PILED RAFT FOUNDATIONS –

CHARACTERISTICS, DESIGN AND

APPLICATIONS

Harry G. Poulos
Tetra Tech Coffey, Australia

Lecture to GEG
20 September 2023
OUTLINE
• The mechanics of rafts with piles

• Design concepts & issues

• Overall load-settlement behaviour

• Design for localized behavior under columns

• Parameter assessment

• Examples of applications to tall buildings

FOUNDATION OPTIONS
DESIGN PHILOSOPHY

• Different from normal piles

• Share load between raft

and piles

• Raft provides extra load

capacity for piles

• Focus more on stiffness

than capacity
ANOTHER PERSPECTIVE
• The piled raft foundation can be considered as a pile-enhanced raft.

• The piles act to reduce the net load on the raft.

• If the pile capacity has not been fully mobilized, then the pile acts as a
spring that carries some of the column load and so reduces the load that
the raft has to carry.

• Even if a pile has reached its full capacity, it provides a negative (upward)
load that counteracts the downward column load.

• Thus, a piled raft (or a pile-enhanced raft) is able to be very effective in

reducing the raft thickness requirements while increasing the foundation
stiffness at the same time.
CIRCUMSTANCES FAVOURABLE FOR PILED RAFTS

• Where the raft can provide a

reasonable amount of stiffness and
load capacity, e.g.

 Relatively stiff clay profiles

 Relatively dense sands, weak
rocks

• Where soil movements due to

external causes do not occur

6
CIRCUMSTANCES UNFAVOURABLE FOR
PILED RAFTS
• Soft clays near the surface

• Loose sands near the surface

• Compressible layers at depth

• Where consolidation settlements may occur

 Soil may move away from raft

• Where swelling movements may occur

 Soil may impose additional uplift loading on piles
EARLY MODEL TEST DATA (Whitaker, 1957)

Optimum number

Optimum number
THE MECHANICS OF RAFTS + PILES

Ultimate geotechnical capacity:

Depends on mechanism of failure; σraft

• n x single pile capacity + net raft capacity for widely spaced piles
• Block capacity + external raft area capacity for closely-spaced piles

With more widely-spaced piles, there are two advantages: z

• Greater likelihood of “single pile” failure mechanism;

• Increased single pile capacity due to greater vertical stress
from raft contact pressure

fs = Ks (σv’ + σraft). tan(δ)

THE MECHANICS OF RAFTS + PILES
THE MECHANICS OF RAFTS + PILES
Settlement:

Depends on:
• Applied load
• Pile dimensions
• Ground stiffness and profile
• Group dimensions and number of piles
• Raft dimensions and raft stiffness

• Pile-soil-raft interactions (4 components)

 Pile-soil-pile
 Raft-soil-raft
 Raft-soil-pile
 Pile-soil-raft
DESIGN ISSUES

• Ultimate load capacity - vertical, lateral, moment

loading
• Maximum settlement
• Differential settlement
• Raft moments & shears
• Pile loads and moments
SETTLEMENT IS THE CRITICAL FACTOR FOR
LARGE FOUNDATIONS
 STAGES OF DESIGN
Preliminary design –

• number of piles for capacity & settlement requirements

• assess pile locations for local requirements

Detailed design - –
• refine pile locations

• optimize raft thickness

• obtain detailed settlement, moment and pile load information

PRELIMINARY DESIGN - ULTIMATE LOAD CAPACITY

• Lesser of:
–raft plus individual piles
–block of piles + soil, plus raft outside block perimeter

• Need to consider effects of soil layering on raft & pile

capacity

NOTE: Raft pressures may enhance pile capacity

 PRELIMINARY DESIGN - OVERALL LOAD -
SETTLEMENT ESTIMATION

• Estimate average settlement versus number of piles -

can use simple hand or computer methods. Allow for
cases where piles are fully utilized.

• Randolph's expressions give useful first estimate of

load sharing between raft & piles.

