sustainability

Article
Assessment of the Bearing Capacity of Foundations on
Rock Masses Subjected to Seismic and Seepage Loads
Rubén Galindo 1, * , Ana Alencar 1 , Nihat Sinan Isik 2 and Claudio Olalla Marañón 1
1 Departamento de Ingeniería y Morfología del Terreno, Universidad Politécnica de Madrid,
28040 Madrid, Spain; at.santos@alumnos.upm.es (A.A.); claudio.olalla@upm.es (C.O.M.)
2 Department of Civil Engineering, Faculty of Technology, Gazi University, 06560 Ankara, Turkey;
nihatsinan@gazi.edu.tr
* Correspondence: rubenangel.galindo@upm.es




Received: 20 October 2020; Accepted: 30 November 2020; Published: 2 December 2020


Abstract: It is usual to adopt the seismic force acting as an additional body force, employing the
pseudo-static hypothesis, when considering earthquakes in the estimation of the bearing capacity
of foundations. A similar approach in seepage studies can be applied for the pore pressure’s
consideration as an external force. In the present study, the bearing capacity of shallow foundations
on rock masses considering the presence of the pseudo-static load was developed by applying an
analytical solution for the Modified Hoek and Brown failure criterion. Calculations were performed
adopting various inclinations of the load and the slope on the edge of the foundation, as well as
different values of the vertical and horizontal components of the pseudo-static load. The results are
presented in the form of charts to allow an affordable and immediate practical application for footing
problems in the event of seismic loads or seepages. Finally, and to validate the analytical solution
presented, a numerical study was developed applying the finite difference method to estimate the
bearing capacity of a shallow foundation on a rock mass considering the presence of an additional
horizontal force that could be caused by an earthquake or a seepage.

Keywords: Hoek–Brown criterion; seismic load; seepage; analytical method; pseudo-static load;
finite difference method

1. Introduction
Natural events such as earthquakes and floods generate impulsive loads and seepages in the
ground, endangering civil constructions. Infrastructures must be built to resist natural risks in adequate
safety conditions and be economically viable. Therefore, an optimized design that reliably considers
the negative effects that these natural phenomena introduce to the foundations is highly advisable to
guarantee the sustainability of the infrastructures.
Today, research in the area of seismic bearing capacity is very much in demand because of
the devastating effects of foundations under earthquake conditions. A high number of failures
have occurred where field conditions have indicated that the bearing capacity was reduced during
seismic events.
In the estimation of the bearing capacity, when the effect of the earthquake is considered, it is
usual to adopt the pseudo-static hypothesis where the seismic force acts as an additional body force
within the soil mass. The vertical and horizontal acceleration are applied both on the ground and in the
structure. Thus, the limit conditions can be evaluated by introducing pseudo-static equivalent forces,
corresponding with the inertial forces in the soil during the seismic excitation. Such an approach is
based on the hypothesis of a synchronous motion for the soil underneath the footing, a hypothesis that
is acceptable only in the case of small footing widths and large values for the soil stiffness. According to

Sustainability 2020, 12, 10063; doi:10.3390/su122310063 www.mdpi.com/journal/sustainability

Sustainability 2020, 12, 10063 2 of 21

numerous studies carried out about this issue in the field of geotechnical engineering, it is known that
the bearing capacity is considerably reduced mostly due to the horizontal component.
These studies are primarily based on (a) the limit equilibrium method [1–4], (b) the method of the
characteristics [5], and (c) the limit analysis method using numerical analysis [6–8].
Limit equilibrium methods have evolved and incorporated different factors, generalizing the
application hypotheses to solve the most general possible configurations. Thus, Sarma and Iossifelis [1]
determined seismic bearing capacity factors for soils with seismic acceleration acting on the structure
only and using the limit equilibrium technique of slope stability analysis with inclined slices.
Later, Richarts et al. [2] considered the inertial mass of the ground using a simplification of the
Prandtl failure surface to eliminate the fan-shaped transition zone and thereby average its effect
concentration. This effect can be considered as a retaining wall with the active lateral thrust from
the region below the foundation pushing against the passive resistance from the external region.
Then, Choudhury and Rao [4], using the limit equilibrium method, studied the seismic forces for
shallow strip footings embedded in sloping ground with a linear failure criterion. They considered
pseudo-static forces acting both on the footing and on the soil below the footing and considered in the
analysis a composite failure surface involving a planar and logspiral.
For its part, Kumar and Rao [5] applied the method of the characteristics using pseudo-static forces
for a Mohr–Coulomb criterion. The analytical methods were limited to simplified configurations where
the inertia of the soil mass was not included and had to be completed with empirical or numerical terms.
Adopting the conventional pseudo-static approach, Yang [7] used a non-linear failure criterion
to evaluate the seismic bearing capacity of strip foundations on rock slopes by means of the
generalized tangential technique, in which the non-linear strength is replaced by an “optimal”
tangential Mohr–Coulomb domain in a limit analysis framework where the upper-bound solutions are
obtained by optimization.
A numerical study was performed by Raj et al. [8] using finite-element limit analysis for shallow
strip foundations embedded in homogeneous soil slopes, applying a pseudo-static approach to consider
the seismic action on both the soil and foundation.
Nova and Montrasio [9] introduced a satisfactory description of the dynamic soil–structure
interaction by means of the macro-element concept (the system as a unique, non-linear macro-element
with a limited number of degrees of freedom), which has been applied by many researchers to different
problems of shallow foundations [10,11].
Applying a similar approach, in recent years, studies have been published where the pore
pressure has been included as an external force, analyzing the slope stability in soil [12] and rock [13],
and affecting the bearing capacity of the shallow foundation on soil [14–16] and in jointed rock mass [17].
In foundations with large widths such as dams, the seepage tends to present an almost horizontal flow
in the influence area of the foundation, as has been possible to verify in real cases [18].
In the case of seismic load, the horizontal acceleration is the same in the whole model, while in
the case of seepage, the horizontal acceleration can vary once there is a loss of force along the water
path. Depending on the direction of the groundwater flow, it can act as a passive (resistant) or active
force [16], once it is in the same direction to the stress path or not. Thus, the wedge formed below
the footing is asymmetric. Additionally, Veiskarami and Kumar [14] show that with an increase
in the hydraulic gradient, the nature of the failure patterns becomes more non-symmetrical; for a
non-symmetrical failure pattern, the footing would have a greater chance of failing in overturning
rather than under simple vertical compression.
It must be pointed out that the traditional formulations of the bearing capacity are based on the
Mohr–Coulomb parameters (the cohesion and friction angle) that are very efficient in the field of soil
mechanics, with a linear behavior. In rock mechanics, the current methods for estimating the bearing
capacity adopt the Hoek and Brown failure criterion [19] (valid only for rock masses not excessively
fractured) and Modified Hoek and Brown failure criterion [20,21] (valid under general conditions of
degree of fracturing of the rock mass). Both failure criteria are non-linear and applicable to the rock
Sustainability 2020, 12, 10063 3 of 21

mass with a homogeneous and isotropic behavior. The linearization of the failure criterion implies
incorporating approximate methods and requires iterative procedures to ensure an optimized upper
or lower bound for the solution; therefore, it is desirable to be able to address the problem using the
non-linear criterion directly.
In addition, most of the formulations are limited to flat ground, the need to analyze the bearing
capacity of shallow foundations on sloping ground of moderate slopes being very common in dam
and bridge foundations. Finally, although the numerical solutions allow solving complex problems
with singular considerations when a seismic load or filtration acts, in the face of non-linear criteria,
a complete analysis of numerical convergence is necessary, which complicates the practical applicability
and the rapid design of foundations in rock masses.
An analytical method for the calculation of shallow foundations that solves the internal equilibrium
equations and boundary conditions combined with the failure criterion was proposed by Serrano and
Olalla [22] and Serrano et al. [23], applying the Hoek and Brown [19] and the Modified Hoek and
Brown failure criterion [20], respectively. It is based on the characteristic lines method [24], with the
hypothesis of the weightless rock, strip foundation and associative flow law. The formulation of the
bearing capacity proposed by Serrano et al. [23] introduces a bearing capacity factor, which makes the
failure pressure proportional to the uniaxial compressive strength of the rock (UCS).
In the present study, the analytical formulation of Serrano et al. [23] and design charts were
developed to study the bearing capacity when there is an increase in load induced by forces of seismic
origin or filtration in rock masses, where it is necessary to use a non-linear failure criterion and the
own weight of the ground is generally negligible compared to the resistant components, considering,
as usual, the possibility of the inclination of the ground to the edge of the foundation. Besides,
a numerical model was created through a finite difference method, assuming a similar hypothesis
for the analytical solution, and it was observed that the results obtained by both methods were
quite similar.

