KSCE Journal of Civil Engineering (2012) 16(7):1170-1177

DOI 10.1007/s12205-012-1700-8

Geotechnical Engineering

www.springer.com/12205

Equivalent Head-down Load vs. Movement Relationships Evaluated from

Bi-directional Pile Load Tests
Sung-Ryul Kim* and Sung-Gyo Chung**
Received September 14, 2011/Accepted February 19, 2012

Abstract
The increasing use of bi-directional (Osterberg cell) load tests in piles necessitates a reliable evaluation of the load-movement
relationship equivalent to the head-down load test. In this study, the existing evaluation methods were reviewed, and a new method
for evaluating the equivalent head-down curve was proposed. This method considers possible situations, such as non-measured axial
load distribution along the pile and the existence of residual load distribution. Three cases, in which bi-directional and head-down
load tests were performed at the same site, were analyzed in order to validate the proposed method. The results of the case studies
showed that the equivalent load-movement curves obtained either without assuming unit shaft resistance distribution or by
considering the effect of residual load agreed well with the measured curves obtained from the head-down test. Therefore, the
proposed method is recommended for practical use.
Keywords: deep foundation, foundation engineering, load test, pile, residual stress

1. Introduction
With the increase in the design load of piled foundations in
recent years, a common head-down load test is sometimes
insufficient to evaluate the bearing capacity of piles. Thus, the bidirectional (Osterberg cell) load test has been increasingly
applied in practice. In this test, a loading device, also known as
the O-cell, is installed at the pile toe. The toe resistance
provides the reaction force to mobilize the shaft resistance and
vice versa. The toe and shaft resistances can then be evaluated
separately, even without strain gauge measurements. The O-cell
test separates the toe and shaft resistances mobilizing them in
opposite directions, whereas the head-down test combines them
in the same direction. Piles for axial load are designed to function
similarly to a head-down test, and the design load is determined
after combining shaft and toe resistances, thus, a method to construct an equivalent head-down load-movement curve is needed.
Osterberg (1998) proposed a method of constructing the equivalent curve, based on an assumption of the pile being a rigid
body. An improved method was proposed by Loadtest (2001),
Kwon et al. (2005) and Lee and Park (2008), who all considered
a pile elastic compression. Unlike such simple approaches, Kim
and Mission (2011) and Xi et al. (2010) performed analysis using
load transfer function for shaft and toe resistances. However,
because their methods were developed based on an assumed unit
shaft resistance, or ignoring the effect of residual load, the appli-

cations may be limited in practice.

The purpose of this study, therefore, is to propose a method to
enable a more practical construction of the equivalent curve. To
achieve this, the existing methods are reviewed, and a new
method is proposed to construct the reliable equivalent curve for
piles under the following conditions: (1) non-measured axial
load distribution along the pile, and (2) the existence of residual
load. In addition, three case studies are analyzed to validate the
proposed method, including two cases involving long drilled
shafts and one case, in which a long concrete pile was driven into
a thick deltaic deposit. The equivalent curves obtained from the
proposed method are compared with those of existing methods
and with the measured load-movement curve from a head-down
test.

2. Existing Methods of Constructing Equivalent

Head-down Load-movement Curve
The following simple methods for constructing the equivalent
curves were reviewed: Osterberg (1998), Loadtest (2001), Kwon
et al. (2005), and Lee and Park (2008).
During the O-cell test, three kinds of load-movement curves
are generally obtained as shown in Fig. 1; (1) (toe resistance, Qd)
vs. (downward pile toe movement, D1), (2) (shaft resistance, Qu)
vs. (upward pile toe movement, D2), and (3) (shaft resistance, Qu)
vs. (upward pile head movement, D3).

