13th World Conference on Earthquake Engineering

Vancouver, B.C., Canada

August 1-6, 2004
Paper No. 3293

BEARING CAPACITY FORMURA FOR SHALLOW FOUNDATIONS

DURING EARTHQUAKE
Yoshito MAEDA1, Tatsuo IRIE2 and Yasuyuki YOKOTA3

SUMMARY
This paper proposes a formula widely applicable for calculating bearing capacity of shallow foundations,
which can evaluate both inclined load action of superstructure and inclined bearing stratum during
earthquake. The formula is derived using seismic coefficient method and admissible velocity field method
from upper bound theorem. Its applicability was verified by series of experiment. In practice, most bearing
capacity formulas assume only the influence of load inclination. However, it was found from the newly
proposed formula and experimental results that in case of strong earthquake most of the present formulas
might have risk of over-evaluating the bearing capacity.
INTRODUCTION
Generally, the inertia force during an earthquake varies with time and place. However, it is known that an
equivalent dynamic model can be made by considering inclined ground, which corresponds to the degree
of inertia (i.e. seismic intensity in seismic coefficient method).
On the other hand, it is typical to consider inclined loads only for inertia force of superstructure in seismic
load-capacity problem of shallow foundation. This is based on the assumption that the inertia of the
superstructure is dominant and the influence of bearing stratum is comparatively small, according to
Yamaguchi [1]. However, this is only valid for a relatively small horizontal seismic coefficient of about
0.2, as indicated in JRA [2]. Moreover, the range of horizontal seismic coefficient where the bearing
stratum can be ignored has never been studied. The seismic design of foundation is shifting to
performance-based-design-method in consequence of recent major earthquakes. Safety is checked for two
earthquake levels, i.e., the ordinary earthquake and the rarely occurring earthquake. Therefore, it is very
important to study the effect of inertia force of bearing stratum to bearing capacity of foundation.
The authors have already proposed a multiple-use bearing capacity formula that considers the inclination
of load and ground; refer to Maeda [3]. It uses the dynamic model of seismic coefficient method and
applies the admissible velocity field method in the upper bound theory of plasticity. In this paper, the
1

Professor, Kyushu Kyoritsu University, Fukuoka Prefecture, Japan. Email: [email protected]

Chief Engineer, CTI Engineering Co. Ltd., Fukuoka Prefecture, Japan. Email: [email protected]
3
Engineer, Oita Prefecture Office, Japan. Email: [email protected]
2

applicability of the proposed formula, and the influence of load and grounds inclination to bearing
capacity are examined by comparing the formula with bearing capacity test results of a two-dimensional
plastic laminated body.
BEARING CAPACITY OF SHALLOW FOUNDATION ACCORDING TO ADMISSIBLE
VELOCITY FIELD METHOD
Breaking mechanism and bearing capacity type considering the load and dip of the ground
An equivalent dynamic model of the ground when inertia force in the supporting ground is uniform and
failure is determined by the horizontal seismic coefficient due to maximum inertia force of superstructure
is shown in Fig.1. The ground can be assumed inclined when the inertia force acts opposite to the
earthquake direction. Under such a condition, a shallow foundation failure mechanism, as shown in Fig.2,
can be obtained. Here, it is assumed that ab and cd are straight lines and bc is logarithmic spiral line.
Thus, the bearing capacity, which considers inclinations of load and ground at the same time, can be
determined.

Figure 1 Input earthquake and equivalent dynamic model

(a)Failure mechanism of whole body

(b) Relationship of inclined load and

active wedge oab

Figure 2 Failure mechanism considering inclinations of load and ground

Here, the inclination angles of the superstructure load and ground vary according to the degree of seismic
response (i.e., vibration mode, response magnification, etc). The inclinations of load () and ground ()
can be express by the following equations where g is the acceleration due to gravity and, s and f are
response accelerations of load and ground, respectively.
tan = s / g (Equation 1)
tan = f / g

(Equation 2)

