Piglet 6-2 Manual-June 2021

PIGLET
ANALYSIS AND DESIGN OF
PILE GROUPS
M. F. RANDOLPH
VERSION 6.2
RELEASED JUNE 2021
(MANUAL LAST EDITED JUNE 2021)

PIGLET
ANALYSIS AND DESIGN OF
PILE GROUPS
M. F. RANDOLPH
VERSION 6.2: RELEASED JUNE 2021
(MANUAL LAST EDITED JUNE 2021)
The accuracy of this program has been checked over a period of years, and it is believed that,
within the limitations of the analytical model, results obtained with the program are correct.
However, the author accepts no responsibility for the relevance of the results to a particular
engineering problem.
Technical support in relation to operation of the program, or in respect of engineering

assistance, may be obtained from the author, who reserves the right to make a charge for such
assistance.
Contact Address: 29/82 King Street

Perth
WA 6000
Australia
Telephone: +61 421 596 075

Email: mark@randolphonking.com
CONTENTS
Page No.
PART A: GENERAL DESCRIPTION
1 INTRODUCTION............................................................................................................... 1
2 IDEALISATION OF SOIL PROPERTIES ........................................................................ 1
3 RESPONSE OF PILES TO AXIAL LOADING ................................................................ 3
3.1 Solution for single axially loaded piles ...................................................................... 3
3.2 Extension of solution to pile groups ........................................................................... 5
4 RESPONSE OF PILES TO TORSIONAL LOADING ...................................................... 6
5 RESPONSE OF PILES TO LATERAL LOADING .......................................................... 8
5.1 Deformation of single laterally loaded piles .............................................................. 8
5.2 Interaction between laterally loaded piles .................................................................. 9
6 ANALYSIS OF PILE GROUP ......................................................................................... 10
6.1 Treatment of raking piles ......................................................................................... 11
6.2 Allowance for free-standing length of piles ............................................................. 11
7 NON-LINEAR PILE RESPONSE .................................................................................... 12
7.1 Elastic-plastic analysis (E-P) .................................................................................... 12
7.2 Non-linear axial and lateral response (NL-P) .......................................................... 12

8 EXAMPLE APPLICATION ............................................................................................. 13
8.1 Example 1: model tests of Davisson and Salley (1970) ........................................... 13
8.2 Example 2: field-scale lateral load tests of small pile group Robbins et al. ............ 14
9 STRUCTURE OF PROGRAM......................................................................................... 16
10 PROGRAM INPUT .......................................................................................................... 16
10.1 Data input and editing .............................................................................................. 16
10.2 Data Items ................................................................................................................ 17
11 PROGRAM OUTPUT ...................................................................................................... 22

