See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/355172197

Pile Design in Seismic Areas: Small or Large Diameter?

Article in Geotechnical Engineering · June 2020

CITATIONS READS

0 184

1 author:

Raffaele Di Laora
Università degli Studi della Campania "Luigi Vanvitelli"
60 PUBLICATIONS 802 CITATIONS

SEE PROFILE

All content following this page was uploaded by Raffaele Di Laora on 11 October 2021.

The user has requested enhancement of the downloaded file.

 Geotechnical Engineering Journal of the SEAGS & AGSSEA Vol. 51 No. 2 June 2020 ISSN 0046-5828

Pile Design in Seismic Areas: Small or Large Diameter?

R. Di Laora1
1
Department of Engineering, University of Campania “Luigi Vanvitelli”, Aversa (CE), Italy
E-mail: [email protected]

ABSTRACT: This work investigates the role of pile diameter in resisting seismic actions, with reference to two example subsoils, namely a
dry sand and a fully saturated NC clay. After a ground response analysis in free-field conditions for different values of peak rock acceleration,
mobilized soil stiffness and surface acceleration are used as ingredients for assessing the kinematic and inertial moment in a concrete pile. An
optimum pile diameter is identified as the one that, while guaranteeing safety, corresponds to the minimum cost. It is also proven that, with a
constant value of reinforcement area and length, increasing pile diameter (i.e. increasing safety factor and cost) leads rapidly to failure.
Likewise, if pile reinforcement is designed only for inertial action, increasing pile diameter is severely detrimental.

KEYWORDS: Pile design, Seismic action, Kinematic interaction, Earthquake-induced bending, Pile diameter

1. INTRODUCTION and total cross-sectional area, moment capacity is roughly

proportional to d3.
The topic of the seismic performance of piles started receiving
attention when post-earthquake investigations revealed the
development of large bending moments: (a) at the head of piles
restrained against rotation by rigid caps and (b) close to interfaces
separating soil layers of sharply differing stiffness, even in absence
of large soil movements such as those induced by lateral spreading
following liquefaction (Kavvadas and Gazetas 1993, Gazetas and
Mylonakis, 1998, Brandenberg et al., 2005, Varun et al., 2013 among
others). Nevertheless, interpretation of the available evidence is not
straightforward. The main reasons are related to the difficulty in
simulating real-life conditions in theoretical models or lab
experiments. Furthermore, the superposition of simultaneous
kinematic and inertial interaction phenomena, whose effects are
difficult to separate, represents an additional difficulty in the
interpretation of data. It is noted that the former type of interaction is
associated to the deformation of the soil surrounding the piles due to
seismic shaking and thereby leads to development of bending over
the whole pile length, whereas the latter is related to the oscillations
of the superstructure and thus generates moments that are maximum
at the pile top and become insignificant below a certain depth
(Figure 1).
A simple method for assessing the kinematic component of pile
bending was first proposed by Margason (1975) and Margason and
Holloway (1977). These articles can be credited pioneering
investigations on the role of pile diameter (to be denoted in the
ensuing by d) and recommend, with some justification, the use of
small diameters to "conform to soil movements”. While several
subsequent studies investigated the problem of kinematic bending
(e.g., Dobry and O’Rourke, 1983, Mineiro, 1990, Kavvadas and
Gazetas, 1993, Mylonakis, 2001, Nikolaou et al., 2001, Maiorano et
al, 2009, de Sanctis et al., 2010, Dezi et al., 2010, Di Laora et al.,
2012, Di Laora et al. 2013, Di Laora and Rovithis, 2015, Martinelli
et al., 2016, Mucciacciaro and Sica, 2018), only a handful of
investigations focused on the effect of pile diameter, mostly for
bending in the proximity of interfaces separating soil layers of sharply
differing stiffness (Mylonakis, 2001, Saitoh, 2005).
Di Laora et al. (2017) explored the role of pile diameter in
resisting seismic actions at the pile top in presence of a cap restraining Figure 1 Kinematic and inertial loading of pile foundations
head rotation, with reference to steel and concrete piles in subsoils
with constant stiffness and stiffness proportional to depth. With This observation revealed a previously unsuspected scale effect
reference to constant stiffness and concrete pile, the work highlighted that causes total moment demand, taken as the mere sum of kinematic
that: and inertial contributions, to increase faster than moment capacity,
a) kinematic bending moment is proportional to d4; thus making yielding at the pile head unavoidable beyond a certain
b) under the assumption of pile in clay with constant undrained "critical" diameter. More specifically, the sole inertial moment
shear strength, if only shaft resistance is considered, for constant provides a minimum diameter below which a pile cannot resist the
values of pile length L and global safety factor against axial demand, while the sole kinematic contribution provides a maximum
bearing capacity SF inertial moment is proportional to d3; diameter. The combination of the two moments reduces the range of
c) under the assumption of constant ratio between reinforcement
172
 Geotechnical Engineering Journal of the SEAGS & AGSSEA Vol. 51 No. 2 June 2020 ISSN 0046-5828