Moreover, in the case of s =f , the responses of the superstructure and bearing stratum are equal, and
= .
For the failure mechanism illustrated in Fig.2, an upper limit of the bearing capacity is found by equating
internal dispersion energy and external work. Also, the admissible velocity field method in this paper
assumes an associated flow rule (=) where the soils yield condition is defined by compatibility of the
Mohr-Coulomb failure criterion and plastic flow.
Internal dispersion energy can be computed as illustrated in Fig.2. The straight-line part is the product of
adhesive strength and a discontinuous quantity of admissible velocity. Its sum with dispersion energy of
internal area represents the transition zone; refer to Yamaguchi [4].
External work is the sum of the work due to the weight of ground, inclined load Q (q=Q/B) and surcharge
load p, where the ground consists of an active wedge zone, transition zone and passive soil pressure zone.
In the following, the inclined load q is obtained by equating the total internal dispersion energy and the
total external work.
q = cN c + pN q +

1
BN
2

(Equation 3)

N c = c N c 0 , N q = q N q 0 ,

N = N 1 + N 2

(Equation 3-1)

Here, c, q, and are inclination coefficients of load and ground that can be found using equations
4, 5 and 6. Nc0 and Nq0 are the bearing capacity coefficients when the load and ground are not inclined, i.e.
==0. Since N contains both N1 and N2, N becomes N0 when ==0 in Equations 3-1. Equations 8, 91 and 9-2 define these coefficients.
c = q =

cos
cos( ) (Equation 4)

cos
cos( ) (Equation 5)

= tan

(Equation 6)

1
tan( )
sin
+
{(1 + sin ) exp(21 tan ) 1}
sin( ) sin
(Equation 7)

N c0 =

N q0

2 sin cos 2 exp(21 tan )

2
4
=
cos sin ( )
(Equation 8)

N 1 =

sin
cos 2 ( )
cos sin ( )

(9 tan

sin
2

+ 1 cos

{ {3 tan cos( + 1 )

+ sin ( + 1 )}exp(31 tan ) + 3 tan cos( )

+ sin ( )}

sin cos + exp(31 tan )

2
4

N 2 =
+

(Equation 9-1)

sin
[ sin ( ) cos( )
cos sin ( )

(9 tan

sin
2

+ 1 cos

{{3 tan sin ( + 1 )

cos( + 1 )}exp(31 tan ) 3 tan sin ( )

+ cos( )}

+ sin sin + exp(31 tan )

2
4

(Equation 9-2)

The bearing capacity component perpendicular to ground surface is expressed in equation 10.
q v = q cos (Equation 10)

In the above equations, represents the correction factor for ground weight. It corrects the bearing
capacity coefficient for ground weight N that is overestimated when using the general bearing capacity
equation based on Prandtls failure mechanism, as compared to precise values determined by stress
characteristic curve methods; refer to PWRI [5]. This paper assumes =1/2, which is suggested in Maeda
[6].
Comparison to past study results
Figure 3 shows the calculation results of this study in comparison to bearing capacity test results of
centrifugal loading of sand, according to Shioiri et al. [7] and numerical solutions of Ktter equation used
in the stress characteristic curve method. Here, the inclination angles of load and ground are the same
(=) and the angle of internal friction of the ground, , is 46 to enable direct comparison with the past

A bearing capacity fall rate q/q0

study. The objective of this paper is only to check the coefficient N defined in the bearing capacity
formula proposed herein. However, results revealed that the test values agree with the numerical solutions.

1.0
Test values
Ktter
Proposed fomula
0.5

20

10
15
(deg)

Figure 3 Relation of bearing capacity reduction rate and inclination angle; ref. to Shioiri et al. [7]
Comparison to Japanese design standard
The computed results, according to bearing capacity equation proposed herein, are compared to Japanese
design standards for highway bridges (ref. to JRA [8]) and railroads (ref. to MTRB [9]), for the case of
level ground. For highway bridges, the bearing capacity coefficient is used. It considers the effect of load
inclination based on results of Komada [10]. However, in the case of railroad, the general bearing capacity
coefficient is multiplied by a correction factor to account for load inclination, applying the results of
Meyerhof [11].

1.0
c

Proposed fomula

Standard for railroad

Standard for road

0.8

0.6
0.4
0.2
0.0
0

10

20

30

40

( (deg)
)
Load inclination angle
(a) Nc

Bearing capacity
coefficient ratio
N /NN/N0

Bearing capacity coefficient ratio

Nc/Nc0

Figure 4 shows the bearing capacity factor ratio when the load inclination angle is in the range of 0 to
30 and the internal friction angle is 30.