12 REFERENCES .................................................................................................................. 22
FIGURE TITLES .................................................................................................................... 25
FIGURES ................................................................................................................................. 26
JUNE 2021 PIGLET MANUAL (Version 6.2) M.F. RANDOLPH
PART A: GENERAL DESCRIPTION
1 INTRODUCTION
The computer program, PIGLET, analyses the load deformation response of pile groups under
general loading conditions. The program is based on a number of approximate, but compact,
solutions for the response of single piles to axial, torsional and lateral loading, with due allowance
made for the effects of interaction between piles in the group. In these solutions, the soil is modelled
as a linear elastic material, with a stiffness that varies linearly with depth. No check of the overall
stability of the pile group is made within the program; such calculations should form a separate part
of the design. However, the current version of PIGLET allows simulation of non-linear response of
individual piles and imposition of limiting axial loads
The program has developed gradually over the last forty years, with the doctoral research of the
author (Randolph, 1977) forming the basis for the original version. This manual is a further revision
of a report describing the program, originally published in 1980. The current manual is based on the
version of PIGLET that has been implemented within an Excel workbook, calling a Fortran dynamic
link library (DLL). The technical basis of the software remains largely unchanged from previous
versions, except for the recent inclusion of the ability to simulate non-linear axial and lateral response
of individual piles. Additionally minor extensions have been implemented: allowing different pile
bending stiffness in the x:z and y:z planes; and a continuous normalised degree of rotational fixity
between pile and pile cap rather than choosing between extreme ‘pinned’ and ‘fixed’ conditions.
In order to minimise the amount of computation required, three separate ‘scopes’ of analysis are
identified, depending on the type of loading to be applied to the group. The three cases are:
(1) vertical loading only;
(2) vertical and horizontal loading in a single plane;
(3) general three-dimensional loading, including torsion.
For the latter two cases, the pile group is assumed capped by a rigid pile cap, with the piles either
pinned or built-in to the cap. In the first case, the user may also specify a fully flexible pile cap.
The pile cap is assumed always to be clear of the ground surface, with no direct transfer of load to
the ground. A non-zero ‘free-standing’ length of pile may be included between the pile cap and the
effective ground surface.
2 IDEALISATION OF SOIL PROPERTIES
Soil is by nature non-linear in its stress-strain behaviour, even at low stress levels. This non-linearity
may be modelled in an approximate fashion for the analysis of single piles by the use of load transfer
methods of analysis, where the soil continuum is replaced by a series of springs acting along the
length of the pile. Extension of such analysis to pile groups is possible by adopting a hybrid soil
model, combining elastic interactive effects with the load transfer analysis of each single pile (O'Neill
et al, 1977). This approach risks inconsistency of the approach between the non-linear single pile
1
response and pile-pile interaction. An alternative has been implemented in this version of PIGLET,
with elastic continuum analysis used as the basis for both the single pile response and pile-pile
interaction.
In many applications, it is sufficient to adopt a linear elastic model for the soil for calculating
deformations and load distributions among piles in a group, under working load conditions.
Independent checks should be performed to ensure that the elastic assumption is reasonable at the
load and deformation levels determined. For pile groups of practical size (in terms of number of
piles) additional deformation due to interactive effects will generally dominate that due to non-linear
effects. A possible exception to this is where significant plastic deformation occurs between pile and
soil. The current version of the software now includes a more general non-linear response of
individual piles, while maintaining purely elastic interaction between piles in the group.
Although the soil domain as a whole has been assumed to deform elastically, less restriction has been
imposed on the relative homogeneity of the soil deposit. It has been assumed that the soil may be
modelled by a material where the stiffness varies linearly with depth. While this does not allow
layered soil profiles to be treated rigorously, such deposits may be analysed by choosing a suitable
average stiffness for the strata penetrated by the piles, and adopting a linear variation of stiffness with
depth that reflects the general trend present in the actual profile. In addition, the special case of
end-bearing (or partially end-bearing) piles has been catered for by the inclusion of a facility for
specifying a soil of increased stiffness below the level of the pile bases.
In summary, the soil is idealised as an elastic material where the stiffness varies as shown in Figure
1. The stiffness is characterised by a shear modulus, G, and Poisson's ratio, , noting that the shear
modulus is related to the Young's modulus, E, by E = 2(1 + )G . The properties that need to be
specified are:
(a) the value of shear modulus at the ground surface, G0.
(b) the rate of increase of shear modulus with depth, m = dG/dz;
(c) the value of shear modulus at the pile base, Gb.
(d) Poisson's ratio for the soil, , assumed constant with depth.
Treatment of the axial and lateral response of the piles independently allows additional freedom when
choosing soil properties. For the majority of piles used in practice, deformation under lateral load
occurs only in the upper part of the pile. Because of this, and to allow for the high strains that occur
locally near the head of a laterally loaded pile, it is advantageous to be able to specify different soil
properties for the analysis of the lateral load deformation response. In particular, it is often advisable
to adopt a value of zero for the shear modulus at ground level, G0, when considering lateral loading.
In the program, the same soil properties are assumed for axial and torsional loading, while different
values of shear modulus may be specified for lateral loading (retaining the same value of Poisson's
ratio).
Separate parameters are included to simulate the non-linear response of individual piles for axial (and
torsional) and lateral loading, in both cases using a generalised hyperbolic relationship each described
by two parameters. Non-linear analysis is undertaken incrementally, according to a specified number
of load steps.
2
3 RESPONSE OF PILES TO AXIAL LOADING
3.1 Solution for single axially loaded piles
An approximate closed form solution for single axially loaded piles has been described in detail for
floating piles by Randolph and Wroth (1978a), and extended to end-bearing piles by Randolph and
Wroth (1978b). The solution is based on the technique of treating load transferred from the pile shaft
separately from that at the pile base. The soil is effectively considered in two layers, divided by an
imaginary line drawn at the level of the pile base (see Figure 2). The upper layer, above the line AB,
is considered to be deformed solely by the shear stresses acting down the pile shaft, while the lower
layer is deformed by the load transmitted to the pile base. Some interaction will occur between the
upper and lower layers, which will serve to limit the radial extent of the deformation in the upper
layer. To illustrate the method of analysis, the solution for a rigid pile will be developed here.
The load settlement ratio for the pile base is obtained directly from the Boussinesq solution as
Pb 2
 (1)
G b d b wb 1  
where P is the load, w the settlement and d the pile diameter, the subscript b referring to the pile base.
Turning to the pile shaft, considerations of vertical equilibrium entail that the shear stress, , at any
depth falls off inversely with the radius, r, as (Cooke, 1974; Baguelin et al, 1975; Frank, 1974)
 0 r0  0 d
  (2)
r 2r
where the subscript 0 denotes conditions at the pile shaft. The shear strains,  =  /G, derived from
this equation may be integrated to give the vertical deformation field, which will vary logarithmically
with radius. In particular, if it is assumed that there is some radius, rm, at which the vertical
deformations are effectively zero, then the settlement of the pile shaft may be written
0d
ws   (3)
2G
where  = n(rm/r0). Effectively, this equation is identical to that assumed in linear load transfer
analysis, or a Winkler spring model for the soil response. In such a model, the load transferred from
the pile at any depth is P = d0, and the stiffness is given by
P 2
k  G (4)
ws 
Equation (3) may be combined with equation (1) to give the overall load settlement ratio for a rigid
pile of
3
Pt 2 2 L
  (5)
G  dwt 1      d
where  = db/d is the ratio of underream,  = GL/Gb is the ratio of end-bearing, and the subscript t
denotes conditions at the top of the pile.
Development of the full solution, which takes account of compression of the pile is given in detail
by Randolph and Wroth (1978a). Effectively, equation (3) is taken to act at each point down the
length of the pile, just as in a linear load transfer analysis. The final expression for the load settlement
ratio is
2 2  tanh   L  L

