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

net/publication/237046498

Vertically Loaded Single Piles in Gibson Soil

Article  in  Journal of Geotechnical and Geoenvironmental Engineering · February 2000

DOI: 10.1061/(ASCE)1090-0241(2000)126:2(189)

CITATIONS READS

40 1,115

1 author:

Wei Dong Guo

University of Wollongong
91 PUBLICATIONS   1,194 CITATIONS   

SEE PROFILE

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

Closed-form solutions for non-linear response of laterally loaded-pile groups View project

Onshore-Offshore piles (beams) subjected to dynamic (cyclic) loading View project

All content following this page was uploaded by Wei Dong Guo on 26 May 2014.

The user has requested enhancement of the downloaded file.

 VERTICALLY LOADED SINGLE PILES IN GIBSON SOIL
By Wei Dong Guo1

ABSTRACT: The available closed-form solutions for vertically loaded piles have been, strictly speaking, limited
to homogeneous soil, or nonhomogeneous soil with the shear modulus as a power of depth. The latter solutions
—based on a zero shear modulus at ground surface—are generally sufficiently accurate for normally consolidated
soil. For overconsolidated soil, however, there is generally a nonzero shear modulus at the surface, which may
affect pile response. In this note, rigorous closed-form solutions are established to account for the nonhomo-
geneity of soil profile with nonzero shear modulus at ground surface. The solutions are developed using a load
transfer approach, and are shown to give satisfactory results in comparison with a more rigorous continuum-
based numerical approach, when the proposed load transfer factors are adopted.

INTRODUCTION First, the distribution of soil shear modulus, G, down a pile

To date, the most efficient way to analyze a large group of is assumed to be a power function of depth (Guo et al. 1998):
piles is by ‘‘hybrid analysis,’’ which combines numerical and G = Ag(␣g ⫹ z)n (1)
analytical solutions to perform a complete analysis (Chow
1987; Guo 1997). Soil nonhomogeneity has a considerable ef- where n, ␣g, Ag = constants, and z = depth below the ground
fect on prediction of the pile-soil-pile interaction and on group surface. Below the pile-base level, shear modulus is assumed
pile behavior (Poulos 1989; Guo and Randolph 1999). There- to remain constant, Gb, which may be different from the shear
fore, it is essential to develop closed-form solutions that can modulus at just above the pile-base level, GL. This difference
handle nonhomogeneity more accurately (e.g., Rajapakse 1990). in shear modulus at the base is expressed by the ratio, ␰b(=GL/
The available closed-form solutions for vertically loaded Gb). In this note, however, ␰b is taken as unity.
piles have been, strictly speaking, limited to homogenous soil Second, corresponding to the shear modulus of (1), the var-
(Murff 1975; Motta 1994); until recently, though, Guo and iation of limiting shear stress, ␶f , with the depth, is assumed
Randolph (1997) developed new solutions, ones that can rig- to be
orously account for soil nonhomogeneity by describing the
␶f = Av (␣v ⫹ z)␪ (2)
shear modulus as a power of depth. The solutions, based on
assuming a zero modulus at ground surface, are generally suf- where ␪, ␣v, Av = constants. For simplicity, ␣v and ␪ are taken
ficiently accurate for normally consolidated soil. For overcon- identical values to ␣g and n, respectively. Therefore, the ratio
solidated soil, however, there is generally a nonzero shear of modulus to shaft limiting stress is independent on the depth,
modulus at the surface (simply referred to as Gibson soil in and is equal to Ag /Av.
this note), which may affect pile response (Guo et al., unpub- Third, the nonhomogeneity factor is expressed by the ratio,
lished, 1998). ␳g, of the average soil shear modulus over the pile length and
In this technical note, the previous closed-form solutions for the modulus at the pile-base level, GL. Therefore