1.0
1.0
Proposed
fomula

Standard
for railroad

Standard
for road

0.8
0.8
0.6
0.6
0.4
0.4
0.2
0.2
0.0
0.0
00

10

20
20

30
30

40
40

( ) (deg)

Load
inclinationangle
(b) N

N ()

Figure 4 Comparison of proposed formula and Japanese design standards

LABORATORY TEST ON BEARING CAPACITY

Experiment Outline
The laboratory test for bearing capacity uses a plastic rod, built-up in a two-dimensional soil layer with
dimensions of W=150cm wide, H=50cm high and L=23cm deep as shown in Fig.5. The loading apparatus
and ground can be rotated, making it possible to combine arbitrary inclination angles of ground () and
load (). The footing is B=10cm wide and L=20cm deep. The load position is adjusted so that the resultant
of applied load acts at the center of footing's base for all load inclination angles , as illustrated in Fig.6. In
order not to restrain displacement at right angles to the load axis, a load is set beforehand to balance the
weight of the apparatus for load inclination adjustment. The displacements taken at point of measurement
V0, H0, and at base of footing V, H are expressed in the following equations with reference to Fig.6.
V = V0cosH0sin

(Equation 11)

H = V0sinH0cos

(Equation 12)

Figure 5 Bearing capacity test apparatus

Figure 6 Illustration of loading apparatus

22

4
P (kPa)
V

Vertical load

33

()P (kPa)

44

P (kPa)
V

Pv(kPa)

0.8

11

00

01
1

22
1

323

V(cm)
V(cm)

Vertical displacementv(cm)

0.6

0.4

0.2

0
3

0.80.8
0.60.6

()P (kPa)

55

Horizontal(parallel) load PH(kPa)

The plastic rod has a diameter of 1.6mm, length of L=20cm and unit weight of d=11.5kN/m3. Its internal
friction angle and adhesive strength determined from laboratory shear testing are =21 and c=0kN/m2,
respectively. A displacement control system using screw jack is applied where the speed of inclined load
application is set to 2.5mm/min. Twenty five tests were performed, using combinations of =0, 5, 10, 20
and =0, 10, 20, 30, 40. The method used to evaluate bearing capacity in this paper is shown in Fig.7. In
this figure, bearing capacity is defined as the intersection of fit curve and the line that bisects the angle of
intersection of two lines tangent to the curve.

0.40.4
0.20.2

00

1
11 2
()
()
(cm)
H
H(cm)

22

Horizontal(parallel) displacement H(cm)

Figure 7 Evaluation method of bearing capacity

Comparison of Experimental Results and Proposed Formula

In this experiment, the bearing capacity coefficient is related only to the weight of ground N, since the dry
plastic rod is set without embedment. Therefore, comparison according to weight of ground N is carried
out as follows.
The figures below show the influence of load and ground inclination to bearing coefficient N. The
vertical axis represents the ratio of bearing capacity coefficient Nt corresponding to load inclination angle
and bearing capacity coefficient N0 when =0. Figure 8(a) illustrates the variation of bearing capacity
coefficient with respect to load inclination angle where ground inclination angle is =0. It is clear from
the figure that bearing capacity decreases as load inclination angle increases. In addition, the figure shows
the test values, computed values according to the proposed equation and prescribed values in
Specifications for Highway Bridges. Comparison reveals that these three values coincide when =10 and
less. For =20, test values exceed computed values and prescribed values. It is conceivable that the
reason for this is the unstableness of the foundation. Since the internal friction angle of plastic is =21,
the foundation will starts to slide at =20. The prescribed value is slightly greater than the computed
value, which implies that Specifications for Highway Bridges gives safer values of bearing capacity
coefficient. This may be explained by the difference in slip planes.

Figure 8(a) Influence of load inclination to N (=0)

Figure 8(b) illustrates the variation of bearing capacity coefficient with respect to ground inclination angle
where load inclination angle is =0. It shows that the bearing capacity coefficient decreases as ground
inclination angle increases. This suggests that the influence of the ground's inertial force is well
evaluated. The reduction rate of bearing capacity coefficient due to ground inclination angle is small
compared to that of load inclination angle . Moreover, the test and computed values coincide well, except
when =20.