Pt

1      L d
(6)
G  dwt 8 tanh   L  L
1
 1     L d
where, summarising the various dimensionless parameters:

 = db/d (ratio of underream for underreamed piles)
 = GL/Gb (ratio of end-bearing for end-bearing piles)
 = G /GL (variation of soil modulus with depth)
 = Ep/GL (pile-soil stiffness ratio)
 = n(2rm/d) (measure of radius of influence of pile)
L = 2 2 /  (L/d) (measure of pile compressibility).
It should be noted that Ep is the Young's modulus of a solid pile with equivalent cross-sectional
rigidity to the actual pile. Thus Ep = 4(EA)p /(d2), where (EA)p is the actual cross-sectional rigidity
of the pile. A suitable expression for the maximum radius of influence, rm, is
rm = {0.25 + [2.5(1 - ) - 0.25]}L (7a)
However, in order to allow for proportionally higher values of rm for piles of low slenderness ratio
(L/d), the above expression is modified to:
rm = 2.5d + {0.25 + [2.5(1 - ) - 0.25]}L (7a)
Figure 3 shows the variation of the load settlement ratio with slenderness ratio L/d for  =  = 1,  =
0.3. It has been found that these values are in reasonably good agreement with those computed using
charts from Poulos and Davis (1980), in spite of the simplifying assumptions adopted in the analytical
solution given above, and making allowance for the possible scope for error when using the various
multiplicative factors taken from the charts in Poulos and Davis. For long compressible piles, the
results from Poulos and Davis, which are based on boundary element analysis, give higher values of
pile stiffness than obtained using equation (6). The higher values may be partly due to relatively
coarse discretisation of the very long piles, leading to numerical inaccuracies. In practice, it is also
necessary to increase the value of rm to allow for the plan area of the group. This is done by adding a
term rg, which has been calibrated to ensure appropriate convergence for very large groups.
4
From Figure 3, it may be seen that there are combinations of slenderness ratio, L/d, and stiffness
ratio, , beyond which the load settlement ratio becomes independent of the pile length. It can be
shown that insignificant load is transmitted to the pile base for such long piles. This limiting
behaviour is the converse of a stiff rigid pile, and corresponds to the case where the pile starts
behaving as if it were infinitely long, with no load reaching the lower region.
The two limits may be quantified. Piles may be taken as essentially rigid where L/d is less than
0.25(Ep/GL)0.5. Equation (6) then reduces to equation (5). At the other extreme, for piles where L/d
is greater than about 1.5(Ep/GL)0.5, tanh(L) approaches unity and equation (6) reduces approximately
(exactly for  = 1) to
Pt
   / 2  (8)
G L dwt
As expected, the load settlement ratio is now independent of the length of the pile (since no load
reaches the lower end). The modulus GL should be interpreted as the soil shear modulus at the bottom
of the active part of the pile, that is, at a depth that corresponds to z/d = 1.5(Ep/GL)05, rather than at
z =L.
An alternative form of equation (6) for homogeneous soils has been presented by Mylonakis &
Gazetas (1998), with the pile head stiffness, K = Pt/wt expressed as
  tanh  L 
K   EA  p  (9)
1   tanh  L 
The parameters  and L represent non-dimensional base stiffness and slenderness ratio for the pile,
expressed as
Pb k
 and L  L (10)
wb  EA  p   EA p
where Pb and wb are respectively the load and displacement at the pile base and (EA)p is the cross-
sectional rigidity of the pile.
3.2 Extension of solution to pile groups
For pile groups, the stiffness of each pile is reduced because of interaction effects. Mylonakis &
Gazetas (1998) have demonstrated that the interaction factor, , (as defined by Poulos, 1968) must
reflect not only the (assumed) logarithmic decay in displacements, but also the reinforcing effect of
the neighbouring pile. This leads to a reduction in the pile head displacement below that calculated
from a logarithmic decay. For piles of the same length and diameter, the interaction factor for a given
spacing, s, may then be expressed as the product of two terms representing the logarithmic decay and
a ‘diffraction factor’,  (Mylonakis & Gazetas, 1998), giving:
5
 n  rm / s  
 
 n  2r / d   (11)
 m 
where the diffraction factor, , is a function of  and L, according to
2L  sinh  2L    2 sinh  2L   2L   2 cosh  2L   1

 (12)
2sinh  2L   2 2 sinh  2L   4 cosh  2L 
Randolph (2003) has also extended this to deal with piles of different diameters.
The analysis presented by Mylonakis & Gazetas (1998) is for piles of identical embedded length.
However, the general approach may also be used for piles of different length, provided interaction is
first calculated for the effect on the shorter pile of loading the longer pile. The reverse interaction,
where the shorter pile is loaded and causes displacement of the longer pile, may then be evaluated
using the reciprocal theorem, to ensure a symmetric stiffness matrix. This extension for piles of
different embedded lengths is only approximate, and results for groups where the pile lengths differ
by more than a factor of about 4 should be viewed with circumspection.
The above approach has been used to evaluate the stiffness of square groups of piles, from 2 x 2 up
to 30 x 30, for L/d = 25, Ep/GL = 1000, = 0.75 and  = 0.3. The results are shown in Figure 4a
where the pile group stiffness, Kg (ratio of total applied load to average settlement) has been
normalized by GLB, where B is the width of the pile group. Plotting the normalized stiffness against
the normalized width, B/L, leads to an envelope of curves that tends to the stiffness of a surface raft
as B/L becomes large. The stiffness envelope may also be matched closely by using an equivalent
pier approximation of the pile group (Poulos & Davis, 1980; Randolph, 1994), demonstrating the
robustness of calculations of pile group stiffness even with quite approximate models (Figure 4b).
4 RESPONSE OF PILES TO TORSIONAL LOADING
The next type of loading to be considered is that of torsion about the pile axis. An analytical solution
for the torsional response of piles has been presented by Randolph (1981b). Development of the
solution follows the same lines as for the case of axially loaded piles, with the load transfer down the
pile shaft being considered separately from that at the pile base.
At the pile base, the torque, T, may be related to the angle of twist, , using the established solution
for the torsion of a rigid punch:
Tb 2
 (13)
G b d b3  b 3
Down the pile shaft, it may be shown that the angle of twist is related to the interfacial shear stress,
0, by (Randolph, 1981b)
0
 (14)
2G
6
For rigid piles, the above two equations may be combined to give an overall torsional stiffness of
Tt 2 3 L
   (15)
G L d t
3
3 d
where the parameters are as defined previously for axial loading.

In practice, few piles will behave as rigid piles under torsional loading. Usually, deformations
induced by torsion reduce to negligible magnitude at some level down the pile shaft. The situation
is then similar to that for most laterally loaded piles, with the pile length no longer affecting the
performance of the pile. For piles of intermediate length, the torsional stiffness may be written
23 tanh  L  L
 
Tt 3 L do
 (16)
GL d t
3
64 3GL tanh  L  L
1
3 G p L d
where Gp is the shear modulus of a solid pile of the same torsional rigidity as the actual pile. The
remaining parameters are the same as in equation (6), except that the quantity L is now given by
L = 4 2GL / Gp  L / d  . The similarity of the above expression with that for axially loaded piles
(equation (6)) is evident.
The torsional stiffness Tt/(GLd3t) for homogeneous soil conditions is plotted against the stiffness
ratio Gp/G for various pile slenderness ratios, L/d in Figure 5. The transition from flexible behaviour
(where the pile length does not affect the stiffness), for L/d  0.5(Gp/G)0.5, to rigid behaviour for
L/d  0.063(Gp/G)0.5, may be clearly seen. The limiting form of equation (16) for long piles is
Tt
 0.125 2G p / GL (17)
GL d 3t
where GL is interpreted as the shear modulus at a depth of z = 0.5d(Gp/ GL)0.5.