冉 冉 冊冊
a vertically loaded pile in a nonhomogeneous, elastic-plastic n
soil (Guo and Randolph 1997) are extended to account for 1 ␣g ␣g ␣g /L
␳g = 1⫹ ⫺ (3)
nonzero shear modulus at ground surface. The solutions are 1⫹n L L 1 ⫹ ␣g /L
obtained in a similar way to that described by Guo et al.
(1998), which were expressed in modified Bessel functions of Fourth, the pile-soil relative stiffness factor, ␭, is defined as
noninteger order. Numerical estimates of the solutions are per- the ratio of pile Young’s modulus, Ep, to the shear modulus at
formed by either Mathcad and/or a newly designed spread- pile-base level, GL, i.e.
sheet program. ␭ = Ep /GL (4)
The solutions are based on a load transfer approach (Coyle
and Reese 1966; Randolph and Wroth 1978; Kraft et al. 1981;
Load Transfer Models
Guo and Randolph 1997, 1998), treating the soil as indepen-
dent springs. For the approach, load transfer factors are gen- Closed-form solutions established in this note are based on
erated to account for the nonzero input of shear modulus, in a theoretical load transfer model for nonhomogeneous soil. In
light of the proposal by Guo and Randolph (1998). The ac- the model, the shaft displacement, w, is correlated to the local
curacy and suitability of the new solutions are checked using shaft stress and shear modulus by (Randolph and Wroth 1978):
a more rigorous numerical solution of the finite-difference
code, FLAC (1992). ␶oro
w= ␨ (5)
G
LOAD TRANSFER MODELS
and
Expressions of Nonhomogeneity
The soil profile addressed and the relevant nondimensional
parameters adopted in this note are briefly described below.
␨ = ln 冉冊
rm
ro
(6)

1
Postdoctoral Res. Fellow, Dept. of Civ. Engrg., National Univ. of Sin- where ␶o = local shaft shear stress; ro = pile radius; rm = max-
gapore, Singapore 119260. E-mail: [email protected] imum radius of influence of the pile, beyond which the shear
Note. Discussion open until July 1, 2000. To extend the closing date stress becomes negligible, and may be expressed in terms of
one month, a written request must be filed with the ASCE Manager of
Journals. The manuscript for this technical note was submitted for review
the pile length, L, as (Guo 1997; Guo and Randolph 1997,
and possible publication on February 17, 1999. This technical note is part 1998)
of the Journal of Geotechnical and Geoenvironmental Engineering,
Vol. 126, No. 2, February, 2000. 䉷ASCE, ISSN 1090-0241/00/0002- 1 ⫺ ␯s
rm = A L ⫹ Bro (7)
0189–0193/$8.00 ⫹ $.50 per page. Technical Note No. 20287. 1⫹n

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING / FEBRUARY 2000 / 189

 Nevertheless, for predicting load and/or displacement distri-
bution profiles down a pile, more accurate expression is
needed (Guo 1997; Guo and Randolph 1998).

OVERALL PILE-SOIL INTERACTION

Generally, a pile is supposed to behave elastically, with con-
stant diameter and Young’s modulus. Therefore, the governing
equation for pile-soil interaction may be written as (Randolph
and Wroth 1978)
d 2u(z) 2␲ro␶o
= (13)
d 2z Ep Ap
where Ap = cross-sectional area of an equivalent solid cylinder
pile, and u(z) = axial pile deformation.

Elastic Solution
Within the elastic range, shaft stress, ␶o in (13) can be ex-
pressed by the local displacement as prescribed by (5). Thus,
the basic differential equation governing the axial deformation
for a pile fully embedded in the nonhomogeneous soil, de-
scribed by (1), is derived as
d 2u(z) Ag 2␲
FIG. 1. Typical Pile-Soil System Addressed = (␣g ⫹ z)nw(z) (14)
d 2z Ep Ap ␨
where ␯s = Poisson’s ratio of the soil; B may generally be taken The axial pile displacement, u(z), should equal the pile-soil
as 5; A is dependent on the ratio of the embedded depth of relative displacement, w(z), if any external soil subsidence is
underlying rigid layer, H to the pile length, L (Fig. 1). For the ignored. Normally the load transfer factor ␨ can be taken as a
shear modulus distribution of (1), A may be estimated from constant along a pile depth (Guo and Randolph 1998). There-
Eq. (8), in which the original factor n (Guo and Randolph fore, (14) can be solved using Bessel functions of noninteger
1998) has been replaced with an equivalent nonhomogenous order:

