slope in spatially variable c–φ soil

Probabilistic analysis of retaining walls

A 17 m-high steel strip reinforced soil retaining wall was instrumented to compare field measurements with predictions given by the design guidelines of the American Association of State Highway and Transportation Officials (AASHTO) 1996 Standard Specifications and the AASHTO 1999 Interim Revisions. The AASHTO models were conservative with respect to external lateral earth pressures and lateral earth pressures on the facing panels. On average, the AASHTO 1996 and 1999 models overestimated lateral pressure at the facing by 94 and 142%, respectively. Measured values of foundation bearing stress were generally in good agreement with values calculated using soil unit weight and depth, except that the average force from the facing panels on the leveling pad was twice that of the weight of the panels themselves. This discrepancy is attributed to shear stress on the back of the facing panels and vertical loads transferred to the panels through the strip connection clips. The location of the zone of maximum strip...

Field Performance of a 17 m-High Reinforced Soil Retaining Wall

Assessing the Liquefaction Susceptibility at a Site Based on Information from Penetration Testing

Modeling of membrane action in a two-layer reinforced soil system

Many types of geosynthetic-reinforced soil are subjected to repeated or cyclic loading. In transportation infrastructures such loading conditions occur in pavements supported by reinforced soil ret...

Philippe L. Bourdeau

Papers

Probabilistic analysis of retaining walls

Field Performance of a 17 m-High Reinforced Soil Retaining Wall

Assessing the Liquefaction Susceptibility at a Site Based on Information from Penetration Testing

Modeling of membrane action in a two-layer reinforced soil system

Geosynthetic-Reinforced Soils Under Repeated Loading: A Review and Comparative Design Study