In applying these solutions to the torsional response of piles within a group, two results noted by
Poulos (1975) are of benefit. Firstly, he showed from a series of model tests, that values of shear
modulus for the soil, deduced from axial load tests, gave good predictions of the response of a pile
under torsional loading. Thus, in choosing soil properties as input to PIGLET, the same shear
modulus profile may be adopted for both axial and torsional response of the pile group.
The second observation made by Poulos (1975) was that there was no evidence of an interaction
effect between neighbouring piles under torsional loading. This finding conforms with what might
be anticipated intuitively, and enables the torsional response of piles within a group to be estimated
directly from the equations given above, with no additional factors to allow for effects of interaction.
7
5 RESPONSE OF PILES TO LATERAL LOADING
5.1 Deformation of single laterally loaded piles
The analysis of laterally loaded piles is much more complex than that for axially or torsionally loaded
piles. Even for soil idealised as an elastic continuum, no simple closed form solution is forthcoming.
The solution that has been adopted in the program is one developed by Randolph (1981a) by curve
fitting the results of finite element analyses of laterally loaded piles embedded in elastic ‘soil’. It
was found that, for piles that behave flexibly under lateral load, simple power law relationships could
be developed giving the lateral deflection, u, and the rotation, , of the pile at the soil surface, in
terms of the pile stiffness and the soil properties. The relationships are similar in form to those arising
from considering the soil as a Winkler material characterised by a coefficient of subgrade reaction
(e.g. Reese and Matlock, 1956; Matlock and Reese, 1960). As in the latter type of analysis, the
concept of a ‘critical’ length of pile is used, this depth being the depth to which the pile deforms
appreciably. The term ‘flexible’ is taken to refer to piles where the load deformation characteristics
would not be altered by increasing the length of the pile. Thus piles that are longer than their critical
length behave as ‘flexible’ piles. The large majority of piles used in practice fall into this category.
Since the solution is, by its nature, approximate, a further simplification has been introduced
concerning the soil properties - the shear modulus, G, and Poisson's ratio, . Randolph (1977)
showed that the effect of Poisson's ratio could be allowed for to sufficient accuracy by considering a
single elastic property given by
G* = G(1 + 3/4) (18)
The solution detailed below is in terms of the single parameter G* rather than the true elastic
parameters G and .
The critical length of the pile is determined as
Lc  d  E p / G c 
2/7
(19)
where Ep is the equivalent Young's modulus of the pile, given by
Ep = (EI)p/(d4/64) (20)
(EI)p being the bending rigidity of the pile. The quantity Gc in equation (19) is the value of G* at a
depth of half the critical pile length. For a soil idealised as an elastic material, with a stiffness varying
linearly with depth as
G = G0 + mz (21)
the parameter Gc is given by
8
Gc = G0* + 0.5m*Lc = (1 + 3/4)( G0 + 0.5mLc) (22)
The evaluation of the critical length from equations (19) and (22) requires some iteration except in
the extreme cases of a homogeneous soil (where Gc = G0*) or a soil where the modulus is proportional
to depth (G0* = 0, then Lc = d(2Ep/m*d)2/9 ).
For piles that are longer than their critical length, the lateral deflection, u, and rotation, , at the soil
surface may be evaluated as
E / Gc 
1/ 7
p
 H M 
u  0.27  0.30 
 c Gc   Lc / 2   Lc / 2  
2
(23)
E / Gc 
1/ 7
p
 H M 
  0.30  0.80  
 Lc / 2   Lc / 2  
c
 c Gc 
2 3
where H and M are, respectively, the lateral load and bending moment acting at the soil surface. The
factor c gives the degree of homogeneity for the soil in a similar manner to the factor  in the analysis
of axially loaded piles. It is conveniently defined as the ratio of the value of G* at a depth of Lc/4 to
the value of G* at a depth of Lc/2 (see Figure 6). Thus
Go*  m* Lc / 4 Go*  m* Lc / 4
c   (24)
Go*  m* Lc / 2 Gc
It should be noted that c varies from unity for a homogeneous soil down to 0.5 for a soil where the
stiffness is proportional to depth. In equations (23), the product cGc is merely the value of G* at a
depth of Lc/4. Thus for piles of a given critical length (i.e. stiffness ratio, Ep/Gc), the deformation
under given loading conditions is inversely proportional to the soil stiffness at a depth of one quarter
of the active, or critical, length of pile.
Comparison of results calculated from equations (23) with existing solutions obtained by boundary
element analyses shows good agreement over a wide range of pile-soil stiffness ratios. Detailed
results from such comparisons have been reported by Randolph (1981a).
5.2 Interaction between laterally loaded piles
The complexity of the displacement field around a laterally loaded pile precludes a similar treatment
of the interaction between laterally loaded piles as was possible for axially loaded piles. However,
for piles that are loaded laterally with the pile head restrained against rotation (so-called fixed head
or socketed piles), Randolph (1981a) has shown that the interaction factors, f, may be estimated to
sufficient accuracy from the expression
  1  cos2   ds
1/7
 f  0.3c E p / Gc (25)
where s is the spacing between the axes of the piles and  is the angle that the direction of loading
9
makes to a line passing through the pile axes (see Figure 7).
The same form of expression may be used for interaction of deflection between two free head piles
subjected to force loading (zero moment at the soil surface). In that case, it is found that the
coefficient 0.6 in equation (25) should be replaced by 0.4 to give a reasonable fit to factors computed
by Poulos’ program DEFPIG (Poulos, 1980). In addition, at very close spacings, the 1/s variation of
 can lead to unrealistically high interaction factors. In order to avoid this, and to allow to tend to
unity as s tends to zero, the hyperbolic variation of  is replaced by a parabolic variation wherever 
is calculated to be greater than 1/3. To summarise, the interaction factor uH, giving the increase in
deflection for free head piles subjected to lateral load H, is calculated from
  1 cos2  ds
1/7
  0.2c E p / Gc (26)
where
 uH   for   0.333
and
2
 uH  1  for  > 0.333
27 
Randolph (1981a) has compared values of uH calculated from these expressions with values obtained
from Poulos' program DEFPIG.
The other interaction factors, uM (deflection due to moment loading), H (rotation due to force
loading) and  (rotation due to moment loading) may be estimated to sufficient accuracy by
uM = H  uH2