2. Problem Statement

2.1. Mathematical Model

As is generally known, in rock mechanics, the non-linear Modified Hoek and Brown failure
criterion is the most used, and it is applicable for a rock mass with a homogeneous and isotropic
behavior, meaning that by the inexistence or by the abundance of discontinuities, it has the same
physical properties in all directions.
In this research, the Modified Hoek and Brown failure criterion [21,25] was used, and it was
formulated as a function of the major principal stress (σ3 ) and minor principal stress (σ1 ) according to
the following equation:
σ1 − σ3 σ
 a
= m· 3 + s (1)
σc σc
The uniaxial compressive strength (UCS) is σc , while the parameters m, s and a can be evaluated
with (2)–(4) and depend on the intact rock parameter (mo ), quality index of the rock mass (geological
strength index (GSI)) and damage in the rock mass due to human actions (D), which in shallow
foundations, is usually equal to zero.
GSI−100
m = mo ·e 28−14·D (2)
GSI−100
s=e 9−3·D (3)
1 1 −GSI
 
−20
a = + e 15 − e 3 (4)
2 6
Serrano et al. [23] proposed an analytical formulation for estimating the ultimate bearing capacity
of the strip footing for a weightless rock mass, based on the characteristic method, which allows solving
the internal equilibrium equations in a continuous medium together with the boundary equations and
 Sustainability 2020, 11, x FOR PEER REVIEW 4 of 22

Serrano et al. [23] proposed an analytical formulation for estimating the ultimate bearing
capacity of the strip footing for a weightless rock mass, based on the characteristic method, which
Sustainability
allows solving2020, 12,the
10063internal equilibrium equations in a continuous medium together with4 of the
21
boundary equations and those that define the failure criterion. This solution is based on the
Modified Hoek and Brown failure criterion [20], taking into account the associated plastic flow rule.
those that define the failure criterion. This solution is based on the Modified Hoek and Brown failure
According to this analytical formulation, the ground surface that supports the foundation is
criterion [20], taking into account the associated plastic flow rule.
composed of two sectors (Figure 1): Boundary 1 (free), with the inclination i1, where the load acting
According to this analytical formulation, the ground surface that supports the foundation is
on a surface is known (for example, the self-weight load on the foundation level or the load from
composed of two sectors (Figure 1): Boundary 1 (free), with the inclination i1 , where the load acting
installed anchors), and Boundary 2 (foundation), where the bearing capacity of the foundation
on a surface is known (for example, the self-weight load on the foundation level or the load from
should be determined (acting with the inclination of i2).
installed anchors), and Boundary 2 (foundation), where the bearing capacity of the foundation should
be determined (acting with the inclination of i2 ).

Boundary 2: Foundation

Ph i2 i1 Boundary 1: Free surface

𝜋 𝜌2 1
( + )
4 2
Rankine-2 region 𝜋 𝜌1
( − )
4 2


Prandtl region Rankine-1 region

Figure1.1.Mathematical
Figure Mathematicalmodel
modelofofthe
thebearing
bearingcapacity
capacityofofthe
thestrip
stripfooting.
footing.

The
Thesolution
solutionbased basedon onthe
thecharacteristic
characteristiclines
linesmethod
methodrequires
requiresthe
theequation
equationofofthe
theRiemann
Riemann
invariants
invariants(I(Ia )a)[26]
[26]fulfilled
fulfilledalong
alongthe
thecharacteristic
characteristicline:
line:
𝐼I𝑎a ((𝜌
ρ11))++ψ𝜓1 1==Ia𝐼(𝑎ρ(𝜌 )+𝜓
2 )2+ ψ2 2
(5)
(5)
1   ρ  𝜌
𝐼𝑎 (𝜌) 1 ∙ [𝑐𝑜𝑡𝑔(𝜌) + 𝑙𝑛 (𝑐𝑜𝑡𝑔 ( ))]
Ia (ρ=

)= · cotg ( ρ ) + ln cotg (6)
(6)
2 ∙2·k
𝑘 2 2
InInthis
thisequation,
equation,the
theinstantaneous
instantaneousfriction
frictionangle
angleatatthe
theboundary
boundary22(ρ(ρ2 )2)isisthe
theonly
onlyunknown,
unknown,
because
becausethe other
the variables
other can be
variables candefined at Boundary
be defined 1: the instantaneous
at Boundary friction angle
1: the instantaneous at Boundary
friction angle at
1 (ρ1 ) and the angle (Ψ1 ) between the major principal stress and the vertical axis in this sector (Figure 1).
Sustainability 2020, 12, 10063 5 of 21

Thus, expressing Ψ2 (the angle between the major principal stress and the vertical axis in Boundary 2,
as indicated in Figure 1) as a function of ρ2 , it is possible to estimate the ultimate bearing capacity.
Through the analytical method [23], the bearing capacity was obtained by (7).
 
Ph = βa · Nβ − ζa (7)

The resistant parameters βa and ζa were applied to make dimensionless the calculation of the
Modified Hoek and Brown failure criterion. βa represents the characteristic strength, which has the
same units as the UCS and was used to make the pressures dimensionless, while ζa (the “tenacity
coefficient”) is a dimensionless coefficient that, multiplied by βa , corresponds to the tensile strength.

!1
s m·(1 − a) k (1 − a)
βa = Aa ·UCS; ζa = ; Aa = 1
; k= (8)
(m·Aa ) 2 a a

Aa , k and the exponent a are constants for the rock mass and depend on the rock type (m),
UCS and GSI.
Nβ is the bearing capacity factor, and it can be calculated, according to the problem statement,
as follows.
The angle of internal friction (ρ1 ) can be obtained by iteration from the load at Boundary 1.
From the value of ρ1 and by the iteration of (5), the value of the internal friction angle at Boundary 2
(ρ2 ) can be calculated.
Finally, knowing ρ2 , the bearing capacity factor (Nβ ) can be calculated, and using, again, parameters
βa and ζa , the ultimate bearing capacity (Ph ) was estimated as an expression that depended on the
instantaneous friction angle at Boundary 2 (ρ2 ), the inclination of the load on the foundation (i2 ) and
the exponent of the Modified Hoek and Brown criterion (a; k = (1 − a)/a):
1 
1−sin(ρ2 ) a·(1+sin(ρ2 ))

k
Nβ = cos(i2 ) k·sin(ρ2 ) sin(ρ2 )
cos(i2 )
r 2  (9)
a·(1+k·sin(ρ2 ))

+ 1− sin(i2 ) 

sin(ρ2 )