admissible diameters. As far as a soil with stiffness proportional to Proceeding along these lines, this work expands the investigation
depth is concerned, kinematic interaction moment increases at a of the role of pile diameter in resisting seismic forces under different
smaller rate with pile diameter; this results in a larger minimum assumptions and with reference to more realistic subsoil profiles, as
diameter which, however, is mainly due the large kinematic bending reported in the ensuing.
which develops in such soft soils (Figure 2).

Figure 2 Pile size limitation in idealized subsoil (after Di Laora et al., 2017)

2. SUBSOILS AND FREE-FIELD SEISMIC RESPONSE with z the depth from ground surface expressed in meters. A unit
weight γ = 16 kN/m3 and a Poisson’s ratio ν = 0.3 have been
Since the problem of size limitation is relevant only for soils having considered. The resulting equivalent shear wave velocity is Vs,30 = 200
low-to-moderate stiffness at shallow depths, two cases are considered m/s, thereby corresponding to a class C soil according to Eurocodes.
in this study, namely a dry sand with medium relative density and a With reference to the NC clay, initial stiffness has been
normally-consolidated (NC), fully saturated clay. Following Hardin considered variable proportionally to depth according to the relation:
and Drnevich (1992) the sand (profile A in Figure 3) is assumed to
possess a low-strain stiffness proportional to the square root of depth,
according to the equation: G0  MPa  = 3  z (2)

G0  MPa  = 20  z 0.5 (1) with z in meters. A saturated unit weight γsat = 18 kN/m3 and a
Poisson’s ratio ν = 0.5 have been chosen, while the resulting
equivalent shear wave velocity Vs,30 is 116 m/s (class D).