and (27)
M  uH3
Poulos and Randolph (1983) have compared tabulated values of interaction factors obtained from
this approach and from the boundary element program DEFPIG. In general, the agreement is
reasonably good, with a tendency for interaction factors given by the present approach to decay more
rapidly with increasing pile spacing, than shown by the DEFPIG results.
6 ANALYSIS OF PILE GROUP
The separate solutions for axial, torsional and lateral response of piles must be combined in order to
analyse a pile group under general loading conditions. The program caters for important practical
features of pile groups - namely the presence of raking piles, the possibility of a free-standing length
of pile between pile cap and bearing strata, and also redistribution of load away from highly loaded
piles due to non-linear response.
10
6.1 Treatment of raking piles
The main reasons for using raking piles instead of vertical piles are:
(a) to transfer a portion of the horizontal load at the pile cap into axial load down the pile;
(b) to increase the average spacing between piles, thus transferring the load from the foundation
over a greater volume of soil and, in effect, decreasing the amount of interaction between
neighbouring piles.
The treatment of the first of these effects in the analysis is straightforward. The solutions outlined in
the previous sections are used to calculate the stiffness matrices in terms of local pile axes (i.e. in
terms of axial, torsional and lateral loads and deflections). When the overall group stiffness matrix
is formed, the coordinate axes of each pile are transformed to global axes (vertical and horizontal).
It should be noted that the bending moments induced in the piles by horizontal loading are relatively
sensitive to the angle of rake of the piles. The use of raking piles instead of vertical piles for a
particular foundation may well enable economies to be made in the choice of pile section. To balance
this benefit of raking piles, the difficulties (and possible inaccuracies in positioning) in installing
such piles must be borne in mind, as must the danger of using raking piles in circumstances where
large vertical movements of piles or soil are possible. Fleming et al (1985) discuss this point in more
detail.
Some discussion concerning reason (b) above is appropriate. Raking piles may be used to spread the
foundation load over a greater volume of soil. The ability of the program to cope with piles raking
in any direction (rather than in any one particular vertical plane) is an important one, since it enables
the true spacing between pile centres to be calculated at a given depth. Consider, for example, a
square 2 x 2 group of piles with L/d = 20 and a pile spacing at ground level of s = 3d. If the piles
rake diagonally outwards at 1 in 8, the true average spacing down the shafts of the piles at adjacent
corners is 5.5d. If the analysis is restricted to piles raking in one direction only, the spacing between
adjacent corner piles normal to this direction would remain at the surface spacing of 3d.
It is also necessary to consider the mode in which interaction is assumed to take place. For two piles
that rake away from each other (see Figure 8), axial and lateral loading on pile A may be assumed to
cause interactive displacement of pile B in mode (i) (Figure 8b - where the induced movements are
parallel and normal to pile A) or in mode (ii) (Figure 8c - where the induced movements are parallel
and normal to pile B). Poulos (1979) has discussed the merits of either choice in the analysis of pile
groups with raking piles. He points out that the assumption of interaction in mode (ii) conforms with
the reciprocal theorem of Betti, while that in mode (i) does not. Clearly both modes are idealisations
of the real situation. However, in order to satisfy the reciprocal theorem, the second mode has been
adopted in the present analysis. Adoption of this mode of interaction between piles has the additional
advantage of enabling the axial load deformation behaviour of the piles to be considered
independently from the lateral load deformation behaviour, before combining the two to obtain the
overall deformation characteristics of the foundation.
6.2 Allowance for free-standing length of piles
In many situations, the soil immediately below the pile cap may be relatively soft and should be
ignored in the analysis of the load deformation characteristics of the pile group. In effect the pile cap
11
is considered suspended above the top of the soil strata in which the piles are founded (see Figure 9).
The resulting free-standing length of pile must be taken into account. This is achieved by modifying
the axial, torsional and lateral flexibility matrices of the piles (which relate deformations and loads
at the top of the bearing stratum), treating the free-standing section of pile as a simple cantilever.
New flexibility matrices are formed relating the deformations and loads at the underside of the pile
cap before combining these to give the required load deformation characteristics of the complete
group. To allow for the situation where the upper part of a pile is cased as it passes through softer
soil, it is possible in the program to specify different pile properties in the free-standing section than
in the main part of the pile.
7 NON-LINEAR PILE RESPONSE
PIGLET is primarily limited to elastic response of the pile-soil system, and hence it is generally
inappropriate to design individual piles to withstand the high loads predicted at the edges and corners
of the pile group.
7.1 Elastic-plastic analysis (E-P)
In order to assess what maximum axial load the piles should be designed for, the program allows the
user to specify a limiting axial load (or ‘pile capacity’) for each pile within the group. The ratio of
capacities for piles loaded in tension or compression is also input, with the program choosing the
appropriate capacity depending on the sign of the axial force. The program then performs an
incremental elastic analysis, eliminating any pile within the group that reaches the specified limiting
load. This results in a non-linear group response with redistribution of any additional loads away
from ‘failed’ piles, but does not change the underlying ‘elastic’ nature of the pile responses prior to
failure.
7.2 Non-linear axial and lateral response (NL-P)

