651
Technical Committee 103 /
Comité technique 103
problems are now becoming more and more common and
important in geotechnical engineering practice as interaction
between the old existing structures and the newly installed geo-
structures will lead to problems that did not exist in the past. It
is a result of rapid urbanization. For example, Rezaei et al.
(2013) reported that urban development and increasingly
growth of population have been accompanied by a considerable
growth in mechanized Shield tunnelling. For example, Mayoral
et al. (2013) considered numerical analysis of the static behavior
of the intersection of two major metro lines located in a soft
lacustrine clay deposit overlaid by a very dense clayed sand
deposit, in Mexico City. The intersection consists of a new
tunnel excavated under an existing metro station-tunnel system,
using the earth pressure balance, EPB, construction technique.
This required the construction of a support structure for the
station foundation. Actually, these kinds of problems exist in
many developing cities in the world. For example, the MTR
project of Shatin to Central link in Hong Kong also faced
similar problems. Building new railway in developed and
densely populated urban areas is a very challenging task.
Everaars and Peters (2013) considered the case studies of a
large infrastructural railway project through the historical city
centre of Delft, The Netherlands and of an underground
expansion project of the Drents Museum in Assen, The
Netherlands.
2.6.3. Retaining walls
There are 4 papers considering the application of numerical
methods to retaining walls. These are Sadrekarimi and
Monfared (2013), Smith et al. (2013), Mirmoradi and Ehrlich
(2013), and Bui et al. (2013). Retaining wall design is probably
one of the oldest geotechnical problems that was dealt with
rigor and mathematics. However, today it remains one of the
most difficult problems in geomechanics.
2.6.4 Slopes
There are 5 papers investigating the deformation of slopes (both
natural and cut slopes), including Yerro et al. (2013), Hoshina
and Isobe (2013), Kamalzare et al. (2013), Bryson and El
Naggar (2013), and Lu et al. (2013). Progressive failure of
slopes remains an elusive problem as it is a highly nonlinear
problem. The propagation of shear crack in slopes may play a
crucial role in such process (Palmer and Rice, 1973).
Unfortunately none of the submitted papers summarized in this
report addresses such problem. Numerical simulation of such
progressive failure will be highly sensitive to the constitutive
models. If bifurcation type of instability of slope sliding occurs
(Chau 1995, 1999), numerical modeling will be very difficult.
2.6.5 Levees
Three studies addressed the erosion and instability of levees.
They are Fujisawa and Murakami (2013), Kamalzare et al.
(2013), and Smith et al. (2013). This topic becomes extremely
since the levee failure occurred at New Orleans during the
attack of the South Asian Tsunami in 2004 and the Hurricane
Katrina in 2005. Recently, tsunami disasters in Japan after 2011
Tohoku earthquake further reinforce the importance of this topic.
2.6.6 Breakwaters
Stickle et al. (2013) considered wave-induced nonlinear
dynamic soil response in vertical breakwaters foundation using
Biot’s (1941) theory extended to include dynamic terms. Note
that Biot’s (1941) theory allows for the coupling between soil
deformations and pore water pressure fluctuations.
2.7
Experimental validation
2.7.1
Centrifuge tests
Kamalzare et al. (2013) used the experimental results of a 150
g-ton geotechnical centrifuge to calibrate their models for
modeling levee’s erosion. Kwon et al. (2013) compared their
results of dynamic analysis based on the finite difference
method under seismic loading to the observations in dynamic
centrifuge tests. Although centrifuge test can provide the
required high stress level typically observed in the field, the size
of particles in soil samples may also present an unwanted scale
effect in soil deformation mechanism.
2.7.2
1-g experiments
As expected, 1-g experiments in laboratory have been the most
commonly used tool in calibrating and validating numerical
models. Various experiments (including direct shear box and
triaxial test) have been reported in the following papers:
Wardani et al. (2013), Stirling and Davie (2013), Ramon and
Alonso (2013), Tanchaisawat et al. (2013), Kamalzare et al.
(2013), Hashash et al. (2013), Sokolić and Szavits-Nossan
(2013), Pinkert and Klar (2013), and James et al. (2013).
2.8
Constitutive modeling
The constitutive modeling of responses of soils and rocks were
considered by Biru and Benz (2013), Biru et al. (2013), Dong
and Anagnostou (2013), Ebrahimian and Noorzad (2013),
Pereira et al. (2013), Siddiquee and Islam (2013), and Yao et al.
(2013). To avoid singularity at the yield vertex of the Mohr-
Coulomb failure surface, Dong and Anagnostou (2013) replaced
the original Mohr-Coulomb yield surface by the Matsuoka-
Nakai criterion. However, this leads to the issue of whether
yield surface vertex is real. Rudnicki and Rice (1975) and
Rudnicki (1984) argued based on the mechanics of sliding
microcrack that yield vertex should exist in rocks. However, as
summarized in Chau (2013) the existence of such yield surface
vertex is inconclusive.
3 VALIDATION OF FEM MODELS
As remarked by Brinkgreve and Engin (2013), the use of the
Finite Element Method for geotechnical analysis and design has
become quite popular. It is often the younger generation of
engineers who operate easy-to-use finite element programs and
produce colourful results, whilst the responsible senior
engineers find it difficult to validate the outcome. Brinkgreve
and Engin (2013) presented the recent finding by the NAFEMS
Geotechnical Committee of the Netherlands on the validation of
geotechnical finite element analysis. This is an extremely
important topic that has largely been ignored in the past. The
NAFEMS Geotechnical Committee has concluded that there is
a need for guidelines on validation of geotechnical finite
element calculations. This section summarizes the essential
findings reported in Brinkgreve and Engin (2013).
3.1
Sources of discrepancies
Brinkgreve and Engin (2013) classified the sources of
discrepancies into 6 categories: Simplifications, Modelling
errors, Constitutive modeling, Uncertainties, Software and
hardware issues, Misinterpretation of results. Examples of these
are:
3.1.1
Examples of simplifications
Simplifications are normally made on the geometries of the
problem, on the selection of model boundaries, on material
behaviour, and on the presumed construction process.
3.1.2
Examples of modelling errors
Modelling errors can include input errors, discretisation errors
(meshing), boundary conditions, time integration, tolerances
(tolerated numerical errors), and limitations in theories and
methods (e.g. the use of small deformation theory for problems
with large deflections).