• Can thus obtain estimate of required number of piles

for specified ultimate capacity & settlement criteria.
SIMPLIFIED METHODS OF ANALYSIS

Poulos-Davis-Randolph (PDR) method - tri-linear

load-settlement curve

Requires estimates of:

• pile group stiffness
• raft stiffness
• ultimate pile group & raft capacities
• raft-pile interaction factor
PILED RAFT UNIT (Randolph, 1994)

rc
Young' s Modulus E s
E so E sav E sl E sb

Soil L

Bearing
d= 2ro Depth
stratum
Fig.2 Simplified representation of pile-raft unit

Stiffness of piled raft depends on:

- Raft stiffness
- Pile group stiffness
- Interaction between piles and raft
POULOS-DAVIS-RANDOLPH (PDR) METHOD
Construct a Tri-Linear Load-Settlement Curve

Pu
B

Load

P1 A

Pile + raft
elastic
Pile capacity fully utilised, Pile + raft ultimate
raft elastic capacity reached

Settlement
Fig.3 Simplified load- settlement curve for preliminary analysis
 POULOS-DAVIS-RANDOLPH (PDR) METHOD
• Piled raft stiffness:

Kpr = [Kp + Kr.(1-2αcp)]/(1-αcp2.Kr/Kp) Kp = pile group stiffness

Kr = raft stiffness
• Pile-raft interaction factor: αcp = pile-raft interaction factor
rc = raft area divided by number of piles
αcp = 1-ln(rc/r0)/ζ r0 = pile radius
ζ = ln(rm/r0)
rm = “magic radius” – see Randolph (1994)
• Proportion of load on Raft: Pup= ultimate capacity of piles
Pu = Pup + Pur
X = Kr(1- αcp)/[Kp+Kr.(1-2αcp )] Pur = ultimate capacity of raft

• Load at which pile capacity mobilized:

P1 = Pup/(1-X)
 TYPICAL RESULTS FROM PRELIMINARY
ANALYSIS

40 80
Ultimate Load Capacity (MN)

Total Load = 12 MN

Central Settlement (mm)

30 60

20 40

10 20

0 0
0 5 10 15 20 25 0 5 10 15 20 25
Number of Piles Number of Piles
(a) Ultimate Load Capacity (b) Settlement
Fig. 10 Effect of number of piles on ultimate load capacity and settlement

Increasing the number of piles beyond a certain point has no benefits

 ESTIMATING DIFFERENTIAL SETTLEMENTS

Can make use of approximations of Randolph (1994).

∆S/Sav = function of R = { n s / L }1/2

= f.R/4 for R <4

=f for R>4

f= 0.3 for corner to mid-side, and 0.5 for centre to

corner
DESIGN FOR LOCALIZED BEHAVIOR

When is a pile required under a column?

When:
• Maximum moment exceeds allowable

• Maximum shear exceeds allowable (punching shear)

• Maximum contact pressure exceeds allowable

bearing capacity of soil

• Local settlement exceeds allowable value

COLUMN LOAD ON RAFT
2c

x
t Raft E r , νr

Layer 1 E s1 , ν s

Layer 2 E s2 , ν s

Layer 3 E s3 , ν s

Figure 5 Individual Column Load-Raft on Layered Soil Profile

 SOIL PROFILE FOR LOCALIZED DESIGN

• May be different to that for overall load-settlement estimate

• Consider average parameters within effective depth = 3a

a = characteristic length of raft

= function of raft thickness & modulus, & soil modulus

a = t.{Er.(1-νs2)/6.Es.(1-νr2)]1/3
where t = raft thickness, ν = Poisson’s ratio of soil, ν = Poisson’s ratio of raft, E = raft modulus,
E = soil modulus. s r r
s
 MAXIMUM MOMENT CRITERION

Use solutions of Selvadurai (1979) to develop criterion.

