Download as pdf or txt
Download as pdf or txt
You are on page 1of 12

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

net/publication/334820628

Nomographs for predicting allowable bearing capacity and elastic settlement


of shallow foundation on granular soil

Article  in  Arabian Journal of Geosciences · August 2019


DOI: 10.1007/s12517-019-4644-1

CITATIONS READS

3 527

6 authors, including:

Maher Omar Am Shanableh


University of Sharjah University of Sharjah
112 PUBLICATIONS   1,331 CITATIONS    293 PUBLICATIONS   3,317 CITATIONS   

SEE PROFILE SEE PROFILE

Khaled Hamad Ali Tahmaz


University of Sharjah University of Sharjah
103 PUBLICATIONS   652 CITATIONS    20 PUBLICATIONS   75 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Conference Paper Blended recycled clay masonry and crushed concrete aggregate in bases View project

Experimental study on shear strength of trapezoidal corrugated steel webs View project

All content following this page was uploaded by Mohamed Arab on 14 January 2020.

The user has requested enhancement of the downloaded file.


Arabian Journal of Geosciences (2019) 12: 485
https://doi.org/10.1007/s12517-019-4644-1

ORIGINAL PAPER

Nomographs for predicting allowable bearing capacity and elastic


settlement of shallow foundation on granular soil
Maher Omar 1 & Abdallah Shanableh 1 & Khaled Hamad 1 & Ali Tahmaz 1 & Mohamed G. Arab 1,2 & Zaid Al-Sadoon 1

Received: 5 February 2018 / Accepted: 10 July 2019 / Published online: 31 July 2019
# Saudi Society for Geosciences 2019

Abstract
Prediction of allowable bearing capacity of granular soil requires an intensive field investigation program. This research proposes
empirical correlations to predict the allowable bearing capacity and elastic settlement of shallow foundation on granular soils. The
current correlation using only standard penetration blow count number and soil unit weight an estimation of the bearing capacity
can be attained. Such correlations can be used at the preliminary stage of estimating the allowable bearing capacity and elastic
settlement of shallow foundation on granular soils and can help site engineers make immediate decisions in cases of field
variations given in soil reports. Moreover, it can be used to create a map for the country in basis of the allowable bearing capacity
based on few soil parameters. In this study, database of granular soil properties obtained from 650 boreholes collected from
various locations in Sharjah, United Arab Emirates, were used to develop the governing predictive equations. Multiple regression
analyses were conducted to develop mathematical models and nomographic solutions to predict the allowable bearing capacity
and elastic settlement of shallow foundation. Following development of predictive equations, a set of data collected from 40
boreholes and 20 zone load tests was used to verify validity of the predictive model. The results indicated that the nomographs
could effectively predict allowable bearing capacity within ± 15% confidence interval and the elastic settlement within ± 10%.

Keywords Shallow foundations . Allowable bearing capacity . Elastic settlement . Multiple regression . Nomograph

Introduction

Allowable bearing capacity and settlement are the two main


Editorial handling: David Giles
criteria that control design practices of shallow foundations so
that safety and serviceability requirements are realized. While
* Maher Omar
momar@sharjah.ac.ae assessing the soil, geotechnical engineers face considerable
uncertainty in terms of soil variability, time effects, construc-
Abdallah Shanableh tion effects, human error and sampling, and laboratory testing
shanableh@sharjah.ac.ae inaccuracy. All these factors make the use of empirical corre-
Khaled Hamad lations preferable to conventional methods, as they are practi-
khamad@shsrjah.ac.ae cal and less expensive than conventional methods.
Ali Tahmaz
A review of the literature revealed that empirical correla-
atahmaz@sharjah.ac.ae tions have been used extensively in many applications of geo-
technical engineering such as pile capacity prediction, site
Mohamed G. Arab
marab@sharjah.ac.ae characterization, modeling soil behavior, liquefaction, earth-
retaining structures, soil penetration resistance, slope stability,
Zaid Al-Sadoon
zalsadoon@sharjah.ac.ae
design of tunnels and underground openings, settlement of
structures, soil compaction, soil permeability and hydraulic
1
Department of Civil and Environmental Engineering, University of conductivity, soil swelling, and many others.
Sharjah, Sharjah, United Arab Emirates Over the past century, researchers have presented several
2
Department of Structural Engineering, Faculty of Engineering, theories and studies to estimate the ultimate bearing capacity.
Mansoura University, Mansoura, Egypt Prandtl (1921) and Reissner (1924) presented theories to
485 Page 2 of 11 Arab J Geosci (2019) 12: 485