*Member, Associate Professor, Dept. of Civil Engineering, Dong-A University, Busan 604-714, Korea (Corresponding Author, E-mail: sungryul@dau.
ac.kr)
**Member, Professor, Dept. of Civil Engineering, Dong-A University, Busan 604-714, Korea (E-mail: [email protected])
1170

Equivalent Head-down Load vs. Movement Relationships Evaluated from Bi-directional Pile Load Tests

Loadtest (2001) considered the additional pile compression

induced by the head-down load. Because D2 in the Original
method includes the pile compression by the O-cell load at toe,
toe, the additional pile compression by the different loading
direction between the O-cell and head-down tests should be
expressed as:
= d + head toe
where,

Fig. 1. Load-movement Curves Obtained from the O-cell Tests

Osterberg (1998) proposed a construction method for the

equivalent curve based on the assumption that the pile is a rigid
body. This method is denoted as Original method in this paper.
If the pile is assumed to be rigid, the top and bottom have the
same deflection but different loads. An arbitrary movement is
selected, after which the toe and shaft resistances are determined
as the loads corresponding to the arbitrary movement at (Qd, D1)
and (Qu, D2) curves, respectively. (The sum of the shaft and the
toe resistances) vs. (the arbitrary movement) becomes a single
point at the equivalent load-movement curve. The equivalent head
down curve is obtained by repeating this process at different
movement points. This method, however, does not satisfy the
real condition because the toe movement D2 includes the pile
compression by the O-cell load.
In general, the head-down load at the pile head induces larger
pile compression compared with the equivalent upward load at
the pile toe. For example, Fig. 2 shows the load distributions obtained from the head-down and O-cell tests. The pile compression is proportional to the area of the load distribution curves,
that is, area ABCE for the head-down test and area FGH for the
O-cell test. Therefore, at the head-down test, the pile compression consists of d and head induced by Qd and Qu, respectively
(Fig. 2a), whereas it consists only of toe induced by the Qu at the
O-cell test (Fig. 2b).

d: Pile compression induced by equivalent head-down

load of Qd
head: Pile compression induced by equivalent head-down
load of Qu
toe: Pile compression induced by upward toe load of Qu

The equivalent curve was obtained by adding to the corresponding movements in the curve obtained from Original method.
In this method, the magnitudes of head, toe, and d were calculated using Eqs. (2), (3), and (4), respectively. Various shapes
of the unit shaft resistance can be taken into account with the
centroid factor C.
QuL
head = ( 1 C ) -------EA

(2)

Qu L
toe = C --------EA

(3)

Qd L
d = --------EA

(4)

where, C: Centroid factor = (the distance to the centroid from the

bottom of the unit shaft resistance)/(overall length
of unit shaft resistance): C=1/3 for a linear distribution of unit shaft resistance; C=1/2 for a constant
distribution
E: Youngs modulus of pile
A: Sectional area of pile
L: Pile penetration length
Kwon et al. (2005) and Lee and Park (2008) modified the
Original method to obtain the equivalent curve of a true rigid
pile. They evaluated the equivalent head-down load Qu using
(Qu, D3) curve instead of the (Qu, D2) curve in Fig. 1 because D3
does not include the pile compression. The pile compression of
(head+d) was then added to the corresponding movements in
the curve of the true rigid pile. The head was calculated using
either Eq. (5) based on measured pile compressions (Kwon et al.,
2005) or Eq. (2) based on theoretical pile compressions (Lee and
Park, 2008). Eqs. (2) and (5) use an assumed unit shaft resistance
distribution, theoretically producing the same results.
head = ( D2 D3 )

Fig. 2. Load Distributions and Pile Compressions Developed from

the Head-down and O-cell Tests: (a) Head-down Test, (b)
O-cell Test
Vol. 16, No. 7 / November 2012

(1)