Max. column load without pile is:

Pc1 = Md / Ax or Md / Bx
0.2

where Md = design moment

Ax , Bx = moment factors 0.1
B

Moment factors A, B
0
x

P
-0.1 O
A
Load
location

-0.2
0 0.05 0.10 0.15
x/a

Fig.6 Moment factors A & B for circular column

 OTHER COLUMN LOAD CRITERIA

The other critical column loads:

• Pc2 for shear
• Pc3 for local bearing pressure;
• Pc4 for local settlement

Can be obtained similarly from Selvadurai’s elastic

solutions. Refer to Poulos (2001).

Date A
27
presentation
to <insert in
 PILE REQUIREMENTS FOR A COLUMN LOCATION

A pile is required if the column load exceeds any of the

critical column loads, Pc1 to Pc4.
• For moment, shear or contact pressure requirement, the pile is
designed to provide deficiency in load capacity.

• For settlement, pile is designed to provide the required

additional stiffness.
CRITICAL COLUMN LOADS

Max. column load without a pile :

• increases as raft thickness increases
• increases as ultimate bearing pressure increases
• increases as soil modulus increases

Raft thickness is very important for

“local” design.
DETAILED DESIGN - REQUIREMENTS

• Analysis which can account for interaction between

piles & raft
• Ability to consider practical soil profiles
• Nonlinear soil - raft & soil - pile responses
• Non - uniform loadings
• Different pile types & sizes at various locations
below raft.
GASP & GARP are able to take the above factors into
account.
APPROXIMATE COMPUTER METHODS
Strip on Springs - GASP
• Strip sections of a raft analyzed
• Piles are modelled as non-linear interacting
springs

Plate on Springs - GARP

• Piles as for GASP
• Raft as elastic plate

ELPLA – Commercially available via GEOTEC SOFTWARE

GARP vs PDR METHOD
30 30
(a) 0.5 m Raft + 3 Piles (b) 0.5 m Raft + 9 Piles
Group Load (MN)

Group Load (MN)

20 20

10 10

0 0
0 20 40 60 80 100 120 0 20 40 60 80 100 120
Central Settlement (mm) Central Settlement (mm)
40
(c) 0.5 m Raft + 15 Piles
Group Load (MN)

30

GARP5
20 Approximate PDR method

10

0
0 20 40 60 80 100 120
Central Settlement (mm)
Fig. 9 Comparison between GARP and approximate analysis
MORE RIGOROUS COMPUTER METHODS

Two - Dimensional analyses:

• FLAC
• PLAXIS2D
• RS2

Full Three-Dimensional Analyses:

• FLAC 3D
• PLAXIS3D
• RS3
EXAMPLE PROBLEM APPLICATION
y

A A 1m
P1 P2 P1
2
A A
P1 P2 P1 x Bearing capacity of raft = 0.3 MPa
2
A A Load capacity of each pile = 0.873 MN (Compression)
P1 P2 P1 1 = 0.786 MN (Tension)

1m 2 2 2 2 1
E p = E r = 30000 MPa
ν p = νr = 0.2 P P2
1 P1
t r = 0.5 m

E = 20 MPa
l = 10 m
ν = 0.3
H = 20 m

d = 0.5 m
s=2 2 2 2m

Fig. 11 Hypothetical example used to compare results of various methods of piled raft analysis
COMPARISON BETWEEN PILED RAFT ANALYSES

20
2 3 4
1
5
15
Total load (MN)

10
1 P D R m e th o d
2 GARP5
5 3 GASP
4 F L A C 3-D
5 F L A C 2-D

0
0 20 40 60 80 10 0 12 0 14 0
C en tra l S e ttle m e n t (m m )

35
EFFECT OF IGNORING INTERACTIONS -
FOUNDATION STIFFNESS (MN/m)
2000
1800
1600
1400
1200
WITH INTERN.
1000
GROUP ONLY
800
NO INTERN.
600
400
200
0
9 PILES 15 PILES

Inclusion of interaction effects is

essential!

2D analysis is unreliable
SOME OTHER ISSUES
Lateral Loading:
Raft tends to take a larger proportion of the lateral load than for
vertical load. Full 3D analysis needs to be used at final design
stage.

