Japanese Geotechnical Society Special Publication

The 15th Asian Regional Conference on

Soil Mechanics and Geotechnical Engineering

Design and performance of the piled raft foundation for Shanghai World Financial Center
W. D. Wang i, ii), J. B. Wu i, ii) and Q. Li i, ii)
i) Department of Underground Structure and Geotechnical Engineering, East China Architectural Design
and Research Institute Co., Ltd., Shanghai 200002, Peoples Republic of China
ii) Shanghai Engineering Research Center of Safety Control for Facilities Adjacent to Deep Excavations,
Shanghai 200002, Peoples Republic of China

ABSTRACT
A case history of the piled raft foundation for Shanghai World Financial Center which is a 101-storey, 492 m high
building is presented in this paper. To monitor the performance of the piled raft foundation system and to verify the
foundation design work, field measurements were carried out on the raft settlements, pile head axial loads, contact
pressures of the raft and pore-water pressures underneath the raft during the entire construction period. The measured
raft settlements were comparable to the computed results. At six months after topping out, the maximum settlement
was 130 mm in the tube area, decreasing to 90 mm at the edges. Axial loads at pile head increased gradually and
varied in magnitude from 1000 to 5000 kN, less than the proposed compression capacities of the steel pipe piles.
Both contact pressures of the raft and pore-water pressures beneath the raft varied considerably during the initial
construction period but changed slightly after the end of construction. Based on the field measurement results, the
piled raft foundation design proved to be appropriate.
Keywords: foundations, piled raft, design, settlement, field monitoring

solution. To verify the design assumptions, extensive

instruments were installed on site to monitor the raft
settlements, pile head axial loads, contact pressures of
the raft and pore-water pressures underneath the raft
during the entire construction period. Comparisons are
made between the observed raft settlements and
computed values. Finally the performance of the piled
raft is investigated based on the field monitoring
results.

1. INTRODUCTION
Shanghai is one of the fastest growing cities in
terms of skyscraper construction in the world. In recent
decades, an increasing number of high-rise buildings
have been constructed or are under construction. Most
of these tall buildings are founded on piled rafts. Since
Burland et al. (1977) presented the concept of
settlement reducers, there has been an increasing
recognition that the use of piles to reduce raft
settlements and differential settlements can lead to
considerable economy without comprising the safety
and performance of the foundation (Poulos, 2001). As
the raft settlement is crucial to a piled raft foundation
system, it is demanded to estimate the settlement
distributions of a piled raft at the design stage by
considering the complex interaction among the soil, raft
and piles. In addition, field monitoring is required to
verify the validity of the foundation design and to
improve the understanding of the mechanics of
behavior of piled rafts (O'Neill et al., 1996, Mandolini
et al., 2005).
A case history of design and performance of the
piled raft foundation for Shanghai World Financial
Center (SWFC) is presented in this paper. The
settlement behavior of the piled raft is analyzed by
using a computer program PWMI which can take the
pile-soil-raft interactions into account based on Geddes

http://doi.org/10.3208/jgssp.CHN-44

Shanghai World Finacial Center

Jin Mao Tower

Fig. 1 Photo of Shanghai World Financial Center

162

Shanghai World Financial Center is a mixed-use

skyscraper located in the heart of Pudong Lujiazui
financial district on the eastern bank of Huangpu River
in Shanghai. The building consists of a 101-storey, 492
m high main tower above the ground surface and a
three-storey basement below ground. As shown in Fig.
1, the towers basic form is that of a square prism, 58 m
on a side, intersected by two sweeping arcs to form a
six-sided shape, and finally tapering into a single
diagonal line at the apex (Katz and Robertson, 2008).
To resist lateral loads from wind and earthquake, a
unique structural system has been developed for this
building. The system comprises three main components,
i.e., the mega-frame structure consisting of
mega-columns, diagonals and belt trusses, the concrete
shear walls and the outrigger trusses connecting the
mega-columns and the concrete shear walls.

Soil
strata

h
(m)

(kN/m3)

(kPa)

()

1
2
3
1
2
3

1.3
4.0
10.7
6.0
5.0
13.0
23.0
6.4
4.4
17.0
56.0

18.4
17.7
16.7
18.4
19.8
18.7
19.2
19.1
20.2
20.2
20.2

20
10
14
15
45
3
0
2
/
/
/

18.0
22.5
11.5
22.0
17.0
32.5
33.5
34.0
/
/
/

GEOLOGICAL CONDITIONS

Shanghai is situated on the Yangtze River delta

where the geology is generally composed of alluvial
and marine sediments deposited during the Quaternary
period. The ground conditions at this site comprise
variable horizontally stratified clayey and sand strata
due to alternating climates and sea level changes. The
upper clay layers are underlain by successions of
various aquifers inter-bedded with silt and clay layers.
Field exploration and laboratory tests show that the
sub-soils down to a depth of about 28.7 m can be
divided into fill (0-1.7 m), silty clay (1.7-3.0
m), soft silty clay (3.0-7.0 m), soft clay
(7.0-17.7 m), firm silty clay (17.7-23.7 m),
stiff silty clay (23.7-28.7 m). Between depths of
28.7-71.1 m, there lies fine sand and silt (layer ). A
fine-coarse sand layer (layer ) is located below a
depth of 71.1 m. Layers and are artesian and
referred to as Aquifer I and II, respectively. These two
sandy aquifers are inter-connected as a silty clay layer
(layer ) is not found between them at this site.
Groundwater levels are generally high across the site
and located about 0.5m below the ground surface. A
typical ground profile and geotechnical parameters
derived from the investigation data and experience of
similar soils on adjacent sites are summarized in Table
1.