2.2. Consideration of Pseudo-Static Load: Mathematical Transformation

In the pseudo-static approach, static horizontal and vertical inertial forces, which are intended
to represent the destabilizing effects of the earthquake or seepage, are calculated as the product of
the seismic/seepage coefficients and the distributed load applied to the boundaries. In the case of the
rock mass, the weight collaboration is usually negligible compared to the resistance of the ground,
and therefore, the inertial forces are applied both to the foundation and to the free boundary.
The vertical seismic/seepage coefficient kv is supposed to be a fraction of one horizontal kh , and in
particular, the vertical acceleration is thus assumed to be in phase with the horizontal acceleration.
The present study is divided into three parts: (a) The first one considered the horizontal (kh ) and
vertical (kv ) components of the pseudo-static load on both boundaries, with a free boundary inclined
by α at the edge of the foundation, which resembles the hypothesis of a seism. (b) In the second part,
only the horizontal component (kh ) on the foundation boundary was adopted (it being possible to
consider both the horizontal and vertical components on the free boundary depending on the direction
of the seepage), with the free boundary inclined by α at the edge of the foundation. This hypothesis is
more similar to the presence of a seepage (both hypotheses are represented schematically in Figure 2,
and they are solved and shown in new charts including additional horizontal and vertical loads).
(c) The final section is the comparison of the analytical result with that obtained numerically through
the finite difference method.
Sustainability 2020, 12, 10063 6 of 21
Sustainability 2020, 11, x FOR PEER REVIEW 6 of 22

Figure2.2.Scheme
Figure Schemeof
ofthe
thepseudo-static
pseudo-staticload
loadacting:
acting:(a)
(a)Seismic
Seismicload
load(F
(Fss);); (b)
(b) Seepage
Seepage (F
(Fww).

The
The application
application of of the
the analytical
analytical method
method [23] [23] with
with anan increase
increase in in the
the horizontal
horizontal and/or
and/or vertical
vertical
loads
loads produced
produced by by inertial
inertial forces
forces cancan be
be carried
carried out out by
by means
means of of aaparametric
parametric transformation
transformation from from
the
the incidence angles of the loads in the static configuration to the final configuration including the
incidence angles of the loads in the static configuration to the final configuration including the
seismic
seismic or seepage loads. loads. Thus,
Thus,for fora ageneral
generalcasecase ofof a pseudo-static
a pseudo-static force
force on onthethe
twotwo boundaries,
boundaries, the
the starting
starting point
point is theis the
load load acting
acting in the
in the static
static hypothesis
hypothesis (subscript
(subscript 0), 0),
as isasrepresented
is represented in Figure
in Figure 3a.3a.
In
In this
this case,
case, thethe inclinations
inclinations of of
thethe
loadload
ononthethe foundation
foundation (p)(p)
and and of the
of the loadloadon on
thethefreefree boundary
boundary (q)
(q) 𝑖02i02and
areare and𝑖01i01, ,respectively.
respectively. However,
However, considering
considering the the pseudo-static
pseudo-static load, the inclinations
inclinations of of the
the
loads
loads in
in both
both boundaries
boundaries are aredifferent
different(i(i22 for
for the
thefoundation
foundation boundary
boundary and andii11 for the free boundary).
boundary).
Figure
Figure 3b 3b allows
allows the the deduction
deduction of of the
the mathematical
mathematical transformations
transformations of of these
these angles
angles from
from thethe static
static
configuration
configurationtotothe final
the pseudo-static
final pseudo-static configuration
configuration as a function of the angle
as a function of the(α)angle
of the(α)freeofboundary
the free
and of the horizontal
boundary and of the or kh2 , depending
(kh1horizontal (kh1 or konh2, the boundary)
depending onand thevertical
boundary) (kv1 or kv2 ,vertical
and depending (kv1 on
or the
kv2,
boundary)
dependingcomponents
on the boundary) of the pseudo-static
componentsload. These
of the transformations
pseudo-static load.areThese
expressed in (10) and (11).
transformations are
expressed in (10) and (11).
tan (i02 ) + kh2
tan (i2 ) = 𝑡𝑎𝑛 (𝑖02 ) + 𝑘ℎ2 (10)
𝑡𝑎𝑛 (𝑖 ) = 1 − kv2 (10)
2
1 − 𝑘𝑣2
(1 − kv1 )tan(α + i01 )
tan (α + i1 ) = (1 − 𝑘𝑣1 )𝑡𝑎𝑛(𝛼 + 𝑖01 ) (11)
𝑡𝑎𝑛 ( + 𝑖1 = 1 − kh1 tan(α + i01 )
) (11)
1 − 𝑘ℎ1 𝑡𝑎𝑛(𝛼 + 𝑖01 )
Therefore, for the pseudo-static calculation, the analytical formulation of the characteristics
method expressed by (5) can be used using the load inclinations on each boundary obtained through
Boundary
the transformations indicated2in(Foundation)
Equations (10) and (11) Boundary 1 (Free)
as a function of the horizontal and vertical

a
components of the added inertial load and of the angle of inclination of Boundary 1.
p
seepage load, kv1 = 0; thus, for clearer
i02 that kh1 = kh2 and kv1 = kv2 , while in the case of the
In the case of the seismic load, it vis considered
i01
phnotation, it is denoted qv load, the horizontal
that for the seismic
and vertical components of the pseudo-static load are kh and kv , respectively, and for the seepage load,
the horizontal component of the pseudo-static load on the foundation boundary qh is called ia .


expressed in (10) and (11).
𝑡𝑎𝑛 (𝑖02 ) + 𝑘ℎ2
𝑡𝑎𝑛 (𝑖2 ) = (10)
1 − 𝑘𝑣2
(1 − 𝑘𝑣1 )𝑡𝑎𝑛(𝛼 + 𝑖01 )
Sustainability 2020, 12, 10063 𝑡𝑎𝑛 ( + 𝑖1 ) = (11)
7 of 21
1 − 𝑘ℎ1 𝑡𝑎𝑛(𝛼 + 𝑖01 )

Boundary 2 (Foundation) Boundary 1 (Free)

a pv
ph
i02
i01
qv
qh

Sustainability 2020, 11, x FOR PEER REVIEW 7 of 22

Boundary 2 (Foundation) Boundary 1 (Free)

b pv-kv2pv
ph+kh2pv qv-kv1qv
qh-kh1qv

𝑝 +𝑘 ℎ 2 ·𝑝 𝑣 𝜋 𝑞 −𝑘 ·𝑞
𝑡𝑎𝑛 (𝑖2 ) = (1ℎ 𝑡𝑎𝑛 ( −  − 𝑖1 ) = (1ℎ ℎ 1) 𝑣
− 𝑘 𝑣2 ) 𝑝 𝑣 2 −𝑘 𝑞 𝑣1 𝑣

Figure
Figure 3.3. Scheme
Schemeofofthethe seismic
seismic loadload estimation:
estimation: (a)configuration
(a) static static configuration and (b) pseudo-static
and (b) pseudo-static inclination
inclination
on Boundary 1 (external boundary of foundation). Note: In this figure, the subscriptsthe
on Boundary 1 (external boundary of foundation). Note: In this figure, subscripts
“v” “v”
and “h” refer
and “h” refer to the vertical and horizontal projections
to the vertical and horizontal projections of the load. of the load.