Hamada et al (2015):

αp = proportion of load carried by piles

Kgp=n*Kp1
n = no. of piles
Kp1 = lateral stiffness of single pile
Kr = raft stiffness = 2πGr/(2-ν)
r = equivalent radius of raft
β = (Es/4.EI)0.25
SOME OTHER ISSUES

Effect of Basement Walls:

They take some of the lateral load and reduce the bending
moments in the pile. Can be taken into account during the final
design stage.

Disconnected Piles:

• Piles act as ground reinforcement. OK for vertical load;

• Not ideal for lateral load
GEOTECHNICAL PARAMETER ASSESSMENT

Require:

• raft bearing capacity

• pile capacity
• soil stiffness for raft
• soil stiffness for piles – vertical & lateral.

For preliminary design, can use correlations with

SPT or CPT for soils, or with UCS for rocks.

For final design, large scale pile load tests are highly
desirable
 CORRELATIONS WITH SPT

Decourt (1989, 1995):

Raft capacity: pu = K1 Nr kPa

Pile shaft : fs = a( 2.8Ns + 10) kPa

Pile base: fb = K2 Nb kPa

Soil modulus (raft): Es = 2Nr MPa

Soil modulus (pile): Es = 3N MPa

 CORRELATIONS WITH UCS
(Conservative)
Raft capacity: pu = 1.5(UCS)

Pile shaft : fs = 0.3(UCS)0.5 MPa

Pile base: fb = 4.8 (UCS)0.5 MPa

Rock modulus (raft): Es = 100(UCS)

Rock modulus (pile): Es = 150(UCS)

MEASURES TO REDUCE RISKS
• Independent peer review of ground investigation report

• Independent peer review of foundation design

• Make use of relevant precedent information

• Independent supervision of foundation construction

• Foundation element (pile/barrette) testing

• Monitor settlements during construction

 Compare measured versus predicted performance

Treat these measures as INSURANCE INVESTMENTS

 EXAMPLES OF PILED RAFT DESIGN

• Artique – Gold Coast Australia

• Burj Khalifa – Dubai, UAE

• Convention Centre – Doha, Qatar

• Diamond Tower, Jeddah, KSA

• Incheon Tower – Incheon, South Korea

• Examples of excessive piles & foundation economy – Italy, Germany

EACH OF THESE CASES HAS PROVIDED LESSONS.

GOLD COAST PROJECT - ARTIQUE
•A 28 storey building on the
Gold Coast, Australia

•Structural Engineer designed a

fully piled to rock foundation
system

•Piling contractor engaged

Coffey to assess feasibility of
piled raft

•Based on results of feasibility,

piling contractor engaged
Coffey to optimize piled raft
design
 ORIGINAL DESIGN

136 piles
founded
on rock

Slab
0.7m
thick

Contiguous bored pile wall Shear joint between core and podium
GEOTECHNICAL MODEL Es pu Es
Av. Su fs fb
DESCRIPTION (RAFT) (RAFT) (PILES)
SPT kPa kPa MPa
MPa MPa MPa
0
ASSUMED
BASE OF RAFT

-5
SAND
60 - 90 5.4 120 100 9.9
D-VD

-10

PEATY CLAY
(SOME SAND) 10 80 8 0.5 20 22 0.7
-15
F-St

SAND
60 - 90 5.4 120 100 9.9
D
RL (m)

-20

SANDY CLAY (H)

/CLAYEY SAND 14 250 25 1.5 40 60 2.0
-25 MD

-30 SAND WITH

SOME GRAVEL 25 - 37.5 2.25 50 48 4.1
MD

-35 SANDY GRAVEL 100 - 150 9.0 200 100 10.0

METASILTSTONE
- - 2000 - 2000 - 10.0
SW

-40
 FEASIBILITY RESULTS
• Indicated that a raft foundation alone would have a
factor of safety of approximately 10 for ultimate
loading

• Settlements would govern. Estimated to be of the

order of 35mm to 60mm

• The number of piles would be of the order of 140 –

as per the foundation design supplied by contractor.
However, piles only 18m long not 35m
Initial Assessment