Figure 3 Results of ground response analysis

173
 Geotechnical Engineering Journal of the SEAGS & AGSSEA Vol. 51 No. 2 June 2020 ISSN 0046-5828

To take into account some overconsolidation at very shallow 3.2 Inertial moment
depth, the stiffness in the first meter has been kept constant and equal
to the one deriving from the above equations for z = 1 m; this location By interpreting results of some numerical analyses, it was found that
also corresponds the position of pile head. the moment at the top of a fixed-head pile is roughly equal to 1/4 the
Ground response under seismic action has been evaluated through horizontal force Fh times the active length La. Considering that Fh is
the freely available program Strata (Kottke and Rathje, 2008). Seven proportional to the axial load Pp carried by the pile, the inertial
accelerograms have been selected to match class A design spectrum moment may be therefore expressed by:
with reference to 3 different levels of peak rock acceleration ar,
namely 0.15, 0.20 and 0.25 times the gravity acceleration g. An elastic 1 as S a
M in = Pp La (6)
bedrock has been considered at the depth of 30 m, having shear wave 4 g q
velocity Vs = 800 m/s. Earthquake signals have been applied as
outcrop motions. where as·Sa is the elastic spectral acceleration, to be divided by the
To take into account soil non-linear behaviour, an Equivalent- behavior factor q, which accounts for structure ductility, to obtain the
Linear approach has been adopted, taking advantage of literature structure design acceleration.
curves of Modulus Reduction and Damping vs. shear strain. For the
dry sand, the mean curves proposed by Seed and Idriss (1970) have 3.3 Section capacity
been used, while the curves by Vucetic and Dobry (1991), with
reference to a Plasticity Index equal to 30%, have been employed. With reference to a concrete pile, the section moment capacity (the
Results of the ground response analyses are plotted in Figure 3. yield moment is assumed in this work for simplicity) may be
The following aspects are noteworthy: (a) maximum acceleration is estimated through the simplified formula (Di Laora et al., 2019):
always at surface; (b) the clay exhibits a higher amplification, mainly
due to a more pronounced increase in acceleration when approaching 3
2 d  2d 
surface; (c) for the sand the earthquake-induced decrease in shear M y = M y , c + M y , s =   sin 3  f ck +  − c  As sin  f yk (7)
modulus is more pronounced; (d) the mobilized shear modulus profile 3 2  2 
is consistent with the suggestion of Eurocode of simply scaling down
the initial stiffness profile of a factor which is function of the surface where My,c and My,s denote, respectively, the relative contributions of
acceleration. concrete and steel, f ’ck = 0.9 fck, the latter being the characteristic
The mobilized stiffness of the soil computed from the above compressive strength of concrete, f ’yk = 0.95 fyk , with fyk the yield
analyses is employed for the assessment of kinematic and inertial strength of steel reinforcement, c is the thickness of the concrete
bending, as detailed in the following. cover, θ is a characteristic angle expressed by:

3. PILE BENDING DEMAND AND CAPACITY

  4   + 
2
 32
 =  1 + 2 −   −1 + 1 +  (8)
3.1 Kinematic moment 4     (1 + 2 − 4 /  ) 
2
 
Di Laora and Rovithis (2015) investigated the kinematic moment at
pile head in a subsoil whose stiffness varies in a continuous manner where ω = As f ’yk / (Ac f ’ck) is the mechanical percentage of
according to the generalized power law: reinforcement and ν = Pp / (Ac f ’ck) is the dimensionless axial force
parameter, As and Ac are reinforcement and total section areas,
 z
n respectively.
G ( z ) = Gsd  a + (1 − a )  (3)
 d
4. EFFECT OF PILE DIAMETER
where G(z) is the soil shear modulus at depth z, Gsd the corresponding The role of pile diameter in the development of kinematic bending is
value at the depth of one pile diameter, a and n are coefficients quite straightforward inspecting Equations 4 and 5. It is easy to verify
regulating soil stiffness at surface and rate of stiffness increase with that for the case of constant stiffness (n = 0) kinematic moment is
depth, respectively. It is straightforward to notice that n = 1 proportional to d4, while for depth-proportional stiffness (a = 0, n =
corresponds to a linear variation of stiffness with depth, while for a = 1) kinematic moment is proportional to d3.2; all cases with a and n
0 soil stiffness vanishes at soil surface. With reference to a fixed-head, between 0 and 1 are bounded by these two extreme scenarios.
'long' (i.e. L > 8-10d) pile, the authors showed that pile bending Identifying the effect of pile diameter on the inertial moment
moment at the head is accurately evaluated through the relation: requires further considerations. In the work by Di Laora et al. (2017)
pile length L and safety factor SF were considered constant, resulting
as  s in a working load Pp proportional to diameter in the hypothesis of
M kin = E p I p (4) constant undrained shear strength and negligible tip resistance; this
G ( La / 2 ) leads to an inertial moment proportional to d2. Consideration of tip
resistance and/or increasing strength with depth results in the
where Ep is the Young’s modulus of pile material, Ip is the cross- exponent of diameter to raise to values up to 3. An alternative line of
sectional moment of inertia, as soil surface acceleration, ρs soil reasoning, and perhaps more oriented towards a design problem, is to
density and G(La/2) is soil stiffness evaluated by means of Equation keep constant the working load Pp that the pile should safely carry. A
3 at the depth of La/2, with La the pile active length expressed by (Di variation in pile diameter therefore corresponds, for a given safety
Laora and Rovithis, 2015, Karatzia and Mylonakis, 2016): factor SF, to a different length. This way, it is easy to derive from
Equations 5 and 6 that inertial moment is proportional to d for
constant stiffness (n = 0) and to d0.8 for the other extreme case of soil
 4