estimate ultimate bearing capacity based on the concept of Sharjah Directorate of Town Planning, provided the soil
plastic equilibrium of loading a strip footing until it penetrated reports. Field and laboratory test results extracted from
into the soil. Afterward, these theories were modified for geo- these soil reports include coordinates of the borehole,
technical applications by Terzaghi (1943), who formulated a ground level, water table SPT blow count, modulus of
semi-empirical equation for computing the ultimate bearing elasticity, unit weight, and values angle of internal friction
capacity. This work was followed by an improved solution based on direct shear test conducted in the laboratory.
presented by Meyerhof (1963) proposing a general bearing Table 1 summarizes some of the statistical characteristics
capacity equation similar to that of Terzaghi (1943) which of the collected data.
included different shape and depth factors. Shortly after, it Figure 2 shows the range of grain size distribution extract-
was further modified by Hansen (1970) and then by Vesic ed from the 650 boreholes. From these data, the following
(1973), who developed a bearing capacity prediction equation general observations can be made:
similar to that proposed by Hansen (1970).
During the last decade, several researchers such as & Most soils had fines (P#200) in the range of 5 to 15%.
Soleimanbeigi and Hataf (2005), Padmini et al. (2008), Only a few soils had fines exceeding 20%. Hence, based
Kalinli et al. (2011), Ornek et al. (2012), Marto et al. (2014), on the AASHTO soil classification (AASHTO M 145 or
Nazir et al. (2015), and Gupta et al. (2016) have established ASTM test designation D-3282), and the unified soil clas-
statistical artificial neural network (ANN) models for sification systems (ASTM test designation D-2487), all
predicting the ultimate bearing capacity of soils based on soil soils are granular.
parameters. & The plasticity index (PI) of most soils ranged between 0
In most cases of shallow foundation design, settlement is and 6. Only a few soils had PI exceeding 10.
the governing parameter rather than bearing capacity, particu- & The specific gravity of soil solids (Gs) generally ranged
larly for sands. A number of empirical equations are available between 2.55 and 2.75 with an average of about 2.68.
in the literature that can be used to estimate settlement of
footings in sands, such as in Meyerhof (1956), De Beer
(1965), Schmertmann et al. (1978), and Mayne and Based on the extracted field and laboratory test results,
Illingworth (2010). Hansen’s method (1970) shown in Eq. (1) was used for the
Das and Sivakugan (2007) reported that the most popular estimate of ultimate bearing capacity of vertically loaded
methods for settlement predictions, discussed commonly in foundations on cohesionless soils. For the investigation sites
textbooks, are the ones proposed by Terzaghi and Peck considered in the study, the water table was taken at the base
(1948), Schmertmann (1970), Schmertmann et al. (1978), of the footing and the factor of safety was taken equal to 3 to
and Burland and Burbidge (1985). Sivakugan and Johnson calculate the allowable bearing capacity.
(2004) proposed a probabilistic approach quantifying the un-
certainties associated with settlement prediction methods. In
recent years, some authors such as Shahin et al. (2002), Nazir qu ¼ qN q F qs F qd þ 0:5γBN γ F γs F γd ð1Þ
et al. (2014), and Erzin and Gul (2014) have utilized ANN
modeling of soil settlement and demonstrated a high degree of
success in this field. The purpose of this paper is to develop where B is the width of foundation (m), γ is unit weight of soil
locally calibrated empirical correlations to predict both allow- (kN/m3), q = γDf, Df is the depth of footing embedment (m),
able bearing capacity and elastic settlement of shallow foun- Nq and Nγ are the bearing capacity factors, Fqs and Fγs are the
dation on granular soils. shape factors, and Fqs and Fγs are the depth factors.
For a uniform load applied on a flexible foundation of
dimension L × B at embedment depth Df in a deep elastic
Study area and data collection layer, the elastic settlement by Bowles (1987) shown by Eq.
(2) was used to calculate the actual elastic settlement:
Sharjah is the third largest emirate in the United Arab
Emirates with a total area of 2590 sq. km. It is located in
the geological window between (25 25′N and 25 14′N) 0 1−ν 2
s e ¼ qo B Is I f α ð2Þ
longitude and (55 45′E and 55 20′E) latitude. Over a period Es
of 2 years, in this study, more than 650 boreholes located at
different locations within Sharjah City were collected, as where se is the elastic settlement, B′ is the adjusted smallest
shown in Fig. 1. These boreholes were drilled with depths foundation dimension, q o is applied pressure, ν is the
ranging between 3 and 40 m, mainly for residential and Poisson’s ratio (assumed 0.3), Is is the Steinbrenner influence
commercial buildings. Both, Sharjah Municipality and factor, If is the influence factor which depends on depth of
Arab J Geosci (2019) 12: 485 Page 3 of 11 485

