d

2n International Conference on New Developments in Soil Mechanics and Geotechnical Engineering,

28-30 May 2009, Near East University, Nicosia, North Cyprus

Observed and predicted settlement of shallow foundation

Rasin Dzceer
Kaskta A..-stanbul, Turkey

KEYWORDS: Settlement, Shallow foundations, Load tests, CPT, SPT, Finite Element Method.
ABSTRACT: The objective of this paper is to compare the predictive capabilities of different
methods of estimating settlements of shallow foundations on sands. For this purpose 2.10 x.2.10 m
square concrete footing was statically load tested. Prior to load test, standard penetration test
(SPT), cone penetration test (CPT) and laboratory tests were performed to determine the
engineering properties of soil layers. Predictions of footing settlement were performed by
conventional (semi-empirical) and finite element method (FEM). The results of static load test
revealed that the settlements were over predicted by Finite element method. Finite element analysis
using either SPT or CPT derived input parameters provided conservative settlement estimates.
However, most of the empirical methods employed in this study provide reasonable estimates
using CPT derived parameters as input.
1 INTRODUCTION
The design of shallow foundations on cohesionless soils is often controlled by settlement, rather
than bearing capacity limitations. Several methods have been proposed for predicting settlement of
shallow foundations on cohesionless soils. Settlement prediction methods can be divided into two
categories, conventional or semi-empirical methods and the finite element based methods.
Semi-empirical methods are the predominant techniques used to estimate settlements of shallow
foundations on cohesionless soils. These methods have been correlated to large databases of tests
such as the SPT and CPT (Kimmerling 2002).
In this paper a 2.1x.2.1 m precast concrete footing was statically load tested to 1.50 times the
proposed design load of 200 kPa to examine the settlement behaviour of the footing. Prior to load
test, SPT, CPT and laboratory tests were performed to determine the engineering properties of soil
layers. Settlement of the footing resting on cohesionless soil was estimated by several methods
based on semi-empirical correlation and FEM. Measured settlement of the footing was compared
with the settlements estimated by conventional methods and FEM.
2 REVIEW OF THE SETTLEMENT PREDICTION METHODS
Allowable bearing pressure for footings on sand is generally limited by the consideration of
settlement rather than safety against bearing capacity failure. Due to the difficulties of obtaining
relatively undisturbed samples of cohesionless soils, semi-empirical approaches rely on
correlations between the observed foundation settlements and some parameters from in situ tests.
Many methods have been proposed to predict the settlement of foundations on cohesionless soils
based on SPT N values and CPT point resistance, qc. Some of the methods used to estimate
settlement are summarized in Table 1. There are several other methods used to estimate settlement
of foundations based on dilatometer (DMT) and pressuremeter (PMT) derived parameters.

590

2n International Conference on New Developments in Soil Mechanics and Geotechnical Engineering,

28-30 May 2009, Near East University, Nicosia, North Cyprus

In addition to the conventional approaches in estimating settlement of the shallow foundations,

new methods and techniques are becoming available as more sophisticated electronic and
computational tools are being developed. These include centrifuge modeling (Sargand et al. 1997),
nondestructive test methods such as the wave-activated stiffness (WAK) test (Briaud and Gibbens
1997), and neural networks (Shahin, et al. 2002).
Several studies were performed to compare the predictive capabilities of different methods of
estimating settlements of shallow foundations on sands.
Gifford et al. (1987) concluded that the methods proposed by DAppolonia et al. (1967) and by
Burland-Burbidge (1984) were more accurate than the other methods. The Peck-Bazaraa method
had a tendency to underpredict the field settlement, while the methods by Hough and Schmertmann
(1970) often overpredicted the field settlement.
Briaud and Gibbens (1997) conducted a survey among bridge and foundation engineers for
research work for FHWA. Briaud and Gibbens concluded that the best predictions resulted from
the methods by Briaud (1992), Burland-Burbidge (1984), Peck-Bazaraa and Schmertmann (1986).
Briaud (1992) and Burland-Burbidge (1984) were somewhat conservative for their methods, while
the other two were slightly unconservative.
Berardi and Lancellotta (Lancellotta 1995) have compared the reliability and accuracy of
different methods. They concluded that the most accurate empirical methods appear to be methods
suggested by Burland and Burbridge and DAppolania et al.
3 SOIL CONDITIONS
Standard penetration test (SPT), cone penetration test (CPT) and laboratory tests were performed to
determine the engineering properties of soil layers. The soil investigation has revealed that very
loose to loose silty sand up to 4 m depth, followed by dense to very dense silty sand down to 12 m
exist at the site. The ground water table was encountered at a depth of 2 m below ground level. The
results of SPT and CPT are presented in Figure 1.
Cone Tip Resistance
qc (MPa)
0