1   n +4 4 5 ( n + 4 )(1 − a )     E p  
1 4 n+4
 stiffness being proportional to depth (a = 0, n = 1). Note that
14
La
= 
 a +      − a (5) consideration of tip resistance does not alter the exponent, as it merely
d 1 − a  16    sd  
2 E 
 reduces length for a given SF.
  As far as section capacity is concerned, Equations 7 and 8 show
that, if the reinforcement area is increased proportionally to the
with Esd soil Young’s modulus at the depth of one pile diameter. section area, My is roughly proportional to d3.

174
 Geotechnical Engineering Journal of the SEAGS & AGSSEA Vol. 51 No. 2 June 2020 ISSN 0046-5828

Figure 4 depicts a typical trend of demand and capacity with pile With reference to the cases analyzed in this work, Figure 5 depicts
diameter. It is easy to recognize that capacity increases with pile inertial and kinematic moment demand for the two subsoils as
diameter more than inertial moment and less than the kinematic function of pile diameter, for different values of surface acceleration
demand. resulting from the ground response analysis described in Section 2.
Results are obtained with reference to a linear stiffness profile
interpolating mobilized stiffness below pile head.
The two selected values of pile load (i.e., Pp = 3000 for the sand
and 1000 for the NC clay) correspond to a safety factor SF of about
2.5 for a pile of 1 m in diameter and 25 m in length, for common
choices in the bearing capacity calculation.
It is worth noticing that:
a) both kinematic and inertial moment increase with acceleration;
b) kinematic bending is much more severe for the pile in clay; this
is due to both the higher accelerations occurring at surface and
the lower mobilized stiffness (see Equation 4 and Figure 3);
c) kinematic moments tend to dominate over the inertial
counterpart for large diameters for the pile embedded in clay;
d) despite the large difference in the working load, inertial bending
in clay is just slightly lower than in sand; this is due to a partial
compensation owing to higher acceleration and lower mobilized
stiffness.
Figure 4 Schematic trend of kinematic, inertial, combined demand
and section capacity as function of pile diameter

Figure 5 Kinematic and inertial demand as function of pile diameter for different acceleration levels and for the two subsoil under
investigation. All curves are for Sa/g = 0.833, Ep = 30 GPa

Towards a deeper understanding of size limitations, Figure 6 5. OPTIMAL PILE DIAMETER

illustrates the above bending contributions, both as individual
contributions and combined, normalized by section capacity for a From the discussion above it is apparent that any pile diameter larger
constant value (= 1%) of the ratio of reinforcement area over total than the minimum value is acceptable to sustain earthquake loading
cross-sectional area. It is noted that for the pile in clay kinematic (for simplicity yielding moments have been used instead of code-
action does not result in a maximum diameter, but tremendously based section capacity for given partial factor, but concept remains
affects minimum diameter, and the design of a pile with such the same).
reinforcement ratio is only possible for the lowest value of rock The selection of an optimal pile diameter therefore relies on the
acceleration (= 0.15g). The pile in sand is not affected by relevant size least cost of installation. Following the above hypothesis of constant
limitations, yet the minimum diameter to withstand seismic action As/Ac, cost may be roughly expressed as function of pile diameter as
with reasonable reinforcement is rather high. follows:
Figure 7 shows the total seismic demand normalized by section
capacity for different values of As/Ac, up to the Codes limit of 4%. It C ( d ) = C f  L  d + Cc  L  d 2 + Cs  L  d 2 (9)
is evident that the pile in clay needs a very strong reinforcement and
quite large diameter to resist seismic forces.