冉 冊冉 冊
factor ne. Thus, it follows that 1/2

冉 冉 冊 冊
␣g ⫹ z C3(z) ⫹ ␹v C4(z)
Ah 1 0.4 ⫺ ␯s 2 w(z) = wb (15)
A= ⫹ ⫹ C␭(␯s ⫺ 0.4) ␣g ⫹ L C3(L)
Aoh 1 ⫹ ne ne ⫹ 0.4 1 ⫺ 0.3ne

冉 冊冉 冊
1/2
(8) ␣g ⫹ z C1(z) ⫹ ␹v C2(z)
P(z) = ks Ep Apwb(␣g ⫹ z)n/2
␣g ⫹ L C3(L)
where C␭ = 0, 0.5, and 1.0 for ␭ = 300, 1,000, and 10,000.
Aoh is Ah at a ratio of H/L = 4, Ah is given by the following (16)
equation: where w(z), P(z) = displacement and load at a depth of z (0 <

冉 冊
z ⱕ L), respectively; and Ci (i = 1 to 4) is expressed as the
Ah = 0.124e 2.23␳g 1 ⫺ e 1⫺H/L ⫹ 1.01e 0.11ne (9) modified Bessel functions, K, and I of noninteger order, m and
m-1
The ne may be assessed using the following equation: C1(z) = ⫺Km⫺1Im⫺1( y) ⫹ Km⫺1( y)Im⫺1;
ne = 1/␳g ⫺ 1 (10) C2(z) = Km Im⫺1( y) ⫹ Km⫺1( y)Im;
As the pile-head load increases, the mobilized shaft shear C3(z) = Km⫺1Im( y) ⫹ Km( y)Im⫺1;
stress will reach the limiting value, ␶f . Therefore, the local
limiting displacement, we , can be obtained by C4(z) = ⫺Km Im( y) ⫹ Km( y)Im (17)

we =
Av
␨ro (11) where m = 1/(n ⫹ 2), Im, Im⫺1, Km⫺1, and Km = values of the
Ag Bessel functions at z = L. The variable y is
Thereafter, as the pile-soil relative displacement exceeds the y = 2mks(␣g ⫹ z)1/(2m) (18)
limiting value, the shear stress is kept as ␶f , (i.e., an ideal
elastic-plastic load transfer model is adopted). With (1) and The stiffness factor ks is

冑 冉 冊
(2), the limiting shaft displacement is a constant down the pile. 1/(2m)

The base settlement can be estimated through the solution for ␣g ⫹ L 2 1

ks = (19)
a rigid punch acting on a half-space: ro ␭␨ ␣g ⫹ L

Pb(1 ⫺ ␯s)␻
wb = (12) and the ratio ␹v is given by
4roGb
where Pb = mobilized base load; ␻ = pile-base shape and depth
factor, which is taken as unity (Randolph and Wroth 1978)
␹v =
2兹2
␲(1 ⫺ ␯s)␻␰b 冑 ␨
␭
(20)

since it can only lead to rather small (normally <6%; Guo and
Randolph 1998) difference in predicting pile-head stiffness. At any depth, z, the stiffness can be derived as
190 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING / FEBRUARY 2000
 P(z)
GLw(z)r0
= 兹2␲ 冑 ␭
␨
Cv (z) (21)

where

冉 冊
n/2
C1(z) ⫹ ␹vC2(z) ␣g ⫹ z
Cv (z) = (22)
C3(z) ⫹ ␹v C4(z) ␣g ⫹ L

At the ground surface, where z = 0, it is necessary to take the

limiting value of Cv (z) as z approaches zero. From (15) and
(16), base settlement can be written as a function of pile-load
and displacement. The accuracy of the above closed-form so-
lutions (hereafter denoted by CF) has been checked by Math-
cad. Further corroboration by continuum-based finite-differ-
ence analysis is shown later.