(5)

where, =1: For a constant unit shaft resistance with depth

=2: For a linear increase of unit shaft resistance with
depth

1171

Sung-Ryul Kim and Sung-Gyo Chung

3. Development of a Construction Method

The previous construction methods were only based on an assumed distribution of unit shaft resistance, or possibly, on a measured distribution using strain gauge instrumentation.
By improving the existing methods, a new construction method
is proposed, which determines a pile compression without assuming the unit shaft resistance distribution. In addition, the residual
load, which may exist in piles because of the rearrangement or
consolidation in the surrounding soil prior to the pile load test, is
considered in the proposed method for the evaluation of the pile
compression. Basically, the proposed method constructs the equivalent curve by applying the same procedure in Kwons method
(2005). However, the proposed method accurately evaluates head
in two possible cases: (1) when residual load can be ignored and
(2) when residual load is significant. The practical evaluations of
these two cases are presented below.
3.1 When Residual Load is Non-significant
3.1.1 When Axial Load Distribution is Unknown
Although an axial load distribution along a pile is unknown,
the pile compression head can be evaluated using the measured
pile compression during O-cell tests.
Theoretically, the sum of toe and head should be (QuL)/(EA)
under the following conditions: (1) the residual load can be ignored
along the pile, (2) the side shear is identical regardless of the
loading direction (i.e., bottom-up or head-down loads), and (3)
the EA is constant along the pile length from pile top to bottom.
Because toe is equal to the measured pile compression at an
O-cell test of (D2 D3), head is thus evaluated using Eq. (6). If
the measured pile compression is accurate, Eq. (6) theoretically
gives an accurate head value without the measurement of pile
compression.
head

Qu L
QuL
= -------- toe = -------- ( D2 D3 )
EA
EA

(Gregersen et al., 1973; Altae et al., 1993), and for bored piles in
sand (Baker et al., 1993). Likewise, Fellenius (2004) argued that
the residual load can be mobilized irrespective of the types of
piles and soils. Therefore, if the residual load is not considered,
the estimated resistance distribution may be different from the
actual (true) distribution. Although this error does not affect the
measured pile capacity, it may affect the designers interpretation
and subsequent use of the shaft resistance distribution.
The effect of the residual load on pile compression is shown in
Fig. 4. It is assumed that the residual load indicated by area EFG
in Fig. 4(a) exists in the pile prior to a load test. The area EFH in
Fig. 4(b) indicates the test resistance, which is mobilized by the
O-cell load, assuming zero pile strain. The true shaft resistance is
the sum of the residual load and the test resistance (Fig. 4c).
Because the pile compression that results from the residual
load has an identical effect on both the O-cell and the head-down
load test, head decreases and is then calculated using Eq. (8).
Qu L area EHI area EFG
-------------------------------------------------head = --------
EA
area EFHI

(6)

(8)

where the notations E, F, G, H, and I are shown in Fig. 4(c).

3.1.2 When Axial Load Distribution is Known

If the axial load distribution is measured during a load test, the
pile compression head can be calculated directly from the load
distribution curve shown in Fig. 3 using Eq. (7). Calculating loaddistributed areas based on the maximum applied load at the final
loading step is convenient. This procedure should provide a more
accurate estimate of the pile compression based on a measured
load distribution.
Qu L area ACD
---------------------------head = --------EA area ABCD

Fig. 3. Combined Shaft Load Distribution of the O-cell and Headdown Tests

If the false load distribution in Fig. 4(b) is used without the

residual load, head is proportional to the area EHI in Fig. 4(b)

(7)

where the notations A, B, C and D are shown in Fig. 3.

3.2 When Residual Load is Significant
In general, large residual load develops due to the negative
skin friction in soft ground. Several researchers have observed
that the residual load may also develop for driven piles in sand

Fig. 4. True Resistance by the Inclusion of the Residual Loads: (a)

Residual Load, (b) False Resistance, (c) True Resistance

1172

KSCE Journal of Civil Engineering

Equivalent Head-down Load vs. Movement Relationships Evaluated from Bi-directional Pile Load Tests

rather than to the [(area EHI) - (area EFG)] in Fig. 4(c). This
may lead to a significant overestimation of the pile compression.