N63.5
2.5
2.0
1.0
5.5
19
33
60
46
52
54
79

fs

fp

(kPa)

(kPa)

15
20
20
45
60
60
70
70
70
70
70

/
/
/
/
/
/
2500
2200
2500
2500
2500

Ground surface
1 Fill
2 Silty clay
3 Soft silty clay

10

4.0 m

4.5 m

20

4.0 m

4 Soft clay

5 Silty clay
6 Silty clay

30

60.7 m

40

7 1 Sandy silt
with beds of
silty fine sand

41.7 m

Table 1 Summary of soil profile and geotechnical parameters

h-thickness; -unit weight; c-cohesion; -friction angle;
N63.5-number of blows to achieve a penetration depth of 10 cm
by a heavy hammer of 63.5 kg; fs-skin friction; fp-tip resistance. fs
and fp values obtained from static cone penetration tests

14.5 m

2 DESCRIPTIONS OF SHANGHAI WORLD

FINANCIAL CENTER

50
7 2 Fine sand
60

70

7 3 Sandy silt
9 1 Silty sand

80
9 2 Coarse sand
with gravels
90

70 m

100

110

120

4 FOUNDATION DESIGN
4.1 Foundation layout
Fig. 2 shows the schematic diagram of the
foundation for the main tower of SWFC which
occupies a total square footprint of 70 m by 70m. The
foundation consists of a piled raft with 985 steel pipe
piles, each 0.7 m in diameter. The raft is embedded at
14.5 m below the ground and constructed using C40

130

9 3 Silty fine sand

225 piles 60.7m long

760 piles 41.7m long
70 m

Fig. 2 Piled raft foundation for the main tower of SWFC

concrete (28-day compressive strength is 19.1 MPa).

The central part of the raft is 4.5 m thick, decreasing to
4.0 m at the edges. There are 225 piles underneath the

163

around 1/850.

central portion of the raft. Each pile is 60.7 m long and

has a wall thickness of 18 mm. These piles are
embedded in coarse sand (layer 2) and have a
compression capacity of 5750 kN. The rest pipe piles
whose wall thickness is 15 mm are driven into the fine
sand layer at a depth of 41.7 m beneath the raft and
have a compression capacity of 4250 kN.

5.1 Instrumentation
A
16

Distance (m)

10m
7m

26

12

10m

7m

25
11

18m

10

20

21

Settlement monitoring points

22
A

23

24

Fig. 4 Layout of settlement monitoring points

The raft settlements were measured by precise

levelling of settlement markers which were made of
stainless steel rods, 10 cm long and 16 mm in diameter.
They were set up in the raft before the concrete
hardened and had a slightly protruding rounded head
that was always findable by a surveyor. A total of 29
settlement monitoring points were placed over the
entire raft as shown in Fig. 4. Three strain gauges were
installed at the head of each instrumented pile to
investigate the axial load distribution. Earth pressure
cells and piezometers were installed beneath the raft
near the instrumented piles to register earth pressures
and pore water pressures, respectively. The field
monitoring extended from the completion of raft
construction to about six months after topping out. Due
to paper length limit, monitoring points for pile-head
axial loads, contact pressures and pore pressures are not
shown in Fig. 4.

20

10

60

27

Concrete
tube
1

30

50

28

13

19

40

40

18

50

30

29

17

60

20

14

70

10

15

18m

4.2 Settlement analysis

In order to estimate the raft settlements, a
computer program PWMI developed by Wang et al.
(2007) has been used as a preliminary design tool. In
PWMI, the raft is modeled as a thick plate, the soil-pile
interactions are analyzed based on Geddes solution and
the walls are treated as beam elements. At each raft
node, the equivalent piled raft stiffness is formed by
adding the soil-pile stiffness, the raft stiffness and the
wall stiffness together. Based on the equivalent raft
stiffness, the settlements and stresses of the raft
subjected to vertical loads can be determined once the
displacement compatibility condition is satisfied.
In the deformation analysis, the vertical loads
transferred to the raft from the columns are modeled as
concentrated loads, while the loads from shear walls are
considered as knife edge loads. All the loads from the
superstructures are directly applied on the raft nodes.
The numerical model of the raft consists of 30591
quadrilateral and 1010 triangular elements. Soil
parameters used in the analyses are given in Table.1.