3. Design Charts
Therefore, forforthe
Estimation of Bearing
pseudo-static Capacity
calculation, the analytical formulation of the characteristics
method expressed by (5) can be used using the load inclinations on each boundary obtained through
3.1. Calculation Cases and Representation of Analytical Results
the transformations indicated in Equations (10) and (11) as a function of the horizontal and vertical
components
Once the of mathematical
the added inertial load and of of
transformation thethe
angleloadof angles
inclination
in theof boundaries,
Boundary 1. for the problem
In theincase
presented of the
Figure 3a,seismic
accordingload, it is and
to (10) considered thatseismic
(11) for the kh1 = kh2and
andseepage
kv1 = kv2load
, while
hasinbeen
the carried
case of out,
the
seepage load, formulation
the analytical kv1 = 0; thus, using
for clearer notation,
the method it ischaracteristic
of the denoted that linesfor the
canseismic load, The
be applied. the horizontal
results are
and vertical
presented as components
graphs, which ofallow
the pseudo-static
the estimationload of thearebearing
kh and capacity
kv, respectively, and the
considering for presence
the seepage of a
load, the horizontal
pseudo-static load. component of the pseudo-static load on the foundation boundary is called ia.
The charts are clustered according to the exponent “a” of the Modified Hoek and Brown criterion,
3.
theDesign Charts
inclination α offor Estimation
Boundary of kBearing
1 and Capacity
v of the foundation boundary, and they were developed based on
io2 and the horizontal component of the pseudo-static load of the foundation boundary (kh ). It is noted
3.1.
thatCalculation Cases andpressures
for high confining Representation of Analytical
in Boundary 1, it isResults
not always possible to obtain a bearing capacity
value; this limit is demarcated by the non-equilibrium
Once the mathematical transformation of the load line.
angles in the boundaries, for the problem
For the
presented in development of the new
Figure 3a, according charts,
to (10) andthree values
(11) for of the GSI
the seismic (geological
and strength
seepage load index)
has been were
carried
adopted (8, 20 and 100), which generated exponents “a” of the Modified
out, the analytical formulation using the method of the characteristic lines can be applied. TheHoek and Brown criterion
equal to
results 0.5,
are 0.55 andas0.6
presented from (4).
graphs, which Based
allowonthe
(7),estimation
the valuesofofthe thebearing
rock type (mo ) and
capacity the uniaxial
considering the
presence of a pseudo-static load.
The charts are clustered according to the exponent “a” of the Modified Hoek and Brown
criterion, the inclination α of Boundary 1 and kv of the foundation boundary, and they were
developed based on io2 and the horizontal component of the pseudo-static load of the foundation
boundary (kh). It is noted that for high confining pressures in Boundary 1, it is not always possible to
obtain a bearing capacity value; this limit is demarcated by the non-equilibrium line.
Sustainability 2020, 12, 10063 8 of 21

compressive strength of the rock (UCS) do not influence the normalized graphs, and to perform the
calculations, mo = 15 and UCS = 1 MPa were adopted.
In the graphs developed for kv > 0 on the foundation boundary, representative of the seismic load,
four kh values were adopted and correlated two values of kv , two slope angles for the free boundary
(α) and three initial inclination angles for the load on the foundation boundary (io2 ), representing the
inclination angle of the load without considering the pseudo-static load. In Table 1, the values used in
the analysis are indicated.

Table 1. Geometric parameters adopted in the model (kv > 0).

kh kv = kh , kh /2 α (◦ ) io2 (◦ ) a
0.1 0.1, 0.05
0.5,
0.2 0.2, 0.1
0, 20 0, 10, 20 0.55,
0.3 0.3, 0.15
0.6
0.4 0.4, 0.2

On the other hand, in the charts developed with kv = 0 on the foundation boundary, representative
of the seepage loads, three values of horizontal load were used, in those cases called ia (additional
inclination); three values of the slope (α) and another three values of io2 were also used, which are
shown in Table 2.

Table 2. Geometric parameters adopted in the model (kv = 0).

ia α (◦ ) io2 (◦ ) a
0.1, 0.2, 0.3 0, 5, 10 0, 10, 20 0.5, 0.55, 0.6

It is emphasized that kh (the horizontal component of the seismic load) and ia (the horizontal
component of the seepage load) were in proportion to the vertical load applied; the values expressed
in Tables 1 and 2 are the relation between the horizontal and the vertical load. The final inclination
of the load in the foundation boundary, in the cases of seepage loads, can be obtained directly as a
simplification of (10):
tan (i2 ) = tan(io2 ) + ia (12)

3.2. Charts for Estimation of Bearing Capacity Considering kh and kv > 0 (Seismic Case)
These design charts allow obtaining the bearing capacity factor Nβ (9), and they are presented
in Figures 4–6. Each graph represented corresponds to determined values of the exponent “a” of
the Modified Hoek and Brown criterion of the rock mass (a function of the GSI of the quality of
the rock mass), the inclination α of the free boundary and the ratio of the vertical and horizontal
components of the pseudo-static load (kv /kh = 1 or 0.5). In each graph, different curves corresponding
to the angles of the inclination of the static load on Boundary 2 (io2 ) and horizontal component of the
pseudo-static load (kh ) are presented, so that in the abscissa, the known value of the normalized main
major stress is presented, normalized on Boundary 1 (σ*01 ), estimated through (13). It is dimensionless
and corresponds to the load acting on Boundary 1, and its value depends on the inclination angle of
the load i1 obtained by (11) (Figure 1).
σ
σ∗01 = 1 + ζa (13)
βa
Among the graphs, it is easy to appreciate that as there is a greater slope for the free boundary (α),
a lower kv /kh ratio and a lower exponent “a” from the Modified Hoek and Brown failure criterion,
the value of Nβ follows a declining trend.
Sustainability 2020, 12, 10063 9 of 21
Sustainability 2020, 11, x FOR PEER REVIEW 10 of 22

a = 0.5 /  = 0  / kv = kh a = 0.5 /  = 20  / kv = kh
15
1 i2=0; kh=0.1 15
Static Case 1 i2=0; kh=0.1
2 i2=0; kh=0.2
1 2 i2=0; kh=0.2
3 i2=10; kh=0.1
4 i2=0; kh=0.3 3 i2=10; kh=0.1
10 2 Static Case
5 i2=10; kh=0.2 10 4 i2=0; kh=0.3
1
N

3 6 i2=20; kh=0.1
5 i2=10; kh=0.2

N
i2=0; kh=0.4 2
7 6 i2=20; kh=0.1
5 4 i2=10; kh=0.3 3
5
i2=20; kh=0.2 5 4 i2=0; kh=0.4
5 7 i2=10; kh=0.3
6 8 i2=10; kh=0.4
Non-equilibrium 6 i2=20; kh=0.2
7 i2=20; kh=0.3
8 boundary 7
9 9 i2=20; kh=0.4 Non-equilibrium
0 9
8
boundary i2=10; kh=0.4
0 8
0,01 0,10 1,00 i2=20; kh=0.3
Normalized main major stress, *01 0,01 0,10 1,00
9 i2=20; kh=0.4
Normalized main major stress, *01

a = 0.5 /  = 0  / kv = kh /2 a = 0.5 /  =20  / kv = kh /2

15 15
1 i2=0; kh=0.2 1 i2=0; kh=0.2

2 i2=0; kh=0.3 2 i2=0; kh=0.3

Static Case
3 i2=10; kh=0.2 3 i2=10; kh=0.2
10 1 10 Static Case 4 i2=0; kh=0.4
4 i2=0; kh=0.4

N
5 i2=10; kh=0.3 5 i2=10; kh=0.3
2 1
N

6 i2=20; kh=0.2 6 i2=20; kh=0.2

2
5 3 7 i2=10; kh=0.4 5 3
7 i2=10; kh=0.4

4 4 i2=20; kh=0.3
i2=20; kh=0.3 8
5 8 5 i2=20; kh=0.4
i2=20; kh=0.4 6
6 Non-equilibrium
7
8 boundary 7 Non-equilibrium boundary
0 0 8

0,01 0.10 1.00 0.01 0.10 1.00

Normalized main major stress, *01
Normalized main major stress, *01

Figure 4. Seismic charts for estimation of the bearing capacity factor Nβ in the case of an exponent from Modified Hoek and Brown failure criterion: a = 0.5.
Sustainability 2020, 11, x FOR PEER REVIEW 11 of 22
Sustainability 2020, 12, 10063 10 of 21
Figure 4. Seismic charts for estimation of the bearing capacity factor Nß in the case of an exponent from Modified Hoek and Brown failure criterion: a = 0.5.