4. Case Studies
Three cases were analyzed, namely, two drilled shafts and a
driven concrete pile to verify the proposed construction method
of an equivalent curve. The third case involved the monitoring of
the residual load. In these cases, the equivalent curve using the
proposed method was compared with the measured curve
obtained from the head-down load tests, which were performed
at the same site.
4.1 Case 1: Drilled Shaft in Busan, Korea
The O-cell and head-down load tests were performed separately for two drilled shafts on the same site as shown in Fig. 5.
The pile diameter was 1.5 m, the penetration depth was 33.5 m,
and the pile toe was embedded about 5 m into the weathered
rock. The axial load distribution was measured using strain
gauge pairs at 12 levels. The load test was performed one month
after the concrete placement, the period of which would be
insufficient to mobilize the residual load. The detailed results are
described in the paper of Kwon et al. (2005).
Fig. 6 shows the load movement curve of the O-cell test. The
maximum applied load was about 19 MN, which induced a large
downward toe movement of 214 mm.
Pile compressions have a large influence on an equivalent
curve. Thus, the pile compression toe was evaluated using three
methods, after which the results were compared in Fig. 7. The
methods were as follows: (1) the conversion of strain into pile
compression using 12 levels of strain gauge measurements; (2)
LVDT measurement using the telltale reading; and, (3) theoretical
calculation assuming linear unit shaft resistance distribution. The
Strain gauge measurement curve increased proportionally with
the increase in applied load, whereas the LVDT measurement
curve increased rapidly about 3.6 mm at about 16 MN. This abrupt

Fig. 5. Schematic Drawing of the Test Piles and Definition of

Movement and Pile Compression
Vol. 16, No. 7 / November 2012

Fig. 6. Load-movement Curve in the O-cell Test (The buoyant

weight of pile was 1.36 MN)

Fig. 7. Comparison of Pile Compressions, toe

increase in toe movement may have occurred due to measurement error, because pile damages were not detected from strain
gauge results. Therefore, the strain gauge measurement gave a
more reliable estimation of pile compression. In addition, the pile
compression of strain gauge measurement became identical with
that of the theoretical calculation at the final loading step. Thus,
the assumption of the linear increase of unit shaft resistance is
reasonable.
The equivalent load-movement curves obtained using the proposed method were compared with that of the head-down load
test (Fig. 8).
The curve, denoted as Unknown distribution, was obtained
assuming that the axial load distribution was unknown, and by
inputting the measured movements of D2 and D3 into Eq. (6) in
the evaluation of the pile compression head. The curve, denoted
as Known distribution, was obtained by applying the area ratio
of (area ACD)/(area ABCD) as 2/3 in Eq. (7), based on the linear
increase of unit shaft resistance confirmed in Fig. 7. Given with
C=1/3 in Eq. (2), the predicted curve from the Loadtest method
(2001) becomes identical with the Known distribution curve of
the proposed method.
In Fig. 8, the movements of the proposed method are a little

1173

Sung-Ryul Kim and Sung-Gyo Chung

Fig. 8. Comparison of Equivalent Load-movement Curves

larger than that of the measured curve. The larger movement of

the equivalent curve can be attributed to the large toe downward
movement D1 at the O-cell test because of the poor filling of
concrete near the pile toe as reported by Kwon et al. (2005).
The comparison showed that the equivalent curve by the proposed method predicted a similar trend with the measured curve,
and the method can be applied successfully to the case with an
unknown axial load distribution.
4.2 Case 2: Multiple O-cell Test for a Drilled Shaft in the
Singapore MRT Project
The head-down and the O-cell pile load tests were performed
separately for two drilled shafts at Singapore MRT (Mass Rapid
Transit) project. The soil layer, as shown in Fig. 9, consisted of a
silty clay layer, a clayey silt layer, and a very stiff silty clay layer
below 37 m, which comprised the bearing stratum of the test
piles. The diameter of the test piles was 1.2 m, and the multi-level
O-cells were installed at depths of 26.9 m and 36.9 m, respec-