SPT N
Blows / 30cm

Friction Ratio.
fR (%)
0 2 4 6 8

0 10 20 30 40 50

DESCRIPTION
( SPT )

(SP - SM)

Depth ( m )

Depth ( m )

Light brown to dark

gray, very loose to
medium dense, fine to
medium SAND with
traces of silt

Depth ( m )

Depth ( m )

0 10 20 30 40 50

Sleeve Friction
fs (kPa)
0 100 200 300 400

Figure 1: SPT and CPT Results

591

Observed and Predicted Settlement of Shallow Foundation

Dzceer, R.

Table 1 Summary of Settlement Prediction Methods

METHOD

EXPRESSION FOR
SETTLEMENT

D'Appolonia
(1967)

S 0 1

Department of the
Navy (1982)

Peck and Bazaraa

(Anderson et al.
2007)

DEFINATIONS

EXPLANATIONS

S = settlement (inches);0 = embedment influence

factor; 1= compressible strata influence factor;
q
= applied pressure (tsf); M = modulus of
compressibility (tsf). B = footing width (ft)

qB
M

4q B
K v1 B 1

2q 2 B
S CD CW

N B 1

KV1 = modulus of subgrade reaction (tons/ft3);

Valid for B 20 feet

CD = embedment correction factor;

Cw = water table correction factor;
N = corrected SPT-N value;

C D 1 0.4
Cw

C f
q

v
v '
Dw
D B
f

C w 0.5 0.5

Cw = water table correction factor;

N1 = average corrected SPT-N value within depth of
1B below the base of footing.

Peck-HansonThornburn
(D'Appolonia
1967)

Anagnostropoulos
(1991)

Bowles (1996)

Burland and
Burbidge (1984)

1.25( L / B )
S 1 2 3
Bq '
0.25 ( L / B )

Meyerhof
(Anderson et al.
2007)

Schultze and
Sherif (Anderson
et al. 2007)

Terzaghi and Peck

(Anderson et al.
2007)

Buisman- De Beer
(1965)

1.2

(1 2 )qB '
Is I f
Es

S= settlement (ft); 1 = a constant (0.14 for normally

consolidated sands; 0.047 for overconsolidated sands);
2 = compressibility index; and 3 = correction for the
depth of influence; q' = applied stress at the level of
foundation (tsf);

8q
N'
12 q B

N ' B 1

q = applied bearing pressure (ksf); B = footing width

(ft); N ' Corrected SPT- N value

S = (in); C1 = foundation depth correction factor; C2 =

soil creep factor; q = applied pressure; Iz = strain
influence factor; and Es = modulus of elasticity.

Zz
I
S C1C2 q z z
0 Es

3q 2 B
S C D Cw

N B 1

CD = embedment correction factor; Cw = water table

correction factor; N = average uncorrected SPT-N
value for depth B below the base of footing; q =
applied pressure (tsf); and B = footing width (ft).

H =thickness of layer; C=Compressibility of sand

'o =Effective stress; v =Change in Effective stress

H
0 '
log

C
0'

S Hc
i 1

S H
i 1

B 4 feet
B 4 feet
N ' 15 0.5 N 15
D f
C1 1 0.5

q
t
C2 1 0.2 log

0.1

f = influence factor; q = applied bearing pressure (tsf);

B = footing width (ft); N = average SPT-N value
within 2B from the base of footing; and Df = footing
embedment depth (ft).

fq B
0. 4 D f
0.87
N 1
B

Janbu (CFEM
1992)

N1 2 N for 'v 0

S = settlement (mm); q = applied bearing pressure

(kPa) B = footing width (m);
N = average uncorrected SPT-N value
= Poissons ratio; q = applied bearing pressure
(ksf);B = B/2 for footing center and = B for footing
corner (ft); Es = modulus of elasticity of bearing soil
(ksf); Is and If influence factor.