The most recent version of PIGLET allows individual pile responses to be non-linear in axial and
lateral deformation modes. For the axial response, the initial pile head stiffness kv,0 is modified using
a generalised hyperbolic expression relating the current secant stiffness k,v to the axial load P
normalised by the imposed limiting axial capacity Plim:
g
kv  P 
 1 f   (28)
kv ,0  Plim 
Two parameters f and g allow a wide range of non-linear forms to be represented, with the reduction
in secant stiffness given by 1 – f and the parameter g changing the curvature, as shown in Figure 10.
Note, the torsional response is also softened to maintain consistency with the axial response.
Under lateral loading, it is often not really possible to identify a (geotechnical, as opposed to
structural) capacity and so it is more convenient to define the reduction in stiffness in terms of the
elastic pile head displacement u0,el normalised by the value at which the secant stiffness reduces by
50%, u0,50. A suitable expression is
12
kh 1
 p
k h ,0 u  (29)
1   0,el 
 u 0,50 
As shown in Figure 11, high values of the power p will provide an almost elastic, perfectly plastic
response, with a limiting horizontal load of Hult ~ kh,0u50. By contrast, lower values will give gradually
decreasing degrees of non-linearity.
In the previous version (PIGLET6.1a), the reduction in stiffness was applied to all flexibility terms
relating displacements u and rotations  to the ground level loads H and moments M (see equation
(23)). However, this leads to underprediction of moments for either a free-head or fixed head pile. In
the current version (PIGLET6.2), this has been corrected by applying that factor to the u:H flexibility,
but using (kh/kh,o)2/3 for the u:M and :H terms and (kh/kh,o)1/3 for the :M term and also for the
critical length Lc. The powers 2/3 and 1/3 follow naturally from the form of equation (23).
Following the principle proposed by Caputo and Viggiani (1984) (see also Randolph 1994),
interaction between piles is still based on elastic conditions. That is, the displacement of a
neighbouring pile is assumed proportional to the elastic displacement of the loaded pile.
The non-linear relationships given in Equations (28) and (29) are phenomenological in nature and
the non-linear parameters need to be assessed in the light of load test data, or alternatively based on
experience or more sophisticated numerical analysis of the single pile response, for example using
load transfer analysis software such as RATZ (Randolph, 2003), LAP (Doherty, 2018) or LPILE
(Ensoft, 2016). An example is shown later for the case of a laterally loaded 9-pile group.
8 EXAMPLE APPLICATION
8.1 Example 1: model tests of Davisson and Salley (1970)
As an example of the application of PIGLET, model tests on a group of six piles embedded in sand
(Davisson and Salley, 1970) have been analysed.
A group of six tubular aluminium piles of external diameter 12.7 mm (0.5 in) and wall thickness 0.8
mm (0.03 in), embedded in a tank of dry fine sand to a depth of 0.533 m (21 in), were loaded through
a pile cap suspended just above the level of the sand surface. Figure 12 shows the pile layout and
applied loads. Equivalent Young's modulus for the piles may be calculated as
Axial loading: Ep = Eal[1 - (di/d)2] = 16,300 MPa (2.37 x 106 psi)
Lateral loading: Ep = Eal[1 - (di/d)4] = 28,900 MPa (4.19 x 106 psi).
Each of the six piles in the group was load tested axially, prior to forming the pile cap, in order to
determine the axial stiffness. Davisson and Salley (1970) report an average stiffness of 0.82 kN/mm
(4860 lbf/in) with a standard deviation of 0.15 kN/mm (840 lbf/in). It is reasonable to assume that
the shear modulus of the sand is proportional to the effective stress level (and thus to depth below
the sand surface). With this assumption, and adopting a value for Poisson's ratio of 0.25, equation
13
(6) may be used to deduce the shear modulus profile necessary to yield the above value of axial
stiffness for the piles. This process leads to an expression for the shear modulus, G, of
G = 4.2z MPa (15.3z psi) [z in metres (in)].
The above variation of shear modulus has been used to analyse the complete group of piles under the
loading shown in Figure 12.
Table 1 summarises measured values of axial load, lateral load and bending moment at the tops of
the six piles. These results compare favourably with those obtained from the program PIGLET, the
error in the predicted bending moments and in the largest axial loads being generally less than 10%.
The computed lateral deflection of 0.29 mm (0.011 in) is some 1.2 standard deviations larger than
the measured deflection of 0.23 mm (0.09 in).
Also shown in Table 1 are loads obtained from the program PGROUP (the values are taken from the
PGROUP users' manual). Although predictions of axial and lateral loads are good, there is
considerable discrepancy in the values of bending moments. Much of this discrepancy may be
attributed to the assumption of a homogeneous value of shear modulus for the soil in the PGROUP
analysis. This assumption is likely to be a less good approximation for the sand than taking a shear
modulus that is proportional to depth, and in this case leads to significant under-prediction of the
bending moments induced in the piles.
Table 1 Comparison with Model Test Results of Davisson and Salley (1970)
LOADS AT HEAD OF EACH PILE
Pile No. Axial Load (N) Lateral Shear Load (N) Bending Moment (Nm)
Meas'd PIGLET PGROUP Meas'd PIGLET PGROUP Meas'd PIGLET PGROUP
1 80.1 73.4 69.0 22.2 14.2 17.4 1.23 1.13 0.79
2 56.9 73.4 69.0 23.1 14.2 17.4 1.27 1.13 0.79
3 28.9 34.2 31.6 16.9 14.7 16.0 1.20 1.17 0.77
4 24.0 34.2 31.6 16.0 14.7 16.0 1.18 1.17 0.77
5 12.2 5.9 14.0 23.1 19.9 20.5 1.49 1.53 0.96
6 13.6 5.9 14.0 16.9 19.9 20.5 1.22 1.53 0.96
8.2 Example 2: field-scale lateral load tests of small pile group Robbins et al.
An example of non-linear response obtained using PIGLET is based on the field tests reported by
Rollins et al. (2006), for lateral loading of a nine pile group (see Figure 13). The 0.324 m diameter
steel pipe piles have equivalent Young’s modulus (for a solid pile) of about 23,000 MPa for axial
14
loading, and 43,000 MPa for lateral loading. The piles were loaded at a height of 0.44 m above
ground level, using a grid system that gave effectively pinned conditions for the pile heads at that
level. The soil conditions are reported by Rollins et al. (1998) as comprising interbedded desiccated
clayey silt and sand layers, with shear wave velocities of 120 to 150 m/s (so small strain shear
modulus of 26 to 40 MPa assuming a unit density of around 1.8 t/m3).
As discussed by Randolph and Reul (2019), the single pile response has been matched using a
uniform shear modulus of 10 MPa (Poisson’s ratio of 0.3), and non-linear soil parameters of
u0,50 = 0.02d = 6.5 mm and a power of p = 0.8 (see Equation (29)), which gives an excellent fit to the
measured response. These parameters also allow a reasonable fit to the nine pile group response, as
shown in Figure 13. At loads above about 80 kN, where the pile head deflection reaches 10% of the
pile diameter (but deflections at ground level about half that, so 5% of the diameter), the measured
group response starts to soften relative to the PIGLET curve. This is probably due to increasing
effects of ‘shadowing’, with the lateral resistance of any trailing pile lower than that of leading piles.
Such effects were noted by Rollins et al. (1998), with typically each front row pile carrying ~14% of
the load, compared with 10% for each middle row pile and 9% for each trailing edge pile. By contrast,
the loads estimated using Piglet are symmetric (leading and trailing), with the corner piles carrying
the highest loads. However, the range is narrow, with each pile carrying between 90 and 104 % of
the average applied load, compared with a range of 67 to 115 % for elastic conditions.
15
PART B: PROGRAM DOCUMENTATION