Fig. 1 Borehole locations in Sharjah City, UAE

footing (Df/B), and α is the number of contributing corners.


Table 1 Statistical characteristics of collected data
The results from Eqs. (1 and 2) in the dataset, both allow-
able bearing capacity and settlement will be treated as actual
Statistics SPT Effective Footing Angle of Modulus values of output parameters with which the results of models
blow unit weight, width, internal of
compared. Table 2 summarizes the statistical characteristics of
count, γ’ (kN/m3) B (m) friction, ϕ elasticity,
N (degree) E (kPa) estimated values for allowable bearing capacity and elastic
settlement.
Minimum 5 7.19 1.5 28.80 15652
Maximum 23 11.19 4 33.84 33571
Mean 13.58 9.86 2.73 31.20 25330 Regression analysis and development
Median 14 10.19 3 31.32 26192 of nomographic solution
Standard 4.92 0.86 1.03 1.38 4865.9
Deviat-
ion In this research, a nomograph was developed to predict the
allowable bearing capacity and elastic settlement of shallow
•Depth of footing was limited to 1.5 m below existing ground level foundation on granular soils. The first step for constructing
•Safety factor was taken equal to 3 this chart was to obtain a mathematical model that could pre-
•Water table depth was taken at the base of the footing dict the allowable bearing capacity and elastic settlement
485 Page 4 of 11 Arab J Geosci (2019) 12: 485

Table 2 Statistical characteristics of calculated bearing capacity and type, SPT blow count, modulus of elasticity, unit weight, an-
settlement
gle of internal friction, and footing width.
Statistics Elastic settlement, Se Allowable bearing capacity,
(mm) qall (kPa)

Minimum 7.5 134 Formulation of predictive model


Maximum 70 400
Mean 26.6 260
The analysis to formulate a predictive model was conducted to
Median 26 261.4
determine the best general regression model. Such a model
may incorporate linear or curvilinear effects of one or more
Standard 13.19 73.4
deviation of the independent variables, the dependent variable, or both.
Consequently, the equation sought was of the general form:

properties with reasonable accuracy. Such a model is intended


to perform an in-depth statistical analysis on collected data, f ðYÞ ¼ ao þ Σai f ðX i Þ ð3Þ
with the prime purpose being formulation of a simple yet This function describes the relationship between the gener-
relatively accurate equation. The investigation involved anal- al transformed dependent variable f (Y) and the transformed
ysis of variance (ANOVA) and multivariable regression independent variables f (Xi). The transformation function f can
analysis. be selected with any linear or curvilinear effect (logarithmic,
exponential, etc.) to produce the best-fit model. The causal
effect of each of the independent variables was initially inves-
tigated through a stepwise regression to build the predictive
Data arrangement model. Further, this process was used to find the best-suited
transformation of all variables (dependent and independent).
The experimental data were initially divided into two groups.
The first group, termed dependent variables, included allow-
able bearing capacity (qall) and elastic settlement (Se). The
second group encompassed independent variables. Initially,
Model goodness-of-fit and prediction
several soil properties were considered as independent vari-
accuracy
ables, such as water table, depth and thickness of each soil
The coefficient of multiple correlations, denoted by R2, is usually
Table 3 Regression output for Eqs. 4 and 5 used as a goodness-of-fit measure of a regression model. This
coefficient measures reduction in the model error associated with
Equation 4 Equation 5
use of the independent variables. R2 ranges from 0 to 1. A value
Variable Y Ln (qall) Ln (Se) of 0 is obtained when all of the regression constants ai = 0 (i.e.,
Units kPa mm no relationship exists between the independent and dependent
Regression 2.91 − 0.10 variables) and a value of 1 indicates a perfect fit (i.e., all observed
constant data are exactly predicted by the model).
Std. error of 0.16 0.23 The prediction accuracy of the regression model is com-
estimate
monly assessed by the standard error of estimated square root
Model F-value 19995 5736.6
of the mean squared error (SEE). This term is directly related
Model 0.000 0.000
F-significan- to conditional variance of the dependent variable when knowl-
ce edge of the independent variables is used and to the model’s
R2 0.85 0.75 degrees of freedom. Consequently, a model with good predic-
Degrees of 10869 5588 tion capability will have a low value of SEE.
freedom
(Total)
Variable Xi Ln(N) Ln(γ`) Ln(B) Ln Ln(N) Ln(B)
(- Model selection
q-
al-
l)
In order to choose the predictive function(s) that could represent
Units No./30 cm kN/m3 m kPa No./30 cm m the allowable bearing capacity and elastic settlement of shallow
Coefficient of 0.73 0.30 0.15 0.78 − 0.75 0.84 foundation, the multivariable regression equation(s) should pos-
Xi sess certain statistical features. In general, for a model to be
selected it should:
Arab J Geosci (2019) 12: 485 Page 5 of 11 485

Fig. 2 Range of grain size Particle Size, (mm)


distribution extracted from the

0.001

0.300
0.075
650 soil reports collected

0.005

0.150

0.600

2.36

4.75
1.18

37.5
9.5

300
19

75
100

90

80

70

Pe rce nt Fine r, (%)


60

50

40

30

20

10

#50

3/8''
#30

#16
#200

3/4''
#100

#8

11/2''
#4

12''
3''
(Sieve No.)

& Pass the F and t tests with a pre-selected α-significance & Possess a high value of R2
value (usually 0.05) & Have a low value of SEE