175
 Geotechnical Engineering Journal of the SEAGS & AGSSEA Vol. 51 No. 2 June 2020 ISSN 0046-5828

Figure 6 Kinematic, inertial and combined demand over capacity as function of pile diameter for different acceleration levels and for the two
subsoil under investigation. All curves are for Sa/g = 0.833, Ep = 30 GPa, As/Ac = 1%, fck = 25 MPa, fyk = 450 MPa

Figure 7 Combined demand over capacity as function of pile diameter for different values of As/Ac and for the two subsoil under
investigation. All curves are for ar = 0.20g, Sa/g = 0.833, Ep = 30 GPa, fck = 25 MPa, fyk = 450 MPa

where Cf, Cc and Cs are unit costs related to perforation, concrete and in length for a given design axial load does not balance the increase
steel, clearly variable worldwide. in area. Under such an assumption, an increase in pile diameter cannot
For the subsoils considered in the study, given the absence of a be detrimental for pile safety, regardless of cost considerations.
very stiff bearing layer, it is reasonable to think that doubling the Let us now think of designing a pile for seismic actions, so that
length does not lead to a 4 times lower diameter when designing for we select diameter, length and reinforcement area which provide
axial bearing capacity under constant SF. This means that the lower adequate capacity to resist seismic demand. Suppose to increase pile
the diameter, the lower the cost. It is therefore inferred that the diameter, with length and reinforcement area set at a constant value.
optimum pile diameter is the minimum diameter shown in the graphs This way, safety against a bearing capacity failure increases as well
above. Kinematic interaction can remarkably increase this minimum as cost.
value. Figure 8 illustrates this scenario. The selected values of
reinforcement areas correspond to 1%, 2% and 4% of a pile of 0.8 m
6. CAN LARGE DIAMATERS BE DETRIMENTAL? in diameter. It is clear from the graph that increasing pile diameter,
despite the increase in cost and safety against bearing capacity failure,
All the graphs above have been conceived assuming a constant value may lead to a tremendous decrease in safety against pile structural
of As/Ac, so that an increase in pile diameter corresponds to an collapse. This is particularly true for the pile in clay, where despite
increase in the amount of reinforcement, as the consequent decrease the large amount of reinforcement for a pile of 0.8 m in diameter,
176
 Geotechnical Engineering Journal of the SEAGS & AGSSEA Vol. 51 No. 2 June 2020 ISSN 0046-5828

increasing size above this value leads rapidly to failure. This occurs To shed further light on the last concept, one may think of a pile
because the remarkable increase of kinematic moment (proportional designed only for inertial action, for example ensuring that the inertial
to diameter raised to an exponent between 3.2 and 4) is not balanced moment is half of the section capacity. Figure 9 shows seismic
by the increase in section capacity (roughly proportional to d1.4). As demand normalized by capacity for different acceleration levels as
a side comment, note that the increase in diameter under constant load function of pile diameter. It is evident that increasing pile diameter
also results in a decrease of normal stress in the section, and this is leads quickly to failure.
detrimental for the flexural capacity. The above arguments provide a possible justification of the severe
The above leads to the important conclusion that large diameters damage observed in large diameter piles (designed only to withstand
are safer (as shown in Figures 6 and 7) only if accompanied by a inertial loading) during post-earthquake investigations in the past.
substantial increase in reinforcement to withstand the increasing
kinematic demand.