9 STRUCTURE OF PROGRAM
The program PIGLET has been structured so that the complexity, or ‘scope’, of analysis may be
chosen by the user. The user may choose between three alternatives:
(1) analysis for vertical loading only (piles are assumed to be vertical);
(2) analysis for vertical and horizontal loading in one plane only (piles are assumed to be raked
only in the plane of loading);
(3) full analysis of pile group under vertical, horizontal and torsional loading (piles may be raked
in any direction).
The advantages of this choice are that the amount of input data, computer effort and output are all
determined by the scope of the analysis, being a minimum for (1) and a maximum for (3). In addition,
for vertical loading only, the program allows specification of a fully flexible cap, as opposed to the
rigid cap assumed in the other options. As noted below, the latter two alternatives allow a non-linear
response to be computed, by specifying limiting axial loads for each pile.
Data for the program is input on the ‘Data’ worksheet and the program is run by means either of the
button on the spreadsheet, or by typing ‘Ctrl r’. A ‘Checksum’ comprising the sum of all input data
items is output on the ‘Loadcases’ worksheet, and that value is checked before the program is run, to
ensure that the data have changed since the previous analysis. A number of different loading cases
may be considered in any given analysis. A flow chart for the program is shown in Figure 14. The
only major branch point occurs for the case of vertical loading only (NSCOPE = 1), where the extra
option of a fully flexible pile cap entails a different approach than for a rigid pile cap.
As supplied, the maximum problem size is set to 1000 piles. Alternative versions can be supplied
that can analyse larger groups, at the cost of greater memory requirements. The program runs from
an Excel workbook, where a macro calls a Fortran dynamic link library (DLL) in order to perform
the analysis. Both 32-bit and 64-bit DLLs are provided, with the latter relevant for modern versions
of Excel, such as Office 365. In order for Excel to find the appropriate DLL (avoiding a “File not
found” error appearing), it is most convenient if a copy of the DLL is stored in the same directory as
the workbook. After opening the workbook, the file may be saved back to the working directory
(either over-writing itself or using a new filename), which directs Excel to that directory.
Alternatively, the input data sheet now includes an optional path for the directory where the DLL
resides. Other strategies are described in the ‘Readme’ sheet of the workbook.
10 PROGRAM INPUT
10.1 Data input and editing
Data input is on the worksheet ‘Data’ and is confined to the areas with yellow background. Standard
Excel techniques, including formulae (for example, to give the coordinates of piles in large groups)
may be used. It is important within the main pile geometry and loading array areas not to leave any
gaps between successive entries.
16
The description of each item of data is intended to be self-explanatory. However, additional notes
are provided on the following pages to guide new users of the program. As in any engineering
problem, a consistent set of units must be used in the input data. No system of units is assumed by
the program. Essentially, the user must decide what unit of force (F) and what unit of length (L) are
to be adopted. Data are then input in appropriate units according to the type of data. Thus values of
modulus should be in units of F/L2, bending moments in FL, and so forth.
10.2 Data Items
Problem Title and Scope
The title may comprise any sequence of alphanumeric characters or punctuation marks. The ‘scope’
of the problem has been discussed in Section 8 of the manual. Essentially it defines the complexity
of the applied loading, with options of (1) vertical loading only, (2) vertical and horizontal loading
in one plane, or (3) full three-dimensional loading including torsion.
Pile Parameters
The maximum number of piles that the program can analyse is set to 1000 in the standard version.
The overall length of each pile is the sum of the embedded portion and a free-standing length (which
may be zero). It is assumed that the free-standing length is identical for all the piles, but the embedded
length is specified in the block for data for the pile group geometry and may vary from pile to pile.
It should be noted that evaluation of interaction effects for piles of different embedded lengths is only
approximate, and results for groups where the pile lengths differ by more than a factor of about 4
may lose accuracy. The Young’s modulus of the pile is that of a pile having equivalent
cross-sectional rigidity (for axial loading) or bending rigidity (for lateral loading) as the real pile.
Thus, for a pile of diameter d, the value of Young’s modulus for axial loading is
 EA p
Ep  4 (30)
d 2
while for lateral loading,
 EI  p
E p  64 (31)
d 4
In order to allow for the possibility of a change in pile cross-section at ground level, different values
of Young’s modulus may be specified for the free-standing lengths of pile. For torsional loading, the
torsional rigidity of the pile is obtained from the bending rigidity, taking Poisson’s ratio for the pile
material as 0.3.
For non-circular piles, it is important that the diameter of the idealised pile is chosen realistically. It
is suggested that the cross-sectional area of the idealised pile should be chosen so as to equal the
gross (enclosed) area of the actual pile. For H section piles, the gross area should be taken as that of
the encompassing rectangle.
17
For lateral loading, there is a choice between whether the piles are to be assumed fixed into the pile
cap or pinned to the pile cap (zero moment at pile cap level).
Note that, as described in Section 5.1, the kernel solution for lateral loading in PIGLET is only
applicable to piles that are longer than (at least 80 % of) their critical length. A warning
message will appear at the top of the output spreadsheets if that condition is violated.
Soil Parameters
The value of Poisson’s ratio for the soil is assumed the same for all types of loading - axial, lateral
or torsional. Different profiles of shear modulus may be specified for axial and for lateral loading
(the profile for torsional loading is assumed the same as for axial loading). As discussed in Section 2,
the shear modulus profile is assumed to increase linearly with depth. The user specifies the value at
the ground surface (which must be non-negative) and the gradient with depth (also non-negative). In
addition, for vertical loading a sudden increase in modulus at the base of the pile (for end-bearing
piles) may be input. If this value is set to less than the value that would be calculated from the linear
variation of shear modulus, then the program corrects it to that value (thus the program does not
permit any decrease in the value of shear modulus at the pile base).
For irregular soil profiles, it is important that the linear variation of soil modulus with depth is chosen
so as to reflect the true average shear modulus over the depth of penetration of the piles, and also the
trend of variation of soil modulus with depth. Since piles deflect under lateral loading only in the
upper ten diameters or so, it is possible to specify different values of soil modulus for lateral loading
than for axial (and torsional) loading.
In many instances, piles are installed so that they finish at some depth above a significantly stiffer
stratum of soil. While such piles are not strictly 'end-bearing' piles, the stiffer stratum of soil will
reduce the overall settlement of the group. For a stratum with shear modulus Gh, at a depth h (greater
than the pile length L) it is recommended that the value of shear modulus below the pile bases, Gb, is
chosen by means of the expression (Lee, 1991)
1 1  Gz  L   1  e1 h / L 
  1     (32)
Gb Gh  Gh   Gz  L 