Verification of Elastic Theory

The closed-form solutions outlined above are based on the FIG. 3. Effect of ␣g on Pile-Head Stiffness
uncoupled load transfer analysis. Generally, the solutions com-
pare well with more rigorous numerical approaches for ␣g = stiffnesses are tabulated in Table 2, together with those from
0 (Guo 1997; Guo and Randolph 1997, 1998). Thus the cur- FLAC analysis. It shows that in comparison with the FLAC
rent verification will be limited to the new inclusion of the analysis, the CF estimation gives a slightly higher pile-head
nonzero shear modulus (␣g ≠ 0) in the equations. For this stiffness for short piles, but a lower stiffness for long piles
verification, a continuum-based numerical analysis, using the (particularly at n = 1). However, the difference between the
finite-difference program FLAC (Itasca 1992), was performed two analyses is generally less than 5%.
in a way similar to that reported by Guo and Randolph (1997),
and using the reported pile and soil properties, except that the Elastic-Plastic Solution
new factor ␣g in (1) was assigned different values (Fig. 1). In
particular, in current analysis H/L was taken as 4, Young’s As the pile-head load increases, pile-soil relative slip is as-
modulus of the pile, Ep was taken as 30 GPa and Poisson’s sumed to commence from the ground surface and at any stage
ratio, ␯p as 0.2. Given constant values, say, of ␭ = 1,000, GL during loading may be taken to have developed to a depth
= 30 MPa, regardless of n (n > 0), as the value of ␣g increases, called transition depth (L1), at which the shaft displacement,
the value of Ag in (1) must reduce accordingly. Thus, once the w, corresponds to the local limiting displacement. The upper
␣g is sufficiently high, the soil approaches a homogenous me- part of the pile, above the transition depth, is in plastic state,
dium (as may be noted in Figs. 1 and 2). Pile-head stiffness while the lower part below the depth is in an elastic state.
from the FLAC analysis has been illustrated in Fig. 3. It in- Within the plastic state, the shaft shear stress in (13) should
dicates that as the values of ␣g increase, pile-head stiffness be replaced by the limiting shaft stress from (2). Pile-head load
approaches the upper limit of that for homogeneous case as and settlement are, therefore, expressed, respectively, as a sum
well. of the elastic part represented by letters with subscript of e,
The pile-head stiffness was estimated using (21), in which and the plastic part:
z = 0, and A was estimated from (8). In the estimation, the 2␲ro Av [(␣v ⫹ ␮L)1⫹␪ ⫺ (␣v)1⫹␪]
following values were taken: H/L = 4, ␯s = 0.4, ␰b = 1, ␻ = Pt = Pe ⫹ (23)
␪⫹1
1, and B = 1. In particular, the values of the ne and the cor-
responding A are tabulated in Table 1 for each case. The low
value of B = 1, obtained previously (Guo and Randolph 1998),
was adopted herein, since the FLAC analysis normally gives
wt =
2␲roAv
Ep Ap 冉
␣ 2⫹␪
v ⫹ (␣v ⫹ ␮L)1⫹␪((1 ⫹ ␪)␮L ⫺ ␣v)
(1 ⫹ ␪)(2 ⫹ ␪) 冊
slightly higher pile-head stiffness compared with other ap- ␮LPe
proaches, and a lower value of B gives a slightly higher pile- ⫹ ⫹ we
Ep Ap (24)
head stiffness, except for short piles. The estimated pile-head
where ␮ = L1 /L is defined as degree of slip (0 < ␮ ⱕ 1); L1
= the length of the upper plastic part; and Pe = the pile load
at the transition depth. With w(z) = we at the transition depth
of L1, Pe = P(L1) can be readily estimated from (21); therefore,
(23) can be rewritten as
Pt = we ks Ep Ap(␣g ⫹ L)n/2Cv (␮L)