FIELD MONITORING

70

5.2 Field monitoring results

Distance (m)

Raft settlements

Fig. 3 Contours of computed raft settlements (unit: mm)

For clarity, only the raft settlements along

cross-section A-A (see Fig. 4) is presented in this paper
and shown in Fig. 5. The initial values of settlement
were taken right after the completion of casting the
foundation slab. The settlement curves for the
monitoring points in the tube area are almost
overlapped, suggesting the raft settled uniformly in the
area. The raft settlement increased gradually to about
40 mm about 20 months after the monitoring began.

Fig. 3 shows the contours of the computed raft

settlements for the area enclosed by the dashed line as
illustrated in Fig. 2 under both static and live loads. The
raft for the main tower shows a basin-shaped settlement
profile. The maximum settlement is about 113 mm in
the center, decreasing gradually to about 72 mm at the
edges. The largest computed differential settlement of
the raft is 41 mm, giving rise to an angular rotation of

164

Thereafter, the raft settlement increased significantly

with construction progress. When the building was
topped out in December 2007, the raft settlement
reached about 100 mm in the tube area. Additional 30
mm settlement took place in a period of six months
after topping out. Raft settlements at edges show a
similar trend to those in the tube but with smaller
magnitudes. The raft settlements seemed to reach a
steady value at the end of the monitoring period.
5F

25F

63F

86F

The pile-head axial loads kept increasing during

the construction period. At the end of construction, the
pile-head axial loads for piles beneath the raft varied
from 1000 to 5000 kN, less than the proposed
compression capacities. The measured contact
pressures between raft and soil and pore water
pressures beneath the raft agreed fairly well with each.
Both of them increased significantly in the initial
construction period (before the fifth floor), varied
slightly thereafter and tended to decrease after the end
of construction. The contact pressures varied
considerably from 80 to 140 kPa, while the pore
pressures were in the range of 100-190 kPa.

Topping out

Settlment (mm)

20
40

60

The design of the piled-raft foundation for

Shanghai World Financial Center, 101-storey and 492
m high, is presented in this paper. The raft settlement
was estimated by using a computer program PWMI. To
monitor the performance of the piled raft foundation
system and to confirm the foundation design work, field
measurements were carried out on the raft settlements,
pile head axial loads, contact pressures of the raft and
pore-water pressures underneath the raft during the
entire construction period. The measured raft
settlements were comparable to the computed results.
At six moths after topping out, the maximum settlement
was 130 mm in the tube area, decreasing to 90 mm at
the edges. The pile-head axial loads increased gradually
and varied in magnitude from 1000 to 5000 kN, less
than proposed compression capacities. Contact
pressures of the raft and pore-water pressures beneath
the raft agreed fairly well with each other. Both of them
varied considerably in the initial construction period but
changed slightly after the end of construction. Based on
the field measurement results, the piled raft foundation
design for the SWFC proved to be appropriate.

80

S 14
S6
S2
S1
S4
S 10
S 22

100
120
140
0

200

400

600

800

1000

1200

Time (days)

Fig. 5 Development of raft settlements with time

0
20

Settlement (mm)

40
60
Time (Floor)
98 (B1)
268 (11F)
613 (63F)
698 (79F)
982 (99F)
1044
(topping out)
1135
1178
PWMI

80
100
120
140

S 14

S6

S2
S1
S4
Monitoring point

S 10

S 22

CONCLUSIONS

REFERENCES
(1) Burland, J. B., Broms, B. B. and De Mello, V. F. B. (1977).
Behaviour of foundations and structures. Proceedings of 9th
international conference on SMFE, Tokyo, Japan, 2,
496-546.
(2) Katz, P. and Robertson, L. E. (2008). Case history: Shanhgai
World Financial Center. CTBUH Journal, 2, 10-14.
(3) Mandolini, A., Russo, G. and Viggiani, C. (2005). Pile
foundations: experimental investigations, analysis and design.
Proceeding of 16th International Conference on Soil
Mechanics and Geotechnical Engineering (ICSMGE) 1,
177-213.
(4) O'Neill, M. W., Caputo, V., De Cock, F., Hartikainen, J. and
Mets, M. (1996). Case histories of pile-supported rafts.
Report for ISSMFE Technical Committee TC 18. Houston,
TX: University of Houston,
(5) Poulos, H. (2001). Piled raft foundation: design and
application. Geotechnique, 51(2), 95-113.
(6) Wang, W. D., Shen, Z. W. and Wu, J. B. (2007). Introduction
and application of the utility calculation and analysis method
considering the combined action of the soil, pile, raft and
superstructure. Building structures, 37(5), 111-113.

Fig. 6 Comparison between measured and computed raft

settlements

The measured raft settlement profile along

cross-section A-A is shown in Fig. 6. It is clear that the
observed settlements in the tube area were larger than
those at edges. At the time when the building was
topped out, the maximum settlement at the north side
(point 14) was 75 mm that was larger than the value of
64 mm at the south side (point 22), suggesting the raft
was tilted. The largest differential settlement was found
to be 36 mm, giving rise to an angular rotation of 1/500.
The computed settlements are also included in this
figure for comparison. The maximum raft settlement
computed by using PWMI is comparable to the
measured results.
Pile-head axial loads, contact pressures and pore water
pressures

165