Figure 5.
Figure 5. Seismic
Seismic charts
charts for
for estimation
estimation of
of the
the bearing
bearing capacity
capacity factor
factor N
Nβßin
in the
the case
case of
of an
an exponent
exponent from
from Modified
Modified Hoek
Hoek and
and Brown
Brown failure
failurecriterion:
criterion: aa =
= 0.55.
0.55.
Sustainability 2020, 12, 10063 11 of 21
Sustainability 2020, 11, x FOR PEER REVIEW 12 of 22

Figure
Figure 6. Seismic
6. Seismic charts
charts for for estimation
estimation of of
thethe bearing
bearing capacityfactor
capacity factorNNβßin
in the
the case
case of
of an
an exponent
exponentfrom
fromModified
ModifiedHoek
Hoekand
andBrown failure
Brown criterion:
failure = 0.6.
a = a0.6.
criterion:
Sustainability 2020, 12, 10063 12 of 21

Table 3 shows the results of the bearing capacity (Ph ) of some studied cases. It should be noted that
the value of Ph is not directly proportional to Nβ , meaning that a higher value of Nβ does not necessarily
mean that the value of Ph will be higher. Besides, the same value of Nβ is associated with different
values of Ph depending on the other geotechnical parameters, as shown in (7). On the other hand, it is
observed that under equal conditions (only varying the GSI and, consequently, the parameter “a”),
the greater the GSI, the lower the value of Nβ ; however, as expected, the bearing capacity of the rock
mass is higher.
Table 3. Bearing capacity estimation by analytical method [23] (mo = 15 and uniaxial compressive
strength (UCS) = 1 MPa).

βa Ph
GSI a α (◦ ) σ*01 io2 (◦ ) kh kv ζa Nβ
(MPa) (MPa)
8 0.6 0 0.1 0 0.2 0.2 0.1051 0.00062 6.31 0.664
20 0.55 0 0.1 0 0.2 0.2 0.1309 0.00122 5.42 0.709
100 0.5 0 0.1 0 0.2 0.2 1.8750 0.03556 4.93 9.174
8 0.6 0 0.1 0 0.2 0.1 0.1051 0.00062 6.70 0.704
20 0.55 0 0.1 0 0.2 0.1 0.1309 0.00122 5.73 0.750
100 0.5 0 0.1 0 0.2 0.1 1.8750 0.03556 5.21 9.698
8 0.6 0 0.1 10 0.2 0.2 0.1051 0.00062 3.77 0.397
20 0.55 0 0.1 10 0.2 0.2 0.1309 0.00122 3.28 0.429
100 0.5 0 0.1 10 0.2 0.2 1.8750 0.03556 3.01 5.576
8 0.6 0 0.1 10 0.2 0.1 0.1051 0.00062 4.29 0.451
20 0.55 0 0.1 10 0.2 0.1 0.1309 0.00122 3.72 0.487
100 0.5 0 0.1 10 0.2 0.1 1.8750 0.03556 3.41 6.326
8 0.6 20 0.1 0 0.2 0.2 0.1051 0.00062 4.21 0.442
20 0.55 20 0.1 0 0.2 0.2 0.1309 0.00122 3.77 0.493
100 0.5 20 0.1 0 0.2 0.2 1.8750 0.03556 3.53 6.560
8 0.6 20 0.1 0 0.2 0.1 0.1051 0.00062 4.47 0.470
20 0.55 20 0.1 0 0.2 0.1 0.1309 0.00122 3.99 0.522
100 0.5 20 0.1 0 0.2 0.1 1.8750 0.03556 3.74 6.940
8 0.6 20 0.1 10 0.2 0.2 0.1051 0.00062 2.50 0.262
20 0.55 20 0.1 10 0.2 0.2 0.1309 0.00122 2.28 0.299
100 0.5 20 0.1 10 0.2 0.2 1.8750 0.03556 3.51 6.513
8 0.6 20 0.1 10 0.2 0.1 0.1051 0.00062 2.84 0.299
20 0.55 20 0.1 10 0.2 0.1 0.1309 0.00122 2.58 0.338
100 0.5 20 0.1 10 0.2 0.1 1.8750 0.03556 2.45 4.529

It should be noted that in each graph, an area is indicated, in the lower right part, in which it
was not possible to obtain the mechanical balance due to excessively high load conditions on the
free boundary.

3.3. Charts for Estimation of Bearing Capacity Considering Only a Horizontal Pseudo-Static Load on
Foundation (Seepage Case)
The same representations were realized as in the previous section, where the load capacity
factor Nβ of the analytical Equation (9) can be obtained in Figures 7–9. In this case, each graph
represented corresponds to a determined value of the exponent “a”, the inclination α of Boundary
1 and considering a value of kv = 0 on the foundation boundary. In the same way, in each graph,
the different curves correspond to different inclination angles for the load on Boundary 2 (io2 ) and
the horizontal component of the seepage pseudo-static load (ia ). In this case, of the seepage load,
the abscissa of the normalized main major stress normalized on Boundary 1 (σ*01 ) estimated through
(13) corresponds to the transformation of the original inclination of the load from the static configuration
to the pseudo-static situation, where the horizontal and vertical components will appear according to
the seepage trajectories considered.
Sustainability Sustainability
2020, 12, 100632020, 11, x FOR PEER REVIEW 1421
13 of of 22

a = 0.5 /  = 0 
15
1 i2=0; ia =0.1

Static Case 1 2 i2=0; ia =0.2

10 2 3 i2=10; ia =0.1
3
N

4 4 i2=0; ia =0.3

5 5 i2=10; ia =0.2
5 6
i2=20; ia =0.1
6
7 Non-equilibrium i2=10; ia =0.3
8 boundary
0 7 i2=20; ia =0.2
0,01 0,10 1,00
8 i2=20; ia =0.3
Normalized main major stress, *01

a = 0.5 /  = 10 
15
1 i2=0; ia =0.1

2 i2=0; ia =0.2

Static Case 3 i2=10; ia =0.1

1
10
2 4 i2=0; ia =0.3
N

3 5 i2=10; ia =0.2
4
5 6 i2=20; ia =0.1
5 i2=10; ia =0.3
6
7 7 i2=20; ia =0.2
8 Non-equilibrium
boundary 8 i2=20; ia =0.3
0
0,01 0,10 1,00
Normalized main major stress, *01

Figure 7. Seepage charts

Figure 7. for estimation
Seepage of the bearing
charts for estimation of thecapacity factor Nβfactor
bearing capacity in the
Nßcase of case
in the an exponent from Modified
of an exponent HoekHoek
from Modified and Brown failure
and Brown criterion:
failure a = 0.5.
criterion: a = 0.5.
Sustainability 2020, 12, 10063 14 of 21
Sustainability 2020, 11, x FOR PEER REVIEW 15 of 22

Figure8.8.Seepage
Figure Seepagecharts
chartsfor
forestimation
estimationof
of the
the bearing
bearing capacity
capacity factor Nβß in
in the
the case
caseof
ofan
anexponent
exponentfrom
fromModified
ModifiedHoek
Hoekand
andBrown
Brownfailure criterion:a =a =
failurecriterion: 0.55.
0.55.
Sustainability 2020, 12,
Sustainability 10063
2020, 11, x FOR PEER REVIEW 15 of
1621
of 22

a = 0.6 /  = 0  a = 0.6 /  = 5 
15
15 1 1 i2=0; ia =0.1
1 2
Static Case 1 i2=0; ia =0.1 Static Case
2 3 2 i2=0; ia =0.2
3
4 3 i2=10; ia =0.1
4 2 i2=0; ia =0.2
10 4 i2=0; ia =0.3
10 5 5
3 i2=10; ia =0.1 5 i2=10; ia =0.2