Figure 8(b) Influence of ground inclination to N (=0)

Figure 8(c) illustrates the variation of bearing capacity coefficient when =. It is clear from this that
bearing capacity decreases in the case where inclination angles of load and ground are allowed to increase
independently. Moreover, the reduction ratio of the bearing capacity coefficient ratio increases compared
to the case where load and ground are allowed to vary independently. This is because the bearing capacity
coefficient, which is reduced according to inclinations of load and ground, is a summation of results.
Moreover, test values agree well with computed values.

Figure 8(c) Influence of ground inclination to N (=)

Figure 9 shows the failure condition of the ground when the load inclination angle, , is10 and the
ground inclination angle, , is 10, as an example of test results. This explains that the load due to active
wedge zone acts in the direction of ground inclination

Figure 9 Condition of ground failure (=0, =10)

Figure 10 shows the bearing capacity envelope curve for =, where load P is reduced to its vertical and
horizontal components (i.e., normal load PV and horizontal load PH). The figure reveals that the bearing
capacity envelope curve becomes small when inclination angles of load () and ground () become large,
and that the bearing capacity is dependent on these two factors.

Figure 10 Bearing capacity envelope curve

CONCLUSIONS
In this paper, the properties and applicability of the bearing capacity equation are investigated in order to
evaluate the bearing capacity characteristics of shallow foundation during earthquake loading. The method
uses a dynamic model for seismic coefficient method and failure mechanism according to admissible
velocity field method. Results are summarized as follows.
1.

A multi-application bearing capacity equation is proposed. It assumes a failure mechanism that

considers inclinations of load and ground, and applies admissible velocity field method. Using this
equation, evaluation of bearing capacity becomes possible, by considering the degree of inertia force
acting on the superstructure and bearing stratum.

2.

The applicability of the proposed equation is confirmed, since the decrease of bearing capacity
coefficient according to the equation agrees with the test results from a two-dimensional model.

3.

Based on two-dimensional bearing capacity model test results, the bearing capacity coefficient N
decreases according to inclinations of load and ground, similar to results from past studies. The rate
of decrease becomes large as the inclination angle increases.

4.

In regards to reduction rate of bearing capacity N, the effect of load inclination is greater than that of
ground inclination. Furthermore, the reduction rate becomes slightly bigger in the case when
inclinations of load and ground are both considered, compared to the case when only inclination of
load is considered.
ACKNOWLEDGEMENT

The authors gratitude is indebted to Prof. Dawn A. Shuttle of The University of British Columbia for her
great efforts and cooperation in reading and checking the contents of this paper.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.

Yamaguchi H. Soil mechanics (revised edition). Gihodo Publication, 1976: 390.

Japan Road Association. Specifications for highway bridges (e.g. Part IV Substructure, Part V
Seismic Design), 1996.
Maeda Y., Ochiai H., Yokota Y. Bearing capacity equation for shallow foundation considering
inclinations of load and ground, JSCE Journal No.715/III-60, 2002.
Yamaguchi H. Mechanics of soil, Kyoritsu Publication, 1976: 96-97.
Ministry of Construction Public Works Res. Inst. Study on ultimate bearing capacity of shallow
rigid foundation, Public Works Res. Inst. Material no.1611, 1981.
Maeda Y. Study on application of velocity field method in evaluation of bearing capacity of
foundation, Kyushu University doctoral thesis, 2003.
Shioiri M., Yamaguchi H., Kimura M. Bearing capacity on inclined ground based on centrifugal
loading, Proceedings of 31st JSCE Annual Conference III-205, 1975: 365-366.
Japan Road Association. Specifications for highway bridges Part IV Substructure, 1996.
Railway Technical Res. Inst. Railroad structural design standards, 2000.
Komada K. Calculation diagram of soil bearing capacity under two-dimensional inclined load,
Public Works Res. Inst. Report no.135, 1969.
Meyerhof G: The bearing capacity of foundation under eccentric and inclined loads, Proc.3rd Int.
Conf. Soil Mech. and Found. Eng., 1953: 4-24.