Figure 8 Kinematic, inertial and combined demand over capacity as function of pile diameter for different values of As and for the two
subsoil under investigation. All curves are for ar = 0.20g, Sa/g = 0.833, Ep = 30 GPa, fck = 25 MPa, fyk = 450 MPa

Figure 9 Demand over capacity for a pile designed only for inertial action, as function of pile diameter for different acceleration levels and
for the two subsoil under investigation. All curves are for Sa/g = 0.833, Ep = 30 GPa, fck = 25 MPa, fyk = 450 MPa

177
 Geotechnical Engineering Journal of the SEAGS & AGSSEA Vol. 51 No. 2 June 2020 ISSN 0046-5828

7. CONCLUSIONS Di Laora, R., Mylonakis, G., & Mandolini, A. (2017). Size limitations
for piles in seismic regions. Earthquake Spectra, 33(2), 729-
The work herein presented tries to shed light on the long-standing 756.
issue of the role of pile diameter in resisting seismic actions. In some Di Laora, R. & Rovithis, E. (2015). Kinematic Bending of Fixed-
past work, it was recognized that kinematic demand increases with Head Piles in Nonhomogeneous Soil. J. Geotech. Geoenviron.
pile diameter more than section capacity does, and therefore there Eng.141, No. 4, 04014126.
must exist an upper bound for pile diameter beyond which the pile Hardin, B. O., & Drnevich, V. P. (1972). Shear modulus and damping
cannot resist the demand. On the other hand, inertial action increases in soils: design equations and curves. Journal of Soil
at a smaller rate compared to the capacity and thus provides a Mechanics & Foundations Div, 98(sm7).
minimum diameter. This paper investigates seismic demand in Iovino, M., Di Laora, R., Rovithis, E., & de Sanctis, L. (2019). The
concrete piles for realistic subsoils subjected to different acceleration beneficial role of piles on the seismic loading of structures.
levels. Reference is made to a dry sand and a fully saturated NC clay; Earthquake Spectra, Volume 35(3), 1-24.
after a ground response analysis in free-field conditions, surface Kavvadas, M. & Gazetas, G. (1993). Kinematic seismic response and
acceleration and mobilized soil stiffness are used as ingredients for bending of free-head piles in layered soil. Géotechnique 43,
assessing the kinematic and inertial moment. No. 2, 207-222.
The main conclusions of the study are the following: Karatzia, X., & Mylonakis, G. (2016). Discussion of “Kinematic
a) For a constant axial load and safety factor against a bearing Bending of Fixed-Head Piles in Nonhomogeneous Soil” by
capacity failure, if reinforcement area is taken as proportional to Raffaele Di Laora and Emmanouil Rovithis. Journal of
the whole cross-sectional area, there is always a minimum Geotechnical and Geoenvironmental Engineering 142, No. 2,
diameter to resist seismic demand, while the maximum diameter 07015042.
is too large to represent a real limitation; Kottke, A. R., & Rathje, E. M. (2008). Technical manual for Strata.
b) The minimum diameter is remarkably affected by the kinematic University of California, Berkeley.
moment, which is much larger for the pile embedded in the clay; Maiorano, R. M. S., de Sanctis, L., Aversa, S. & Mandolini, A. (2009).
c) From a design perspective, the above minimum diameter also Kinematic response analysis of piled foundations under
represents the optimal pile diameter as it corresponds to the seismic excitations. Can. Geotech. J. 46, No. 5, 571-584.
minimum cost; Margason, E. (1975). Pile bending during earthquakes. Lecture, 6
d) If the total area of reinforcement is set as constant (i.e. does not March 1975, ASCE UC/Berkeley Seminar on Design
increase with increasing cross section), kinematic moment Construction and Performance of Deep Foundations.