N
N

6 4 i2=0; ia =0.3 6 6 i2=20; ia =0.1

i2=10; ia =0.3
5 i2=10; ia =0.2 5 7 7 i2=20; ia =0.2
5 7
i2=20; ia =0.1
8 i2=20; ia =0.3
6
8 Non-equilibrium i2=10; ia =0.3 8 Non-equilibrium
boundary boundary
0 7 i2=20; ia =0.2 0
0.01 0,10 1.00 0,01 0,10 1,00
Normalized main major stress, *01 8 i2=20; ia =0.3 Normalized main major stress, *01

a = 0.6 /  = 10 
15
1 1 i2=0; ia =0.1
Static Case 2 2 i2=0; ia =0.2
3 3 i2=10; ia =0.1
4 4 i2=0; ia =0.3
10
5 5 i2=10; ia =0.2
N

6 6 i2=20; ia =0.1
i2=10; ia =0.3
7
5 7 i2=20; ia =0.2

8 i2=20; ia =0.3

8 Non-equilibrium
boundary
0
0,01 0,10 1,00
Normalized main major stress, *01

FigureFigure 9. Seepage
9. Seepage chartscharts for estimation
for estimation of theof the bearing
bearing capacity
capacity factorfactor
Nβ inNthe
ß incase
the case
of anofexponent
an exponent
fromfrom Modified
Modified Hoek
Hoek andand Brown
Brown failure
failure a = a0.6.
criterion:
criterion: = 0.6.
Sustainability 2020, 12, 10063 16 of 21
Sustainability 2020, 12, x FOR PEER REVIEW 17 of 22

In this case, since the seepage is typical of dam foundations, it is more realistic to consider cases of
4. Numerical Validation
moderate ground slopes, showing the charts for slopes of the free boundary (α) equal to 0◦ , 5◦ and 10◦ .
In order to compare the results obtained by applying the chart proposed in the present study
4. Numerical
with Validationestimated (through FDM: finite difference method), the same hypotheses
those numerically
(weightless
In orderrock, strip foundation
to compare the resultsand associative
obtained flow law)
by applying were proposed
the chart used in the rock-foundation
in the present study
model.
with those numerically estimated (through FDM: finite difference method), the same hypotheses
The 2Drock,
(weightless models were
strip used to calculate
foundation the cases
and associative flowbylaw)
the finite difference
were used in themethod, employing
rock-foundation the
model.
commercial FLAC software [27], applying the plane strain condition
The 2D models were used to calculate the cases by the finite difference method, employing the (strip footing).
Two-dimensional
commercial FLAC numerical models
software [27], havethe
applying been used
plane bycondition
strain many researchers to solve
(strip footing). problems of
Two-dimensional
foundations
numerical models have been used by many researchers to solve problems of foundations underused
under dynamic loads [28] and in rock masses [29]. Figure 10 shows the model and
dynamic
where the and
loads [28] boundaries were located,
in rock masses at a 10
[29]. Figure distance
showsthat did notused
the model interfere with the
and where result; in allwere
the boundaries the
simulations, the associative flow rule, weightless rock mass and smooth interface at the base of
located, at a distance that did not interfere with the result; in all the simulations, the associative flow rule, the
foundation weremass
weightless rock adopted. The modified
and smooth Hoek–Brown
interface at the base ofconstitutive model
the foundation available
were adopted.in FLAC V.7 was
The modified
used, which corresponds
Hoek–Brown constitutiveto (1). available in FLAC V.7 was used, which corresponds to (1).
model

Figure 10.
Figure 2D model
10. 2D model used
used in
in the
the calculation
calculation through
through FDM
FDM (FLAC
(FLAC 2D).
2D).

It is
It is assumed
assumedthat thatthethe
bearing
bearingcapacity is reached
capacity whenwhen
is reached the continuous medium
the continuous does notdoes
medium support
not
support more load because an internal failure mechanism is formed. In the case under study, duethe
more load because an internal failure mechanism is formed. In the case under study, due to to
presence
the presence of a vertical and a and
of a vertical horizontal force, the
a horizontal vertical
force, the force wasforce
vertical considered unknown;unknown;
was considered therefore,
in the calculation,
therefore, a constant
in the calculation, horizontal
a constant load was
horizontal applied,
load while the
was applied, vertical
while load increased
the vertical load increaseduntil
reaching
until reachingfailure.failure.
Thus, the inclination
Thus, of the load
the inclination of applied
the loadwas also unknown,
applied was alsobecause
unknown,it depended
because on it
the ratio between the vertical and the horizontal components.
depended on the ratio between the vertical and the horizontal components.
Therefore, to
Therefore, to estimate
estimate thethe bearing
bearing capacity
capacity for
for aa determinate
determinate load
load inclination,
inclination, it
it is
is necessary
necessary to to
carry out a series of calculations to find the corresponding combination
carry out a series of calculations to find the corresponding combination of the horizontal (σh) and of the horizontal (σh ) and
vertical (σ
vertical (σvv)) components.
components. Figure Figure11 11shows
showsthe the results
results for
for vertical
vertical loads
loads obtained
obtained for
for the
the case
case studied
studied
(kvv ==0,0,io2io2= =
(k
◦ = 15,
0°,0 m, om=o 15, UCSUCS = 100
= 100 MPa,
MPa, = 65=and
GSIGSI 65 and foundation
foundation width B=B
width = 2.25
2.25 m)a as
m) as a function
function of
of the
the equivalent load
equivalent load inclination. inclination.
 Sustainability 2020, 12, x FOR PEER REVIEW 18 of 22

Sustainability
Sustainability 2020, 12,2020, 12, 10063
x FOR PEER REVIEW 17 of 21 18 of 22

Figure 11. Vertical load (σv) as function of the load inclination on foundation boundary (i2).

The vertical load was applied through velocity increments, and the bearing capacity was
determined from the relation between the stresses and displacements of one of the nodes; in this
case, the Figure
central node of the foundation was considered. In Figure 12a, the displacement of the
11. Vertical load (σv ) as function of the load inclination on foundation boundary (i2 ).
Figure
central 11. Vertical
node load (σv) (abscissa)
of the foundation as function of the
with loadtoinclination
respect on foundation
the load applied boundary
to the ground from (ithe
2).

foundation is represented.
The vertical load was In this figure
applied through is observed that the maximum
velocity increments, and the load
bearingthatcapacity
the groundwas
Thedetermined
vertical
supports loadtheto
is limited
from was the applied
asymptotic
relation between through
value velocity
of the
the stresses curve increments,
represented.
and displacements and
of one thenodes;
of the bearing
in thiscapacity
case, was
Additionally,
the central
determined fromnode a convergence
theofrelation
the foundation
between study was
was considered.carried
the stresses out, consisting
In Figure of the analysis
12a, the displacement
and displacements of the values
of theofcentral
of one of
node in this
the nodes;
thethe
of bearing capacity
foundation obtainedwith
(abscissa) under the different
respect increments
to the load appliedoftothe
thevelocity
groundused.
fromWith a decrease in
the foundation is
case, the central node of the foundation was considered. In Figure 12a, the displacement of the
the value of the
represented. velocity
In this figureincrements,
is observedthe thatresult convergedload
the maximum towards theground
that the final value that is the
supports upper
limited to
central the
node
limit inof
thethe
asymptotic
foundation
theoretical
value ofmethod
(abscissa)
the curve(27b). with respect to the load applied to the ground from the
represented.
foundation is represented. In this figure is observed that the maximum load that the ground
supports is limited to the asymptotic value of the curve represented.
Additionally, a convergence study was carried out, consisting of the analysis of the values of
the bearing capacity obtained under the different increments of the velocity used. With a decrease in
the value of the velocity increments, the result converged towards the final value that is the upper
limit in the theoretical method (27b).