Fig. 9. Schematic Drawing of Test Piles and Definition of Movement and Pile Compression

tively. The load test was performed 20 days after the concrete
placement, thus the effect of the residual load was insignificant.
The detailed description about the test is reported in the paper of
Lee and Park (2008).
The O-cell test was performed in two stages. In the first
stage, the bottom O-cell (36.9 m deep) was expanded, and the
toe resistance Qd was measured. In the second stage, the middle
O-cell (26.9 m deep) was expanded to measure the Lower Side
Shear (LSS) and the Upper Side Shear (USS) under zero toe
resistance by the toe opening of the bottom O-cell. Therefore,
the total resistance can be represented by the sum [Qd+LSS
+USS].
If the equivalent load of [Qd+LSS+USS] is applied on the pile
head, the following pile compressions would occur along the pile
(Fig. 9); (1) d-EB = the pile compression by Qd, (2) head-LSS = the
pile compression along lower side by LSS, (3) d-LSS = the pile
compression along the upper side by the reaction of LSS, and (4)
head-USS = the pile compression along the upper side by USS.
Each pile compression was evaluated by adopting Eqs. (4) and
(6) as follows:
Qd L
d EB = -------EA

(9)

( LSS )L
d LSS = -------------------1
EA

(10)

head-LSS = pile compression along lower side during 2nd step (11)
= (D4 D1)
( USS )L
head USS = --------------------1 ( D5 D3 )
EA

(12)

where the notations D1, D3, D4, and D5 are defined in Fig. 9
L1: Pile length of upper side
LSS: Shaft resistance along lower side at (LSS, D1) curve
of second step
Qd: Toe resistance at (Qd, D1) curve of the first step
USS: Shaft resistance along upper side at (USS, D3) curve
of second step
Figure 10 shows the measured load-movement curves of the
O-cell test. At the first stage, the maximum downward toe movement occurred at about 60 mm, which was enough to mobilize
the toe resistance. At the second stage, the middle O-cell expansion induced the upward movement of about 80 mm and the
downward movement of 10 mm. The LSS at the 10 mm movement did not reach the maximum value. The LSS was assumed
to be constant after the maximum test load.
Figure 11 shows the comparison between the equivalent curve
using the proposed method and the measured curve using the
head-down load test. The comparison showed that the equivalent
curve using the proposed method showed good agreement with
the measured curve.
This case study shows that the proposed method can properly
evaluate the equivalent head-down curve without assuming unit
shaft resistance distribution. Moreover, the proposed method can

1174

KSCE Journal of Civil Engineering

Equivalent Head-down Load vs. Movement Relationships Evaluated from Bi-directional Pile Load Tests

Fig. 12. Soil Profile and the Strain-gage Instrumented Pile

Fig. 10. Multi-level O-cell Test Results: (a) Stage 1, (b) Stage 2 (Data
courtesy of Hyundai Engineering & Construction company)

Fig. 11. Comparison of Equivalent Load-movement Curves

especially be applied to the multi-level O-cell tests.

4.3 Case 3: Combined Load Tests for the Driven PHC Pile
in Busan, Korea
A long PHC pile, with a 600 mm diameter, was driven 56 m
deep into the deltaic area (Fig. 12). The soil profile consisted of a
fill layer, sand, and clay layers. The fill of 5 m thickness was placed
on the original ground surface in the mid-1990s. The piezometer
Vol. 16, No. 7 / November 2012