Fig. 3 Actual vs. predicted 450


allowable bearing capacity
400
±15% Interval

350

300
Predicted qall, kPa

250

200

150

100

50

0
0 50 100 150 200 250 300 350 400 450

Measured qall, kPa


485 Page 6 of 11 Arab J Geosci (2019) 12: 485

Fig. 4 Actual vs. predicted elastic 70


settlement
±10% Interval
60

50

Predicted Settlement Se, mm


40

30

20

10

0
0 10 20 30 40 50 60 70

Measured Settlement Se, mm

For practical purposes, an additional imposed condi- Accuracy of derived models


tion could be that the model contains the least number
of independent variables. The accuracy of the derived models was checked by plotting
the values predicted by Eqs. 4 and 5 versus actual collected
data. These plots are shown in Figs. 3 and 4 for qall and Se,
respectively. These plots also show the prediction accuracy
Derived regression model within a confidence interval of ± 15% for qall and ± 10% for
Se. The closeness of the points to the equality line and the
Based on the procedure presented earlier, the SPSS com- inclusion of most within the confidence interval serve to indi-
puter program was used to analyze data shown in Table 1. cate the validity of the derived models.
The stepwise regression yielded that footing width, SPT
blow count, and effective soil unit weight all had a sig-
nificant effect on predicting allowable bearing capacity. Nomographic solution
While SPT blow count values, foundation width, and al-
lowable bearing capacity affected prediction of elastic set- Using Eqs. 4 and 5, Figs. 5 and 6 were developed to estimate
tlement. Consequently, only these independent variables qall and Se, respectively. These nomographs offer a simpler
were used in the final model derivation. The final best- method to estimate these properties. A dashed line in these
fit models obtained were figures indicates the path to be used in evaluating properties.

lnðqall Þ ¼ 2:91 þ 0:73  lnðN Þ þ 0:30  lnðγ 0 Þ þ 0:15


 lnðBÞ ð4Þ
Prediction accuracy of proposed
nomographic solution
lnðS e Þ ¼ −0:75  lnðN Þ þ 0:84  lnðBÞ þ 0:78
Generally speaking, it is practically impossible to predict
 lnðqall Þ−0:10 ð5Þ accurately the allowable bearing capacity and elastic set-
tlement of shallow foundation on granular soils due to the
The regression characteristics for the best functional existence of several variables. The methods presented in
models presented in Eqs. 4 and 5 are summarized in Table 3. this work for the prediction of allowable bearing capacity
Arab J Geosci (2019) 12: 485 Page 7 of 11 485

Fig. 5 Nomograph for estimating Effective Unit Weight (kN/m3 )


allowable bearing capacity of

10
25

9
12

8
7
shallow foundation on granular
soil
20

Obs e rve d SPT, N-Value


15

10

0
Footing Width (m)
1.5 2 3 4
400

Allowable Bearing Capacity (kN/m2 )


350

300

250

200

150

100

50

and elastic settlement are considered estimates only. Yet settlement of shallow foundation represented by Eqs. 4
the simplicity offered by the nomographs developed in and 5 were also compared with data from 20 zone load
this study for such estimations is appealing. However, tests reported from sites near collected boreholes. Based
one must ensure that these charts can produce good pre- on equal settlement values, the allowable bearing capac-
dictions. For this reason, 40 boreholes not included in the ity model developed in this study was compared with
development of the model were used to check the accura- results obtained from zone load tests as shown in
cy of the nomographs. Fig. 9. It is evident that there is good agreement be-
Figures 7 and 8 show measured versus predicted values tween the measured and predicted values that fall within
for allowable bearing capacity and elastic settlement, re- the safe and maximum allowable bearing capacity esti-
spectively. The closeness of the points to the equality line mated by the proposed model. Elastic settlement predic-
attests to the validity of the model. Furthermore, the pre- tion was much closer to the values obtained from the
dicted values for both allowable bearing capacity and regression models and less than the maximum allowed
elastic settlement position within a confidence interval settlement (25 mm), as shown in Fig. 10. These results
of ± 15% and ± 10%, respectively, of measured values. indicate that the model provides a good conservative
The empirical relationships developed for the estima- approximation of settlement and can be used with a safe
tion of the allowable bearing capacity and elastic degree of confidence.
485 Page 8 of 11 Arab J Geosci (2019) 12: 485

Fig. 6 Nomograph for estimating Applied Pressure (kN/m2 )


elastic settlement of shallow

300

200
350

250

150

100
400
foundation on granular soil 25

20

Observed SPT, N-Value


15

10

100
4.0
90
Footing Width (m)

80
3.0

Elastic Settlement (mm)