Figure 12. Cont.

Sustainability 2020, 12, 10063 18 of 21
Sustainability 2020, 12, x FOR PEER REVIEW 19 of 22

b
200

Bearing capacity (MPa)

150

100

50

0
1,00E-05
10-5 10-6
1,00E-06 1,00E-07
10-7 10-8
1,00E-08 1,00E-09
10-9

Velocity increments (m/step)

12. Estimation
Figure Figure of bearing
12. Estimation capacity
of bearing by FDM
capacity = 0.47):
by(iaFDM (ia = (a) Stress–strain
0.47): curve at
(a) Stress–strain a point
curve at aof the of the
point
ground; (b) Convergence analysis.
ground; (b) Convergence analysis.

Additionally, a convergence study was carried out, consisting of the analysis of the values of the
In the example studied, a seepage case was studied, where kv = 0, io2 = 0°, mo = 15, UCS = 100 MPa,
bearing capacity obtained under the different increments of the velocity used. With a decrease in the
GSI = 65, α = 0° and B = 2.25 m were adopted (which did not influence the result because of the
value of the velocity increments, the result converged towards the final value that is the upper limit in
assumption of a weightless rock mass). In addition, with io2 = 0°, i2 = arctan(ia) (see (12)). Table 4
the theoretical method (27b).
shows the results obtained numerically (PhFDM) and analytically (PhS&O) (first chart proposed Figure 7)
In the example studied, a seepage case was studied, where kv = 0, io2 = 0◦ , mo = 15, UCS = 100 MPa,
considering different values of ia. According to the margin error ratio observed in Table 4, less than
GSI = 65, α = 0◦ and B = 2.25 m were adopted (which did not influence the result because of the
5%, it can be concluded that the two calculation methods have very similar results.
assumption of a weightless rock mass). In addition, with io2 = 0◦ , i2 = arctan(ia ) (see (12)). Table 4
shows the results
Table 4.obtained
Numericalnumerically (PhFDM
and analytical ) and
results for analytically (PhS&O
bearing capacity (kv =) (first
0, io2 =chart
0°, moproposed
= 15, UCS Figure 7)
= 100 MPa,
consideringGSIdifferent values
= 65 and ia . According to the margin error ratio observed in Table 4, less than 5%,
of m).
B = 2.25
it can be concluded that the two calculation methods have very similar results.
PhFDM PhS&O |𝑷𝒉𝑺&𝑂 − 𝑷𝒉𝑭𝑫𝑴 |
i2 =and
Table 4. Numerical 𝑬𝒓𝒓𝒐𝒓
ia analytical results for bearing capacity (kv = 0,(%)
io2 = =0◦ , mo = 15, UCS = 100 MPa,
(MPa) (MPa) 𝑷𝒉𝑺&𝑂
GSI = 65 and B = 2.25 m).
0 259 270 4.07
i2 = ia
0.1 PhFDM (MPa)
233 PhS&O (MPa)
230 Error (%)= |PhS&O
P
−PhFDM |
1.30
hS&O

0 259 270 4.07

0.2 187 191.5 2.35
0.1 233 230 1.30
0.3
0.2 155.2
187 155.7
191.5 2.35 0.32
0.3 155.2 155.7 0.32
In the numerical calculation, to estimate the bearing capacity, a stress path was formed until
Inwas
thereached,
numericaltaking into account
calculation, the whole
to estimate the wedge
bearingofcapacity,
the ground below
a stress thewas
path foundation. Therefore,
formed until
the graphic output of the vertical component of the total stress tensor
was reached, taking into account the whole wedge of the ground below the foundation. Therefore, at failure was used to
understand
the graphic outputhow
of thethe failurecomponent
vertical mechanism of changed depending
the total stress tensoronat ifailure
a.
was used to understand
Figure 13 shows that the stress
how the failure mechanism changed depending on ia . turned horizontally with an increase in ia, the rotation being
larger when the horizontal force was wider. Additionally, it is noted
Figure 13 shows that the stress turned horizontally with an increase in ia , the that the vertical stress
rotation was well
being
larger distributed in the caseforce
when the horizontal that was
ia = 0.wider. Additionally, it is noted that the vertical stress was well
distributed in the case that ia = 0.
Sustainability 2020, 12, 10063 19 of 21
Sustainability 2020, 12, x FOR PEER REVIEW 20 of 22

13. The variation of the vertical component of the total stress tensor obtained by FDM under
Figure 13.
the foundation.

5. Conclusions
5. Conclusions
An
An optimized
optimizeddesigndesignthat thatadequately
adequately considers
considers thethe negative
negative effects that that
effects natural phenomena
natural phenomenasuch
as earthquakes
such and floods
as earthquakes andintroduce to foundations
floods introduce is highly advisable
to foundations is highlytoadvisable
guarantee to theguarantee
sustainability
the
of infrastructures.
sustainability of infrastructures.
Applying
Applyingaapseudo-static
pseudo-static approach
approach thatthat
considers the seismic
considers force and
the seismic forcethe and
seepage the as an additional
seepage as an
body
additional body force, a series of parameterized charts for estimating the bearing capacityfoundations
force, a series of parameterized charts for estimating the bearing capacity of shallow of shallow
on rock masses
foundations onwere
rockproposed.
masses were The charts
proposed. wereThe calculated
charts through the analytical
were calculated through solution proposed
the analytical
by Serrano
solution et al. [23]bybySerrano
proposed means of a previous
et al. [23] by meansmathematical transformation
of a previous of the transformation
mathematical angles of incidence in
of the
the boundaries of the pseudo-static loads produced by the inertial
angles of incidence in the boundaries of the pseudo-static loads produced by the inertial action. action.
Each
Each chart
chart was
was made
made according
according to to the
the exponent
exponent “a” “a” of of the
the Modified
Modified Hoek Hoek and and Brown
Brown criterion,
criterion,
the inclination α
the inclination α of
of the
the free
free boundary
boundary and and kkvv ofof the
the foundation
foundation boundary boundary (k (kvv =
= 00 in
in the
the seepage
seepage case),
case),
and they were developed
and they were developed based on io2 based on i and the horizontal component of the pseudo-static
o2 and the horizontal component of the pseudo-static load of the
load of the
foundation boundary (called
foundation boundary (calledkkh hininthetheseismic
seismic case
case andandia inia the
in the seepage
seepage case).
case). It is It is noted
noted thathigh
that for for
high confining pressures in Boundary 1, it is not always possible to
confining pressures in Boundary 1, it is not always possible to obtain a bearing capacity value; this obtain a bearing capacity value;
this
limitlimit is demarcated
is demarcated by the by non-equilibrium
the non-equilibrium line.line.
As
As expected,
expected, it it was
was observed
observed that that thethe bearing
bearing capacity
capacity decreased
decreased with with an an increase
increase in in the
the
pseudo-static
pseudo-static load and the original inclination of the load. It was also observed that the value of
load and the original inclination of the load. It was also observed that the value of the
the
bearing
bearing capacity
capacity was
was not not directly
directly proportional
proportional to to the
the bearing
bearing capacity factor N
capacity factor Nβß;; therefore,
therefore, aa higher
higher
value of N β is not associated with a greater value of P h . In addition,
value of Nß is not associated with a greater value of Ph. In addition, the same value of Nß generates the same value of N β generates
different
different values
values ofof PPhh,, depending
depending on on thethe other
other geotechnical
geotechnical parameters.
parameters.
A validation using the finite difference method
A validation using the finite difference method was carried was carried out in aout
particular case. The numerical
in a particular case. The
and
numerical and analytical results, according to the example studied, show a variationthan
analytical results, according to the example studied, show a variation of less 5%.
of less In 5%.
than the
numerical graphic output of the vertical component of the total stress
In the numerical graphic output of the vertical component of the total stress tensor, it is observed tensor, it is observed that the
that the stress turned horizontally with an increase in ia. In addition, it is noted that the vertical stress
was well distributed in the case that ia = 0.
Sustainability 2020, 12, 10063 20 of 21

stress turned horizontally with an increase in ia . In addition, it is noted that the vertical stress was well
distributed in the case that ia = 0.