measurements indicated that excess pore pressure of about

20~30 kPa existed in the middle of soft clay layer at the time of
the pile installation. Therefore, residual load by the negative skin
friction was expected to develop along the pile during the dissipation of the excess pore pressure.
The test pile was a hollow cylindrical precast concrete pile,
known as a PHC pile, with a 600 mm outer diameter and an 85
mm thickness. The pile was instrumented with pairs of vibrating
wire strain gauges placed diametrically opposite at twelve levels.
The strain gauges were attached to a rebar cage, which was
inserted into the void of the pile after the driving. The bottom end
of the cage was connected to an O-cell with a 400 mm diameter.
Finally, the central void was filled with cement grout over its full
length above the O-cell.
The mobilization of the residual load was monitored for seven
months after the pile installation. A combination of the head-down
and O-cell tests was chosen because the soil resistance of the test
pile was estimated to exceed the compressive strength of the pile
concrete.
As a first step, the O-cell test was performed until the downward
movement of the pile toe reached 90 mm (Fig. 13b). However,
no head upward movement D3 occurred during the O-cell test,
because the O-cell load was not enough to fully mobilize the
shaft resistance. The head down test followed, and the ultimate
shaft resistance under the zero toe resistance, which resulted
from the toe opening, was confirmed (Fig. 14). The detailed test
results and the methods of converting strains to loads are also
given by Fellenius et al. (2009) and Kim et al. (2011).
In Fig. 13, two cyclic load-movement curves at the pile toe
obtained from the O-cell test are shown. The first cyclic loading
was applied to separate the O-cell from the pile toe, while the
second one was for the main analysis. Fig. 14 shows the curves
of (the head down load) vs. (the pile toe and head movements).
The pile toe movement was insignificant until the 4,200 kN, i.e.,
the pile compression occurred without the pile penetration.
Figure 15 shows the measured residual load and the load-dis-

1175

Sung-Ryul Kim and Sung-Gyo Chung

Fig. 15. Residual Load and Load Distribution at the Maximum Ocell Load

Fig. 13. Load-movement Curves Obtained from the O-cell Test

(The buoyancy weight of the pile was 230 kN): (a) Upward
Movement, (b) Downward Movement

Fig. 16. Comparison of the Equivalent and Measured Curves

the residual load before the head-down test (area EFJ) was used
for this analysis.
Figure 16 shows the comparison between the equivalent curve
and the directly measured curve from the head-down test. The
equivalent curve, which considered the residual load, agreed
well with the measured curve, whereas the equivalent curve that
ignored the residual load significantly overestimated the pile
movement.

5. Conclusions
Fig. 14. Load-movement Curves Obtained from the Head-down
Test

tribution of the maximum applied load in the O-cell test. The

shape of unit shaft resistance was not obtained by the O-cell test.
Thus, an attempt was made with the assumption that the unit
shaft resistance increased linearly with depth. That is, Eq. (8)
with the ratio (area EHI)/(area EFHI) equal to 2/3, was applied,
which is also proved with the area ratio of 0.65 calculated using
the load distribution measured by the head-down test. In addition,

A new method was proposed to construct the equivalent headdown load movement curve on the bi-directional load tests. The
proposed method was verified by comparing the equivalent
curve from the bi-directional load tests and the measured curve
from the head-down load tests. The following conclusions are
drawn from the study.
1. The proposed construction method evaluates the pile compression of the equivalent curves in two ways: (1) using the measured pile compression at a bi-directional load test without

1176

KSCE Journal of Civil Engineering

Equivalent Head-down Load vs. Movement Relationships Evaluated from Bi-directional Pile Load Tests

assuming the unit shaft resistance distribution, and (2) by considering the effect of residual load.
2. The residual load was negligible in the first and second case
studies, in which the load tests were performed within a month
after pile installation. A comparison between the equivalent
curve obtained using the proposed method and the measured
curve obtained from the head-down load tests showed that the
proposed method properly constructed the equivalent curve
using the measured compression in the O-cell tests, and that it
can be successfully applied to a multi level O-cell test.
3. In the case of a long PHC pile driven into the thick deposit, in
which a long term measurement of pile strains was performed,
the estimated equivalent load-movement curve obtained considering the effect of residual load agreed well with the measured curve from the head-down test.