provides a severe limitation for the upper bound of diameter; Margason, E. & Holloway, D. M. (1977). Pile Bending during
increasing diameter at constant length and reinforcement area Earthquakes, Proc. 6th World Conference on Earthquake
(and therefore increasing both safety factor and cost) Engng, Sarita Prakashan, Meerut, India, 1977, 1690-1696, Vol.
surprisingly - at first sight - leads rapidly to failure; II.
e) If reinforcement is designed to resist only inertial action, Martinelli, M., Burghignoli, A. & Callisto, L. (2016). Dynamic
increasing pile diameter is dramatically detrimental for pile response of a pile embedded into a layered soil. Soil Dynamics
structural safety. This may explain the severe damage observed and Earthquake Engineering 87, 16-28.
in large diameter piles (designed only for inertial loading) during Mineiro, A. J. C. (1990). Simplified Procedure for Evaluating
post-earthquake investigations in the past. Earthquake Loading on Piles, De Mello Volume, Lisbon.
It is worth mentioning that this study focuses only on the role of Mylonakis, G. (2001). Simplified model for seismic pile bending at
pile diameter to resist seismic action. Nevertheless, piles may also soil layer interfaces. Soils & Foundations 41, No. 3, 47-58.
provide a filtering action on the seismic demand transmitted to the Mucciacciaro, M., & Sica, S. (2018). Nonlinear soil and pile
superstructure and therefore the selection of a large diameter pile, behaviour on kinematic bending response of flexible piles. Soil
despite detrimental for the pile itself, may help the structure to resist Dynamics and Earthquake Engineering, 107, 195-213.
seismic loading. However, this topic is discussed elsewhere (e.g. Di Nikolaou, A., Mylonakis, G., Gazetas, G. & Tazoh, T. (2001).
Laora and de Sanctis, 2013; Iovino et al., 2019) and lies beyond the Kinematic pile bending during earthquakes analysis and field
scope of this work. measurements. Géotechnique 51, No. 5, 425-440.
Saitoh, M. (2005). Fixed-head pile bending by kinematic interaction
8. REFERENCES and criteria for its minimization at optimal pile radius. Journal
Brandenberg, S. J., Boulanger, R. W., Kutter, B. L., & Chang, D. of Geotechnical and Geoenvironmental Engineering 131, No.
(2005). Behavior of pile foundations in laterally spreading 10, 1243-1251.
ground during centrifuge tests. Journal of Geotechnical and Seed, H. B., & Idriss, I. M. (1970). Soil moduli and damping factors
Geoenvironmental Engineering 131, No. 11, 1378-1391. for dynamic analysis. Report No. EERC 70-10. University of
de Sanctis, L., Maiorano, R.M.S. & Aversa, S. (2010). A method for California, Berkeley.
assessing bending moments at the pile head. Earthquake Engng Varun, Assimaki, D. & Shafieezadeh, A. (2013). Soil–pile–structure
Struct. Dyn 39, 375-397. interaction simulations in liquefiable soils via dynamic
Dezi, F., Carbonari, S. & Leoni, G. (2010). Kinematic bending macroelements: Formulation and validation.
moments in pile foundations. Soil Dyn. Earthquake Engng 30, Vucetic, M., & Dobry, R. (1991). Effect of soil plasticity on cyclic
No. 3, 119-132. response. Journal of geotechnical engineering, 117, No. 1, 89-
Di Laora, R. & de Sanctis, L. (2013). Piles-induced filtering effect on 107.
the Foundation Input Motion. Soil Dynamics and Earthquake
Engineering 46, 52-63.
Di Laora, R., Mandolini, A. & Mylonakis, G. (2012). Insight on
kinematic bending of flexible piles in layered soil. Soil
Dynamics and Earthquake Engineering 43, 309-322.
Di Laora, R., Mylonakis, G. & Mandolini, A. (2013). Pile-head
kinematic bending in layered soil. Earthquake Engineering and
Structural Dynamics 42, 319-337.

178

View publication stats