Author Contributions: Conceptualization, R.G.; methodology, R.G.; validation, R.G.; writing—original draft
preparation, A.A.; writing—review and editing, R.G. and A.A.; supervision, N.S.I. and C.O.M. All authors have
read and agreed to the published version of the manuscript.
Funding: This research was funded by Universidad Politécnica de Madrid.
Acknowledgments: The research described in this paper was financially supported by the Universidad Politécnica
de Madrid by the grant with reference VMENTORUPM20RAGA of the university’s own program for carrying out
research and innovation projects.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Sarma, S.K.; Iossifelis, I.S. Seismic bearing capacity factors of shallow strip footings. Geotechnique
1990, 40, 265–273. [CrossRef]
2. Richards, R.; Elms, D.G.; Budhu, M. Seismic bearing capacity and settlement of foundations. J. Geotech. Eng.
1993, 119, 662–674. [CrossRef]
3. Kumar, J.; Kumar, N. Seismic bearing capacity of rough footings on slopes using limit equilibrium. Geotechnique
2003, 53, 363–369. [CrossRef]
4. Choudhury, D.; Rao, K.S.S. Seismic Bearing Capacity of Shallow Strip Footings Embedded in Slope.
Int. J. Geomech. 2006, 6, 176–184. [CrossRef]
5. Kumar, J.; Rao, V.B.K. Seismic bearing capacity of foundations on slopes. Géotechnique 2003,
53, 347–361. [CrossRef]
6. Kumar, J.; Ghosh, P. Seismic bearing capacity for embedded footings on sloping ground. Géotechnique
2006, 56, 133–140. [CrossRef]
7. Yang, X.L. Seismic bearing capacity of a strip footing on rock slopes. Can. Geotech. J. 2009, 46, 943–954. [CrossRef]
8. Raj, D.; Singh, Y.; Shukla, S. Seismic Bearing Capacity of Strip Foundation Embedded in c-φ Soil Slope.
Int. J. Geomech. 2018, 18. [CrossRef]
9. Nova, R.; Montrasio, L. Settlement of shallow foundations on sand. Géotechnique 1991, 41, 243–256. [CrossRef]
10. Le Pape, Y.; Sieffert, J.P. Application of thermodynamics to the global modelling of shallow foundations on
frictional material. Int. J. Numer. Anal. Methods Geomech. 2001, 25, 1377–1408. [CrossRef]
11. Di Prisco, C.; Pisanò, F. Seismic response of rigid shallow footings. Eur. J. Environ. Civ. Eng. 2011,
15, 185–221. [CrossRef]
12. Veiskarami, M.; Fadaie, S. Stability of Supported Vertical Cuts in Granular Matters in Presence of the Seepage
Flow by a Semi-Analytical Approach. Sci. Iran. 2017, 24, 537–550. [CrossRef]
13. Saada, Z.; Maghous, S.; Garnier, D. Stability analysis of rock slopes subjected to seepage forces using the
modified Hoek–Brown criterion. Int. J. Rock Mech. Min. Sci. 2012, 55, 45–54. [CrossRef]
14. Veiskarami, M.; Kumar, J. Bearing capacity of foundations subjected to groundwater flow. Geomech. Geoengin.
2012, 7, 293–301. [CrossRef]
15. Kumar, J.; Chakraborty, D. Bearing Capacity of Foundations with Inclined Groundwater Seepage.
Int. J. Geomech. 2013, 13, 611–624. [CrossRef]
16. Veiskarami, M.; Habibagahi, G. Foundations bearing capacity subjected to seepage by the kinematic approach
of the limit analysis. Front. Struct. Civ. Eng. 2013, 7, 446–455. [CrossRef]
17. Imani, M.; Fahimifar, A.; Sharifzadeh, M. Upper Bound Solution for the Bearing Capacity of Submerged
Jointed Rock Foundations. Rock Mech. Rock Eng. 2012, 45. [CrossRef]
18. Wang, M.; Chen, Y.F.; Hu, R.; Liu, W.; Zhou, C.B. Coupled hydro-mechanical analysis of a dam foundation
with thick fluvial deposits: A case study of the Danba Hydropower Project, Southwestern China. Eur. J.
Environ. Civ. Eng. 2016, 20, 19–44. [CrossRef]
19. Hoek, E.; Brown, E.T. Empirical strength criterion for rock masses. J. Geotech. Eng. Div. ASCE 1980, 106, 1013–1035.
20. Hoek, E.; Brown, E.T. Practical estimates of rock mass strength. Int. J. Rock Mech. Min. Sci. 1997,
34, 1165–1186. [CrossRef]
Sustainability 2020, 12, 10063 21 of 21

21. Hoek, E.; Carranza-Torres, C.; Corkum, B. Hoek-Brown failure criterion—2002 Edition. In Proceedings of
the NARMS-TAC, Mining Innovation and Technology, Toronto, ON, Canada, 10 July 2002; Hammah, R.,
Bawden, W., Curran, J., Telesnicki, M., Eds.; pp. 267–273. Available online: https://www.rocscience.com/help/
rocdata/pdf_files/theory/Hoek-Brown_Failure_Criterion-2002_Edition.pdf (accessed on 1 September 2020).
22. Serrano, A.; Olalla, C. Ultimate bearing capacity of rock masses. Int. J. Rock Mech. Min. Sci. Geomech.
1994, 31, 93–106. [CrossRef]
23. Serrano, A.; Olalla, C.; González, J. Ultimate bearing capacity of rock masses based on the modified
Hoek-Brown criterion. Int. J. Rock Mech. Min. Sci. 2000, 37, 1013–1018. [CrossRef]
24. Sokolovskii, V.V. Statics of Soil Media; Jones, R.; Schofield, A., Translators; Butterworths Science: London, UK, 1965.
25. Hoek, E.; Kaiser, P.K.; Bawden, W.F. Support of Underground Excavations in Hard Rock; AA Balkema: Rotterdam,
The Netherlands, 1995.
26. Serrano, A.; Olalla, C.; Galindo, R.A. Ultimate bearing capacity at the tip of a pile in rock based on the
modified Hoek–Brown criterion. Int. J. Rock Mech. Min. Sci. 2014, 71, 83–90. [CrossRef]
27. FLAC. Manuals, Complete Set; Itasca: Minneapolis, MN, USA, 2011.
28. Galindo, R.; Illueca, M.; Jiménez, R. Permanent deformation estimates of dynamic equipment foundations:
Application to a gas turbine in granular soils. Soil Dyn. 2014, 63, 8–18. [CrossRef]
29. Benito, J.L.; Moreno, J.; Sanz, E.; Olalla, C. Influence of Natural Cavities on the Design of Shallow Foundations.
Appl. Sci. 2020, 10, 1119. [CrossRef]

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional
affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).