2␲ro Av [(␣v ⫹ ␮L)1⫹␪ ⫺ (␣v)1⫹␪

⫹
␪⫹1 (25)
Similarly, as a result of (24), and substituting for Pe, the pile-
head settlement is expressed as
wt = we[1 ⫹ ␮Lks(␣g ⫹ L)n/2Cv (␮L)]

2␲ro Av ␣ v2⫹␪ ⫹ (␣v ⫹ ␮L)1⫹␪((1 ⫹ ␪)␮L ⫺ ␣v)

⫹
Ep Ap (1 ⫹ ␪)(2 ⫹ ␪) (26)
FIG. 2. Effect of ␣g on ne These solutions provide three important results: (1) For a given
JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING / FEBRUARY 2000 / 191
 TABLE 1. Estimation of ne and Parameter A for CF Analysis

L /ro = 20 L /ro = 40 L /ro = 100

␣g n = 0.5 n = 1.0 n = 0.5 n = 1.0 n = 0.5 n = 1.0
(1) (2) (3) (4) (5) (6) (7)
.50 1.00 .50 1.00 .50 1.00
0
1.57 1.43 1.57 1.43 1.57 1.43
.265 .556 .341 .714 .417 .862
2
1.72 1.54 1.66 1.49 1.61 1.45
.160 .333 .238 .50 .341 .714
5
1.81 1.67 1.74 1.57 1.66 1.49
.097 .20 .160 .333 .265 .556
10
1.88 1.77 1.81 1.67 1.72 1.54
.055 .111 .097 .200 .184 .385
20
1.93 1.86 1.88 1.77 1.79 1.63
Note: Numerator is value of ne estimated by (10), while denominator is value of A by (8). In the estimation, H = 4L, ␭ = 1,000, and ␯s = 0.4.

TABLE 2. Comparison of Pile-Head Stiffness from FLAC and CF Analyses

L /ro = 20 L /ro = 40 L /ro = 100
␣g n = 0.5 n = 1.0 n = 0.5 n = 1.0 n = 0.5 n = 1.0
(1) (2) (3) (4) (5) (6) (7)
32.3 27.4 41.9 35.1 46.2 37.4
0
31.2 25.6 39.9 32.2 43.2 34.2
35.3 31.5 44.9 38.7 — 39.7
2
35.1 30.6 43.7 36.4 46.1 36.8
37.1 34.3 47.1 42.0 50.7 42.5
5
37.3 33.9 46.4 40.4 48.8 39.9
38.2 36.3 49.0 45.1 53.0 45.8
10
38.7 36.4 48.8 44.1 51.6 43.6
39.0 37.9 50.0 48.0 55.7 50.0
20
39.7 38.4 50.8 47.6 54.6 48.3
Note: Numerator is estimated by FLAC analysis, while denominator is predicted by CF solutions. In FLAC analyses, H was taken as 4L, ␭ as 1,000,
and ␯s as 0.4.