2.37q 0.87 B 0.7

Hough
(Kimmerling 2002)

20
N for 0 'v 0.25tsf
'v

N1 0.77 log

N1 0.4 N for 0 'v 0.25tsf

Schmertmann
(1978)

q
0.11Cw N1

due to applied load; qc =Cone resistance

H =thickness of layer ; C=Bearing Capacity Index

' v
1
log o

C'
'o

'o =Effective stress; v =Change in Effective stress

1 '1 'o

mj 'r 'r
j

due to applied load

H =thickness of layer 'o =Initial Effective stress;

'1 = final effective stress; 'r =Reference stress 100

kPa; m =modulus number; j=Stress exponent=0.50 for
sandy and silty soils

592

CW =1.0 for Dw 2B;

CW = 2.0 for Dw B
CD=1.0-

C =1.5

D
D
D
4B
4B
qc

0'

2n International Conference on New Developments in Soil Mechanics and Geotechnical Engineering,

28-30 May 2009, Near East University, Nicosia, North Cyprus

4 LOAD TEST RESULTS

Full scale footing load test was conducted at site. A 2.1 x 2.1 m square concrete footing was used
for the test. The test was conducted at the proposed foundation embedment depth. Load test set up
is presented in Figure 2.

Figure 2 Load Test Set up

The full scale load test was performed on a precast footing up to 1.50 times the proposed safe
bearing capacity of footing. The settlement of the footing and the applied loads recorded during the
test are presented in Table 2.
Table 2 Measured Settlements vs. applied test loads
Loading Sequence (kPa)

Measured
Settlement
(mm)

40

80

120

160

180

200

220

240

260

280

300

0.6

4.3

7.4

8.9

10.2

12.5

13.8

15.2

17

18.8

22.4

5 INPUT PARAMETERS FOR EMPRICAL AND FINITE ELEMENT METHODS

Numerical calculations were performed with finite element method. Plaxis 9.0 (Brinkgreve and
Broere 2008) was used for this purpose. The square footing is represented by an equivalent area
circular footing using axisymetric conditions. The Mohr-Coulomb model was used to conduct the
analysis. The geometry and the generated mesh are given in Figure 3.
The angles of shearing resistance were correlated from SPT N values from the following
equation:
=53.88-27.6x10(-0.0147 N )
(1)
The angles of shearing resistance were correlated from CPT using the correlation based on
the effective overburden pressure v, qc values (Robertson and Campanella 1988)
The following correlations were used to obtain the Modulus of Elasticity with SPT N values and
CPT qc values for normally consolidated sands (Bowles 1996):
E (kPa) =500(N+15)
(2)
E (kPa) =2-4 qc
(3)
The following semi emprical relationship between the soil modulus and the adjusted cone tip
resistance, qc was used to obtain modulus number, m in Janbu method (Canadian foundation
Engineering Manual 1992), as proposed by Massarsch (1994):

593

Observed and Predicted Settlement of Shallow Foundation

Dzceer, R.

ma

qc M

(4)

m = Modulus number ;

a =Emprical modulus modifier, which depends on soil type

r =Reference stress=100 kPa

qc M =Stress adjusted cone stress;

qc M qc

r
'm

(5)

qc = Unadjusted cone resistance;

'm = Mean effective stress

'm = 'v 1 2 K 0
3

(6)

K 0 =Coefficient of horizontal earth pressure

Input parameters used in finite element analysis and empirical methods are given in Table 3.
Table 3 Soil properties from in situ tests
SPT
Description of Soil
Layers (CPT)

Depth
(m)

CPT

( bl/30
cm)

(kN/m3)

(Deg)

E
(MPa)

(MPa)

(Deg)

(MPa)

qc

SAND-GRAVELLY SAND

-1.50 to -2.00

12

19

35

13.50

18

48

48

SAND-SILTY SAND

-2.00 to -3.50

18

32

11

41

21

SANDY SILT- CLAYEY SILT

-3.50 to -4.00

18.5

29

1.5

37

4.5

SAND

-4.00 to -6.00

23

18

41

19

15

43

45

Figure 3 Finite element mesh for analysis

594

2n International Conference on New Developments in Soil Mechanics and Geotechnical Engineering,

28-30 May 2009, Near East University, Nicosia, North Cyprus

6 COMPARISON OF SETTLEMENT PREDICTION METHODS

In this study, ten conventional methods among the available methods have been selected to be
incorporated in settlement predictions.
The comparison of settlement predicted with conventional methods and FEM are summarized
in Table 4 and Table 5 respectively. Comparison of predicted versus measured settlement is
presented in Figure 4.
Table 4 Measured versus predicted settlements by conventional methods
Settlements (mm)