70

60
2.0
50

1.5 40

30

20

10

Conclusion governing predictive equations. Multiple regression anal-


yses were conducted to develop mathematical models and
The purpose of this research was to contribute to the nomographic solutions to predict the allowable bearing
field of foundation design by developing mathematical capacity and elastic settlement of shallow foundation on
models to predict allowable bearing capacity and elastic granular soils. Results showed that SPT blow count and
settlement for shallow foundation on granular soils. The effective soil unit weight had a significant effect on
study presents regression model to predict the allowable predicting allowable bearing capacity, while SPT blow
bearing capacity to predict allowable bearing capacity. count values, foundation width, and allowable bearing
The importance of such models is that they provide first capacity affected prediction of elastic settlement.
estimate for stakeholders, concerned authorities, and pri- Estimations from these nomographs were shown to be
vate consultants who are involved in the area of founda- accurate within ± 15% for allowable bearing capacity
tion design. In addition, such correlations can help site and within ± 10% for elastic settlement. Further, these
engineers make immediate decisions in case of field var- nomographs were found to produce comparable results
iations given in soil reports. of allowable bearing capacity and elastic settlement when
Data obtained from 650 boreholes collected from vari- compared with data from 20 zone load tests reported from
ous locations in Sharjah were used to develop the sites near collected boreholes. Analysis concluded that the
Arab J Geosci (2019) 12: 485 Page 9 of 11 485

Fig. 7 Measured vs. predicted 450


values of allowable bearing
capacity by nomograph shown in
400
Fig. 5
±15% Interval
350

300

Pre dicte d qall, kPa


250

200

150

100

50

0
0 50 100 150 200 250 300 350 400 450

Measured qall, kPa

developed models are reasonably accurate and can be capacity and elastic settlement of shallow foundation on
used as a simple tool to approximate allowable bearing granular soils of United Arab Emirates.

Fig. 8 Measured vs. predicted 70


values of elastic settlement by
nomograph shown in Fig. 6
60
±10% Interval

50
Predicted Settlement Se, mm

40

30

20

10

0
0 10 20 30 40 50 60 70

Measured Settlement Se, mm


485 Page 10 of 11 Arab J Geosci (2019) 12: 485

Fig. 9 Comparison of allowable 375


bearing capacity by analysis

Predicted Allowable Pressure, kPa


developed in this study (Eq. 4)
and zone load test 325

275

225

175

125

75
75 125 175 225 275 325 375

Measured Allowable Pressure, kPa - From Zone Load Test

Although the results of the regression models were future research, and future studies could extend the find-
promising, there is clearly room for improvement. ings of the current study by investigating the effect of
Many factors affect prediction of allowable bearing ca- these factors and others on the proposed regression
pacity and elastic settlement but were not included in our models. To generalize the research conducted in this
models. Examples include depth of water, load assump- study, it is suggested that these models for allowable
tion (assumed vertical and concentric), shape of founda- bearing capacity and settlement be developed to cover
tion (assumed square), and factor of safety (assumed to more areas and cities in UAE.
be three). These factors provide a beginning point for

Fig. 10 Comparison of elastic 30


settlement by analysis developed
in this study (Eq. 5) and zone load Allowable Elastic settlement 25 mm
test