degrees of slip, the pile-head load and settlement can be es- PILE program (Guo 1997), which in turn adopts a load trans-
timated by (25) and (26), respectively; therefore the full pile- fer relationship similar to (5) but includes nonlinear elastic-
head load-settlement relationship may be obtained; (2) for a plastic behavior. For a pile of L/ro = 100, embedded in a soil
given pile-head load, the corresponding degree of slip of the of n = ␪ = 0.5, ␭ = 1,000, Ag /Av = 350, ␰b = 1.0, ␯s = 0.4, H/
pile can be back-figured from (25); and (3) the distribution L = 4.0, the predicted pile-head load displacement relation-
profile of either load or displacement can be readily obtained, ships are demonstrated in Fig. 4(a), using (25) and (26). In the
at any stage of the elastic-plastic development. Within the up- prediction, ␻ was taken as 1, A was estimated using ne of 0.5,
per plastic part, at any depth of z: first, similar to (23), the 0.238, and 0, respectively, which in turn were obtained from
load P(z) can be predicted by (3) for the given ␣g of 0, 12.5, and ⬁, respectively. For the
pile, GASPILE analyses were performed using 20 segments,
(␣v ⫹ ␮L)␪⫹1 ⫺ (␣v ⫹ z)1⫹␪ and using identical parameters to those mentioned above, but
P(z) = Pe ⫹ 2␲ro Av (27)
␪⫹1 the effect of nonlinear stress on the ␨ was taken into account.
The predicted results are shown in Fig. 4(a). It shows that the
Second, similar to (24), the displacement w(z) can be obtained effect of the nonlinear stress on the pile-head load displace-
by ment is rather minor. However, the effect on load-deformation
profiles may become appreciable (Guo and Randolph 1997).
Pe(␮L ⫺ z) 2␲ro Av For instance, given Pt = 2MN, the predicted load profile using
w(z) = we ⫹ ⫹
Ep Ap Ep A p (16) and (27), and the displacement using (15) and (28), are

冉 冊
presented in Figs. 4(b and c), together with the GASPILE anal-
(␣v ⫹ z)2⫹␪ ⫹ (2 ⫹ ␪)(␮L ⫺ z)(␣v ⫹ ␮L)␪⫹1 ⫺ (␣v ⫹ ␮L)2⫹␪ yses. The profiles may be slightly different from a continuum-
⭈
(1 ⫹ ␪)(2 ⫹ ␪) based numerical approach, since the real value of ␻ may be
(28) equal to 1.36, 1.437, and 1.515 for the ␣g of 0, 12.5, and ⬁,
as estimated by Guo and Randolph’s (1998) equation, in which
The current analysis is limited to ␣v = ␣g, and n = ␪, but the the n is replaced with ne. Relevant critical values were ob-
physical implications of n and ␣g (related to elastic state) and tained from the CF solutions, and are shown in Table 3. It
␣v and ␪ (to plastic state) are completely different. Thus, all demonstrates that increase in the ␣g leads to decrease in the
the parameters ␣v, ␣g, n, and ␪ are preserved in the equations. degree of slip, ␮, and base settlement, wb, but increase in the
The accuracy of (25)–(28) has been verified using the GAS- load, Pe, and limiting displacement, we.
192 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING / FEBRUARY 2000
 Guo, W. D. (1997). ‘‘Analytical and numerical solutions for pile foun-
dations,’’ PhD thesis, The University of Western Australia, Nedlands,
Australia.
Guo, W. D., and Randolph, M. F. (1997). ‘‘Vertically loaded piles in non-
homogeneous media.’’ Int. J. Numer. and Anal. Methods in Geomech.,
21(8), 507–532.
Guo, W. D., and Randolph, M. F. (1998). ‘‘Rationality of load transfer
approach for pile analysis.’’ Comp. and Geotechnics, 23(1–2), 85–112.
Guo, W. D., and Randolph, M. F. (1999). ‘‘An efficient approach for
settlement prediction of pile groups.’’ Geotechnique, 49(2), 161–179.
Kraft, L. M., Ray, R. P., and Kagawa, T. (1981). ‘‘Theoretical t-z curves.’’
J. Geotech. Engrg. Div., ASCE, 107(11), 1543–1561.
Motta, E. (1994). ‘‘Approximate elastic-plastic solution for axially loaded
piles.’’ J. Geotech. Engrg., ASCE, 120(9), 1616–1624.
Murff, J. D. (1975). ‘‘Response of axially loaded piles.’’ J. Geotech.
Engrg. Div., ASCE, 101(3), 357–360.
Poulos, H. G. (1989). ‘‘Pile behavior—theory and application.’’ Rankine
lecture, Geotechnique, 39(3), 365–415.
Rajapakse, R. K. N. D. (1990). ‘‘Response of axially loaded elastic pile
in a Gibson soil.’’ Geotechnique, 40(2), 237–249.
Randolph, M. F., and Wroth, C. P. (1978). ‘‘Analysis of deformation of
vertically loaded piles.’’ J. Geotech. Engrg. Div., ASCE, 104(12),
1465–1488.