Notations
The following symbols were used in this paper:
A: Sectional area of the pile
C: (distance to the centroid from the bottom of the unit
shaft resistance)/(overall length of unit shaft resistance)
D1: Downward pile toe movement at the O-cell tests
D2: Upward pile toe movement at the O-cell tests
D3: Upward pile head movement at the O-cell tests
E: Youngs modulus of pile
L: Pile penetration length
L1: Pile length of upper side
LSS: Shaft resistance along lower side below the O-cell
Qd: Downward load at pile toe which initiates toe resistance
Qu: Upward net load at pile toe which arouses shaft resistance
USS: Shaft resistance along upper side above the O-cell
: Coefficient, which was used to evaluate the pile compression by the Qu at the pile head in Kwon et al.s
method
: Additional pile compression by the different loading
direction between the O-cell and head-down tests
d: Pile compression induced by equivalent head-down
load of Qd
d-EB: Pile compression induced by Qd at the multi-level Ocell tests
d-LSS: Pile compression along upper side by the reaction of
LSS at the multi-level O-cell tests
head: Pile compression induced by equivalent head-down
load of Qu
head-LSS: Pile compression along lower side by LSS at the multilevel O-cell tests
head-USS: Pile compression along upper side by USS at the

Vol. 16, No. 7 / November 2012

multi-level O-cell tests

toe: Pile compression induced by upward toe load of Qu

Acknowledgements
This work was supported by the Korea Science and Engineering Foundation (KOSEF) NRL Program grant funded by the
Korea government (MEST) (No. R0A-2008-000-20076-0), and by
Basic Science Research Program through the National Research
Foundation of Korea (NRF) funded by the Ministry of Education,
Science and Technology (No. 2009-0067319).

References
Altaee, A., Evgin, E., and Fellenius, B. H. (1993). Load transfer for
piles in sand and the critical depth. Can. Geotech. J., Vol. 30, No. 3,
pp. 455-463.
Baker, C. N., Park, G., Braid, J. L., Drumright, E. E., and Mensah, F.
(1993). Drilled shafts for bridge foundations, Rep. No. FHWA RD
92 004, FHWA, Wa.
Fellenius, B. H. (2004). Unified design of piled foundations with
emphasis on settlement analysis. Proc., ASCE Geotechnical
Special Publication No. 125, Current Practices and Future Trends
in Deep Foundations, ASCE, Va., pp. 253-275.
Fellenius, B. H., Kim, S. R., and Chung, S. G. (2009). Long-term
monitoring of strain in instrumented piles. Journal of Geotechnical
and Geoenviromental Engineering, Vol. 135, No. 11, pp. 1583-1595.
Gregersen, O. S., Aas, G., and DiBiagio, E. (1973). Load tests on
friction piles in loose sand. Proc. 8th Int. Conf. Soil Mech. and
Found. Engrg, ICSMFE, pp. 109-117.
Kim, S. R., Chung, S. G., and Fellenius, B. H. (2011). Distribution of
residual load and true shaft resistance for a driven instrumented test
pile. Can. Geotech. J., Vol. 48, No. 4, pp. 583-598.
Kim, H. J. and Mission, J. L. (2011). Improved evaluation of equivalent top-down load-displacement curve from a bottom-up pile load
test. Journal of Geotechnical and Geoenviromental Engineering, Vol. 137, No. 6, pp. 568-578.
Kwon, O. S., Choi, Y. K., Kwon, O. K., and Kim, M. M. (2005).
Comparison of the bidirectional load test with the top-down load
test. Transportation Research Record. 1936, Transportation
Research Board, Washington, D.C., pp. 108-116.
Lee, J. S. and Park, Y. H. (2008). Equivalent pile load-head settlement
curve using a bi-directional pile load test. Computers and Geotechnics, Vol. 35, No. 2, pp. 124-133.
Loadtest (2001). Construction of the equivalent top-loaded loadsettlement curve from the results of an O-cell test, Loadtest
Appendix to Reports.
Osterberg, J. O. (1998). The Osterberg load test method for drilled
shafts and driven piles-The First ten years. Proceedings 7th
International Conference on Deep Foundations, Vienna, Austria,
June 15-17, Deep Foundation Institute, Englewood Cliffs, N.J.
Xi, X., Chen, L., and Liu, W. (2010). An analytical solution to
transform O-cell pile test data into conventional load-settlement
curve. GeoShanghai 2010 International Conference, Geotechnical
Special Publication No. 205, pp. 192-199.

1177