Loading Sequence (kPa)

Measured
Buisman- De Beer (3)

60
2.20
6.5

120
7.34
11.3

180
10.2
15.2

240
15.2
18.5

300
22.4
21.4

Burland and Burbidge (1)

Elastic Theory (2)
Anagnostropoulos (1)
D'Appolonia (1)
Hough (1)
Schmertmann (3)
Schultz & Sherif (1)
Department of the Navy (NAVFAC)
Janbu Tangent Modulus (1), (2)

6.2
3.1
7.1
2.1
13.8
6.3
3.2
6.9
4.2

12.3
6.1
13.1
4.3
23.3
12.6
6.4
13.8
10.5

18.5
8. 9
18.5
6.4
30.7
18.9
9.7
20.6
15.8

24.6
12
23.8
8.5
36.9
25.2
12.9
27.5
20.4

30.8
14.9
28.9
10.7
42.1
31.4
16.1
34.4
24.7

(1), ( 3)

(1) SPT based Method; (2) Elastic Theory base Method; (3) CPT based Method

Table 5 Measured Settlements versus finite element analysis with Plaxis using insitu data
Settlements (mm)

Loading Sequence (kPa)

Measured
SPT

60
2.20
3.5

120
7.34
13.2

180
10.2
25.3

240
15.2
39.1

300
22.4
54.1

CPT

1.7

7.4

14.3

21.5

28.9

7 DISCUSSION OF RESULTS
Examination of Table 4 indicates that the six of the conventional methods investigated in this
study, overpredict the settlement of the footing. The most accurate settlement was estimated with
the Buisman - De Beer method, 21.4 mm. The next most accurate method is the Janbu method,
with the settlement of 24.7 mm at 300 kPa.
On the other hand, the Hough method is the least accurate method with 42 mm of total
settlement. This method is considered to be the most conservative among conventional methods in
predicting settlement in sands.
However, the previous studies have shown that, Hough method overpredict settlement by a
factor of 1.8 -2.0 (Gifford et al 1987). It is interesting to note that, an overprediction factor of 1.88
was obtained in this study for Hough method. Following the Hough method Navfac method is the
second conservative method with a total settlement of 34 mm. On the other hand, DAppolania
method, underpredicted the settlement by a factor of 0.48, which is very close to the factor of 0.50
determined in previous studies (Duncan and Tan 1991). Elastic and Shultz - Sherif methods also

595

Observed and Predicted Settlement of Shallow Foundation

Dzceer, R.

underpredict the settlement. The other methods; the Schmertmann, Burland - Burbridge and
Anagnostrospoulos methods are situated in the middle.
As for the finite element analysis, better accuracy of the estimation is obtained using the input
data from CPT testing. The results of settlement estimate corroborate the conclusion from the
Anderson et al (2007) studies. The settlement predicted from the CPT derived input parameters
was smaller than SPT as the CPT estimated modulus of elasticity and angle of shearing resistances
are higher. The predicted settlements from the SPT and CPT input parameters are 54.1 mm and
28.9 mm respectively. The settlement predicted from the SPT input parameters is less accurate then
CPT.

Applied Pressure (kpa)

0

50

100

150

200

250

300

350

0
5
10

Settlement (mm)

15
20
25
30
35
40
45
50
55
60

Measured
Navfac
D'Appolania
Shultz-Sherif
Burland-Burbridge
Anagnostropoulos
Elastic
Buisman De Beer
Hough
Plaxis (CPT)
Plaxis (SPT)
Janbu
Schmertman-1978

Figure 4: Comparison of the predicted and measured settlements.

8 CONCLUSIONS
1. A static load test was conducted to study the settlement behaviour of the footing. The CPT and
SPT data were used to estimate the settlement of shallow foundation on sand.
2. Among the CPT based conventional methods, Buismann-De Beer, provide more accurate
estimations of settlement.
3. Janbu method using CPT derived modulus number m, provide good settlement estimate. The
correlations proposed by Massarsch provide accurate estimates of modulus number.
4. The correlated input parameters from the CPT data are more consistent than the SPT blow
count in both conventional methods and finite element method.
5. Finite element analysis using CPT derived input parameters provided reasonable settlement
estimates whereas the SPT derived input parameters provided poor settlement estimates.
6. The settlement estimations using FEM with CPT and SPT derived parameters corroborate the
results obtained from the previous studies.