• Goal was to assess serviceability criteria and opportunity

for design optimization

• Criteria set by structural engineer:

 Maximum total displacement of building column to be
less than 50mm

 Maximum differential displacement between adjacent

columns of 1/400
Method of Analysis

• Undertaken using GARP 8 program

• Assesses deflections, shear forces and moments in raft and pile

loads

• Final foundation system was a 0.8m thick raft with 136 piles.
CFA piles were 18m long, 0.7m diameter with a maximum
capacity of 4.2MN
Results of Initial Assessment
 136 piles
 Maximum raft
settlement of
36mm
 Maximum
differential
settlement of
10mm (1/400)
 OUTCOMES

• Number of piles reduced by 10% (13 piles)

• Pile length reduced from 35m to 18m

• Total pile length reduced by 2767m

(52% savings)

• Settlement criteria (both total and differential)

satisfied
BURJ DUBAI (KHALIFA)TOWER

Main challenges:

• World’s tallest building

• Foundation capacity for piles in

carbonate soils/rocks

• Concerns re cyclic loading of

piles

• High-rise to low-rise differential

settlements
SITE CHARACTERIZATION

• 30 boreholes

• SPT

• 60 PMT tests in 5 boreholes

• 6 standpipe piezometers

• Geophysics – cross-hole tomography

SIMPLIFIED PROFILE

4 Silty Sand

6 Calcarenite

17 Calcareous Sandstone Base of Tower Raft

4.5 Gypsiferous Sandstone

40 Conglomeritic Calcisiltite

Base of Tower Piles

22.5 Calcareous/Conglomeritic

>47 Claystone/Siltstone
INITIAL PILE DESIGN

• Tower:
196 piles, 1.5m diameter, 47.5m long

• Podium:
750 0.9m diameter piles, 30m long

• Raft: Tower pile layout

3.7m thick (tower)

Learnings from Emirates Towers design were used in the design.

Initial Tower Settlement Predictions
Analysis Settlement mm
(Flexible cap)

HYDER - REPUTE 66

HYDER - ABAQUS 72

COFFEY – FLAC 73
(axisymmetric)
COFFEY - PIGS 74

Subsequent Equivalent pier check (Coffey): 73 mm

LOAD TEST PROGRAM
• 3 static compression tests (1.5m dia.)
Various toe levels (35-55m long)

• 1 static compression test (0.9m dia.)

Shaft grouted

• 1 cyclic compression test (0.9m dia.)

• 1 static tension test (0.9m dia.)

• 1 lateral load test

• For all tests, results exceeded expectations

MEASURED SETTLEMENT CONTOURS –
FEBRUARY 2008
MEASURED TIME-SETTLEMENT – WING C
Predicted approx 75- Settlement in Wing C
27-Jun-06
80mm Distance along wing cross-section (m) 16-Jul-06
16-Aug-06
0 20 40 60 80
18-Sep-06
0
To February 2008 43mm -10
16-Oct-06
14-Nov-06
measured for 80% of dead -20
19-Dec-06

load
16-Jan-07

Settlement (mm)
-30 19-Feb-07
18-Mar-07
-40
12-Jul-07
Estimated 50 to 55mm -50 14-Aug-07
17-Sep-07
final settlement -60
17-Oct-07
-70 14-Nov-07
17-Dec-07
-80
Within design tolerances -90
18-Feb-08
design
CONSTRUCTION PROGRESS

Early 2006 Early 2007 Grand opening, 4 January 2010

EXAMPLE – TOWER IN DOHA

525 piles

d=1.0, 1.2, 1.5m

4 different lengths

Raft 4.0m thick, with local

thickening in core areas
EXAMPLE – TOWER IN DOHA:
Effect of Interaction Assumptions (GARP
Analyses)