APPENDIX II. NOTATION

The following symbols are used in this paper:

Ag = constant for soil shear modulus distribution;

Av = constant for shaft limit stress distribution;
Ap = cross-sectional area of equivalent solid cylinder pile;
Ep = elastic modulus of pile body;
G = soil shear modulus;
Gb = shear modulus at just beneath pile-base level;
GL = soil shear modulus at just above pile-base level;
ks = factor representing pile-soil relative stiffness;
L = embedded pile length;
L1 = depth of transition from elastic to plastic phase;
FIG. 4. Effect of ␣g on Pile Response (L/ro = 100, ␭ = 1,000, n = L2 = length of elastic part of pile under given load;
␪ = 0.5, Ag /Av = 350, ␣g = as shown) n = power for shear modulus distribution;
ne = equivalent nonhomogeneity factor, estimated by (10);
TABLE 3. Critical Values at Pt = 2 MN from CF Analysis P(z) = axial force of pile body at depth of z;
Pe = axial load at depth of transition (L1) from elastic to plastic
␣g 0 2.5 12.5 ⬁ phase;
(1) (2) (3) (4) (5) Pt = load acting on pile head;
␮㛳Pe (kN) 0.527㛳1141 0.443㛳1211 0.306㛳1323 0.126㛳1575 ro = pile radius;
we (mm)㛳wb rm = radius of zone of shaft shear influence;
(mm) 3.25㛳2.04 3.28㛳1.80 3.33㛳1.45 3.43㛳1.02 u(z) = axial pile displacement at depth of z;
wb = settlement of pile base;
we = limiting elastic shaft displacement calculated by using ␶f
CONCLUSIONS in equation;
wt = pile-head settlement;
The analysis outlined in this note provides a simplified ap-
w(z) = deformation of pile body at depth of z;
proach to the analysis of a pile in nonhomogenous soil media ␣g = shear modulus factor for ground surface;
with nonzero input of shear modulus at ground surface. The ␨ = measured influence of load transfer;
approach based on the load transfer approach agrees well with ␭ = relative stiffness ratio between pile Young’s modulus and
those obtained from numerical analyses. soil shear modulus at just above base level (Ep /GL);
␪ = power for shaft limit stress distribution;
APPENDIX I. REFERENCES ␮ = degree of pile-soil relative slip;
␯s = Poisson’s ratio of soil;
Chow, Y. K. (1987). ‘‘Axial and lateral response of pile groups embedded ␰ = end-bearing shear modulus nonhomogeneous factor (GL /
in nonhomogenous soils.’’ Int. J. Numer. and Anal. Meth. in Geomech., Gb);
11, 621–638.
Coyle, H. M., and Reese, L. C. (1966). ‘‘Load transfer for axially loaded
␳p = ratio of average soil shear modulus over pile length and
piles in clay.’’ J. Soil Mech. and Found. Engrg. Div., ASCE, 92(2), 1– modulus at L;
26. ␶f = limiting local shaft stress;
FLAC—Users’ Manual. (1992). Itasca Consulting Group, Minneapolis, ␶o = shear stress on pile soil interface; and
Minn. ␻ = pile-base shape and depth factor.

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING / FEBRUARY 2000 / 193

View publication stats