596

2n International Conference on New Developments in Soil Mechanics and Geotechnical Engineering,

28-30 May 2009, Near East University, Nicosia, North Cyprus

REFERENCES
Anagnostopoulos, A.G., Papadopoulos, B.P.and Kavvadas, M.J.(1991). SPT and Compressibility of
Cohesionless Soils. Proceedings of the 2nd European Symposium on Penetration Testing, Amsterdam.
Anderson B.J., Townsend F. C. and Rahelison L. (2007). Load testing and settlement prediction of shallow
foundation. Journal of Geotechnical and Geoenvironmental Engineering ASCE. Vol 133, No 12:14941502
Briaud, J-L (1992) The pressuremeter. Balkema,Brookfield,Vt.
Briaud, J-L., and Gibbens, R. (1997). Large Scale Tests and Database of Spread Footings on Sand. Report
No. FHWA-RD-97-0680, Federal Highway Administration, McLean, VA,
Brinkgreve, R. B. J., and Broere, W. (2008). Finite element code for soil and rock analysis. Plaxis Version
9.0 Delf University of Technology and Plaxis b.v. Netherlands.
Bowles J. E. (1996). Foundation Analysis and Design 5th Ed. Mc Graw Hill New York
Burland J.B.,and Burbidge, M.C.(1984) Settlement of foundations on sand and gravel. Institution of Civil
Engineers, Glaskow and West Scotland Association, Glaskow , Scotland.
Canadian Foundation Engineering Manual 1992. 3rd Ed. BiTech Publishers Ltd. Richmond, Canada
DAppolonia D. J., DAppolonia D. and Brissette R.F. (1967) Settlement of spread footings on sand. Journal
of Soil mechanics and foundation Division . ASCE., Vol 94, No SM 3 : 735-760
De Beer. (1965) Bearing capacity and settlement of shallow foundations on sands. Symposium on bearing
capacity and settlement of foundations. Duke University. Durham N.C :315-313
Duncan J.M. and Tan C.K. (1991). Settlement of footing on sands- accuracy and reliability. Geotechnical
Engineering Congress, Boulder Colorado. ASCE Geotechnical Special Publication No: 27 Edited by
Francis G., Campbell DW A. and Harris D.W
Gifford, D. G., Kraemer, S. R., Wheeler, J. R., and McKown, A. F. (1987). Spread Footings for Highway
Bridges. Report No. FHWA/RD-86/185, Federal Highway Administration, U.S. Department of
Transportation, Washington, D.C.
Kimmerling Robert E. (2002). Shallow Foundations Report No. FHWA-SA-02-054, Federal Highway
Administration, McLean, VA
Lancellotta Renato. (1995) Geotechnical engineering , Balkema, Rotterdam.
Massarsch K.R. (1994) Settlement analysis of compacted fills. Proceedings of the 13th International
Conference on Soil Mechanics and Foundation Engineering. ICMSFE, New Delhi. Vol I: 325-328
Department of the Navy. (1982). Soil Mechanics. Design Manual 7.1. NAVFAC DM-7.1. Naval Facilities
Engineering Command, Alexandria, VA: 348
Robertson, P. K., Campanella, R. G. (1988). Guidelines for using the CPT, CPTU and Marchetti DMT for
Geotechnical Design Vol II. Report no: FHWA-PA-87-023+84-24. U.S. Department of Transportation,
Washington, D.C.
Sargand S.M., Hazen G.A., Masada T. (1997). Field and laboratory evaluations of spread footings for
highway bridges. FHWA Report no: OH/98-017
Sargand S.M., Masada T. and Abdalla B. (2003). Evaluation of cone penetration test based settlement
prediction methods for shallow foundations on cohesionless soils at highway bridge construction sites.
Journal of Geotechnical and Geoenvironmental Engineering ASCE., Vol 129, No 10 : 900-908
Shahin, M.A., Maier H.R., Jaksa M.B., (2002). Prediction settlement of shollow foundations using neural
network. Journal of Geotechnical and Geoenvironmental Engineering ASCE., Vol 128 No:9 : 785-793
Schmertmann, J.H (1970)Static cone to compute static settlement over sand Journal of Soil mechanics and
foundation Division . ASCE, Vol 96., No SM3 : 1011-1043
Schmertmann, J.H., Brown P.R., and Harman J.P (1978). Improved strain influence factor diagrams. Journal
of Soil mechanics and foundation Division . ASCE, Vol 104, No 8 : 1131-1135

597