Run No. Details

Q1 Normal analysis – all interactions

included

Q2 Zero pile-pile interactions, but raft-raft,

pile-raft and raft-pile interactions
included

Q3 Zero pile-pile, pile-raft and raft-pile

interactions; only raft-raft interaction
accounted for
EXAMPLE – TOWER IN DOHA:
Effect of Interaction Assumptions
90
Maximum settlement mm

80 0.002

Maximum Rotation in x-
70
60 0.0015

direction rad
50
40 0.001
30
20 0.0005
10
0 0
Q1 Q2 Q3 Q1 Q2 Q3
Case Case

Maximum Settlement Maximum Rotation

50
70

Maximum Pile Load MN

Maximum Moment Mx

60 40
50
40 30
MNm

30 20
20
10 10
0
0
Q1 Q2 Q3
Q1 Q2 Q3
Case
Case
Maximum Raft Moment Maximum Pile Load
IGNORING THE PRESENCE OF THE RAFT

Doha Tower Case:

Value Allowing for raft Ignoring the raft

Max settlement mm 81.8 174.6

Min settlement mm 7.6 7.2
Max x-rotation rad 0.00176 0.01260
Max x-moment MNm 66.3 70.3
Min x-moment MNm -46.2 -45.0
Max pile load MN 34.4 67.2
% load on raft 24.6 0
DIAMOND TOWER, JEDDAH, KSA
 KEY CHALLENGES

• Karstic limestone & random cavities

• Potential for building tilt

GEOTECHNICAL PROFILE
 FOUNDATION DETAILS FOR TOWER

• 5.5m thick raft

• 145 bored piles 1.5 m in diameter

• Pile length = 40m

• Total vertical load for serviceability = 2859 MN

PILE LAYOUT
RANDOMLY GENERATED CAVITIES
Young’s Modulus Values from Various Sources
 EFFECTS OF RANDOMLY LOCATED CAVITIES

• Effect of a single cavity was small extra 2-3mm

settlement (55-58mm settlement)

• Effect of multiple cavities: computed settlements

varied from 65 to 74mm, depending on location

• Non-symmetrical settlement pattern

EXAMPLE OF EFFECTS OF RANDOMLY LOCATED CAVITIES
EFFECTS ON RAFT MOMENTS

No Cavities With cavities

Largest increase is about 13%

 OBSERVATIONS FROM ANALYSES

• Cavities cause some increases in settlement

• A single cavity causes relatively small settlement

increase

• Settlement pattern can be non-symmetrical

• Effects are generally not as great as might be feared –

redundant foundation system
EXAMPLE – INCHEON TOWER, S. KOREA

172 bored piles

d=2.5m

Typical pile length = 50m

Raft 5.5m thick

KEY CHALLENGES

• Reclaimed land

• Complex geology

• 600m tall building

• Limited tolerance to differential settlements

SITE CONDITIONS
GEOTECHNICAL CONDITIONS

8 separate soil
profiles modeled
within building
footprint

Contours of depth to bedrock

TYPICAL GEOTECHNICAL MODEL
Strata Ev Eh (MPa) fs fb
(MPa) (kPa) (MPa)

UMD 5 -11 29 - 48 -
7 - 15
LMD 30 21 50 -
Weathered Soil 60 42 75 -
Weathered Rock 200 140 500 -
Soft Rock (above EL-50m) 300 210 750 12

Soft Rock (below EL-

50m) 1700 1190 750 12

Ev = Vertical Modulus
Eh = Horizontal Modulus
fs = Ultimate shaft friction
fb = Ultimate end bearing
 FOUNDATION SYSTEM
• Piles: 172

• Pile size : 2.5m dia.

• Founded minimum 2 pile diameters into soft rock or

below EL-50m

• Mat Thickness : 5.5m

 COMPARISON OF COMPUTED
SETTLEMENTS

• Preliminary (equivalent pier): 75mm (av.)

• Detailed Design (GARP): 67mm (max.)

• Final Design Check (PLAXIS 3D): 56mm (max.)