25
Predicted Elastic Settlement, mm

20

15

10

5
5 10 15 20 25 30

Measured Elastic Settlement, mm - From Zone Load Test


Arab J Geosci (2019) 12: 485 Page 11 of 11 485

References ANN approach. Proc. of the Intl. Conf. On advances in civil, struc-
tural and mechanical engineering – CSM 2014, London, pp 20–24
Bowles JE (1987) Elastic Foundation Settlements on Sand Deposits. Nazir R et al (2015) An artificial neural network approach for prediction
Journal of Geotechnical Engineering, ASCE 113(8):846–860 of bearing capacity of spread foundations. Sand Jurnal Teknologi
Burland JB, Burbidge MC (1985) Settlement of foundations on sand and (Sciences & Engineering) 72(3):9–14
gravel. Proc Inst Civ Eng 78(1):1325–1381 Ornek M, Laman M, Demir A, Yildiz A (2012) Prediction of bearing
Das BM, Sivakugan N (2007) Settlement of shallow foundations on capacity of circular footings on soft clay stabilized with granular
granular soil—an overview. Int J Geotech Eng 1(1):19–29 soil. Soils Found 52(1):69–80
De Beer EE (1965) Bearing capacity and settlement of shallow foundations Padmini D, Ilamparuthi K, Sudheer KP (2008) Ultimate bearing capacity
on sand. In: International proc. of the bearing capacity and settlement of prediction of shallow foundations on cohesionless soils using
foundations symposium. Duke University, Durham, pp 15–34 neurofuzzy models. Comput Geotech 35(1):33–46
Erzin Y, Gul TO (2014) The use of neural networks for the Prandtl L (1921) On the penetrating strengths (hardness) of plastic con-
prediction of the settlement of one-way footings on cohe- struction materials and the strength of cutting edges. Z Angew Math
sionless soils based on standard penetration test. Neural Mech 1:15–20
Comput & Applic 24(3–4):891–900 Reissner H (1924) Zum Erddruck problem (concerning the earth–
Gupta R, Goyal K, Yadav N (2016) Prediction of safe bearing capacity of pressure problem). In: International proceedings of the first
noncohesive soil in arid zone using artificial neural networks. Int J International Congress of Applied Mechanics. Delft, Germany, pp
Geomech 16(2):04015044. https://doi.org/10.1061/(ASCE)GM. 295–311
1943-5622.0000514 Schmertmann JH (1970) Static cone to compute static settlement over
Hansen JB (1970) Revised and extended formula for bearing capacity. sand. J Soil Mech Found Div ASCE 96(3):1011–1043
Danish Geotech Inst Copenhagen Bull 28:5–11 Schmertmann JH, Hartman JP, Brown PR (1978) Improved strain influ-
Kalinli A, Cemal Acar M, Gunduz Z (2011) New approaches to deter- ence factor diagrams. J Geotech Eng Div ASCE 104(8):1131–1135
mine the ultimate bearing capacity of shallow foundations based on Shahin MA, Maier HR, Jaksa MB (2002) Predicting settlement of shal-
artificial neural networks and ant colony optimization. Eng Geol low foundations using neural networks. J Geotech Geoenviron
117:29–38 128(9):785–793
Marto A, Hajihassani M, Momeni E (2014) Bearing capacity of shallow Sivakugan N, Johnson K (2004) Settlement predictions in granular soils:
foundation’s prediction through hybrid artificial neural networks. a probabilistic approach. Geotechnique 54(7):499–502
Appl Mech Mater 567:681–686
Soleimanbeigi A, Hataf N (2005) Predicting ultimate bearing capacity of
Mayne P, Illingworth F (2010) Direct CPT method for footing response in
shallow foundations on reinforced cohesionless soils using artificial
sands using a database approach. The 2nd international symposium
neural networks. Geosynth Int 12(6):321–332
on cone penetration testing, Huntington Beach
Meyerhof GG (1956) Penetration tests and bearing capacity of cohesion- Terzaghi K (1943) Theoretical soil mechanics. John Wiley and Sons,
less soils. Soil Mech Found Div, ASCE 82:1), 1–1),19 New York
Meyerhof GG (1963) Some recent research on the bearing capacity of Terzaghi K, Peck RB (1948) Soil mechanics in engineering practice, 1st
foundations. Can Geotech J 1:16–26 edn. John Wiley & Sons, New York
Nazir R et al (2014) Prediction of spread foundations’ settlement in co- Vesic AS (1973) Analysis of ultimate loads of shallow foundations. J Soil
hesionless soils using a hybrid particle swarm optimization-based Mech Found Div ASCE 99:45–73

View publication stats

You might also like