EXAMPLE – INCHEON TOWER, S. KOREA

Quantity Location Rigid Raft Flexible Raft

Pile Load (MN) Centre Pile 24 49

Centre Edge Pile 65 33

Corner Pile 85 43

Pile Stiffness (MN/m) Centre Pile 511 726

Centre Edge Pile 1418 932

Corner Pile 1604 1292

Raft Settlement (mm) Maximum 52 67

Minimum 26 28
FINAL DESIGN CHECK – 3D FE
ANALYSIS
FINAL DESIGN CHECK – 3D FE ANALYSIS
Visualization

No basement contact With basement contact

 O-CELL TESTING

Date A presentation to <insert in footer> 86

PILE LOAD TEST RESULTS

• Design parameters are

conservative

• Scope exists for foundation

economy (if project resumes)
EXAMPLE OF EXCESSIVE PILES
(Mandolini et al, 2005). Twin towers in Naples

Total 637 piles

(L=20m, d=0.6m)

Designed with no
allowance for the
raft
EXAMPLE OF EXCESSIVE PILES
(Mandolini et al, 2005)
Computed versus measured settlements-excellent agreement
EXAMPLE OF EXCESSIVE PILES
(Mandolini et al, 2005)

Foundation system now analyzed with only 318 piles

Allowance made for raft
EXAMPLE OF EXCESSIVE PILES
(Mandolini et al, 2005)
• Computed settlement with 318 piles = 33mm

• Compared with 30mm for 637 piles!!

• Also, maximum differential settlement reduced by 10%

• Total piling length actually used = 12700m

• Total length that should have been used = 6300m

• So, would have very similar foundation behaviour with

50% savings on piling.

• Therefore, greatly enhanced sustainability.

EXAMPLE OF COST SAVINGS ON PILES
(Katzenbach, 2011)

Fully piled foundation:

•316 piles, 30m long: US$7.4M

Piled raft foundation:

•64 piles, av. 30m long: US$1.5M

•Savings: US$5.9M
EXAMPLE OF COST SAVINGS ON PILES
(Katzenbach et al (2016)

Mirax Tower A, Kiev, Ukraine

Fully piled foundation:

•120 barrettes, 40m long: US$5.8M

Piled raft foundation:

•64 piles, av. 33m long: US$2.5M

•Savings: US$3.3M
SOME GENERAL OBSERVATIONS ON PILED RAFTS
• Settlement decreases as number of piles increases

• Major reduction in settlement can occur for relatively small

number of piles

• Increasing number of piles beyond a certain point is ineffective

• Piles may reduce bending moments in raft if carefully located

• Raft thickness affects differential settlements much more than

total settlements

• Foundation performance may be improved by using piles of

varying length - shorter piles below lighter loads, longer for
heavy loads
CONCLUSIONS
• Piled rafts can provide considerable foundation
economy

• Also increased sustainability

• Considerable experience of successful application

• Combined raft-pile interaction can increase the pile

capacity – “get something for nothing”
BIBLIOGRAPHY
Katzenbach, R.; Leppla, S.; Choudhury, D. (2016). Foundation systems for high-rise structures. CRC
Press Taylor & Francis Group, New York, USA .

Mandolini, A., Russo, G. and Viggiani, C.(2005). “Pile foundations: experimental investigations,
analysis and design”. Proc. 16th Int. Conf. Soil Mechs. Geot. Eng., Osaka, IOS Press, 177-213.

Poulos, H.G. (2001). “Piled Raft Foundations – Design and Applications”. Geotechnique, 51(2): 95-
113.

Poulos, H.G. (2017). “Tall Building Foundation Design”. CRC Press, Boca Raton, Florida.

Poulos, H.G. (2021). “Practical approach for piled raft stiffness estimation”. Australian
Geomechanics, 56 (3): 57-69.

Randolph, M.F. (1994). “Design methods for pile groups and piled rafts”. State of the Art Rep.,
Proc., 13th ICSMFE, New Delhi, Vol. 5, 61–82.

Selvadurai, P. (1979). “Elastic analysis of soil-foundation interaction”. In Developments in

Geotechnical Engineering, Vol. 17, Elsevier, Amsterdam.