For values of h greater than 4L, the presence of the stiffer stratum of soil may be ignored.
For situations where no values of shear modulus are available for the soil, values of G must be chosen
by inspection of the available soil data. For cohesive soil, it is common practice to correlate shear
modulus with the shear strength su. At working load levels, the axial deformation of piles may be
estimated reasonably well by taking shear modulus values in the range
200  G/su  400
Under lateral loading, the high strains that occur in the soil close to the pile give rise to lower secant
modulus. It is suggested that G should be chosen in the range
18
100  G/su  200
for the lateral load deformation behaviour of the pile group.

For non-cohesive soil, or where the only data available are results of standard penetration tests, it is
suggested that the simple (but conservative) guideline of G = N MPa be adopted (see Randolph,
1981b). A less conservative correlation has been proposed by Wroth et al (1979), who suggest
G/pa  40N0.77 (28)
where pa is atmospheric pressure (100 kPa). In general, the variation of shear modulus with depth in
sand (below the water table) may be expressed as G = mz, with m in the range 1 MPa/m (loose virgin
sand) up to 10 MPa/m (dense sand).
In soft rocks, the effects of pile installation must be allowed for. While the in situ modulus of soft
rocks such as chalk can be extremely high, installation of bored or driven piles tends to break up the
block structure of the rock. The relevant shear modulus is then that associated with large strains (see
Wakeling, 1970; Randolph and Wroth, 1978b).
Further guidance on the choice of shear modulus may be found in Wroth et al (1979). For pile groups
under predominantly vertical load, Mandolini and Viggiani (1999) showed that adopting shear
modulus values close to the small strain (seismic) shear modulus gave a reasonable estimate of
measured settlements for large pile groups.
Non-linear soil parameters and number of loading steps
The non-linear pile response parameters, f and g for axial loading, and (u0/d)50 and p for lateral
loading, together with the required number of loading increments for the non-linear analysis, are
input to the right of the elastic soil parameters (Excel columns L & M, rows 12, 14 and 15). For
convenience, the resulting axial and lateral loading responses for the parameters input are shown in
plots to the right of the data input.
Values of the parameters are limited as follows:
f 0 (implying linear response) to 0.999;
g > 0, but the software subsequently forces g ≥ 0.1;
(u0/d)50 ≥ 0.001, but an input value of zero switches off lateral non-linear pile response;
p ≥ 0, but values less than 0.1 not recommended; values greater than 10 give essentially
elastic-plastic response.
The number of loading steps (i.e. increments) is limited to the range 1 to 50. Generally a number
from 10 to 20 gives acceptable accuracy.
It must be emphasised that non-linear pile response is only analysed if a given loading type is set to
a negative quantity. Once that has been set, non-linear axial response requires that (a) f is greater than
1.e-6, and (b) the particular pile has been assigned a positive limiting value. Correspondingly, for
non-linear lateral response (of all piles in the group), the value of (u0/d)50 must be specified greater
than 1.e-6.
19
Load Cases
Up to 100 separate load cases may be specified for each analysis. Loading may be specified explicitly
(as forces and moments) or may be given as imposed deformations of the pile cap. The pile cap is
assumed rigid accept for the case of vertical loading only, when arbitrary loads or deflections may
be specified at the head of each pile.
The user is asked to specify a switch, for each load case, which identifies the loading type. For
vertical loading only (Scope = 1), the switch values are:
Loading type 1: loads applied to a rigid pile cap
Loading type 2: deflections imposed on rigid pile cap
Loading type 3: loads applied to individual piles (fully flexible pile cap)
Loading type 4: deflections imposed on individual piles (fully flexible pile cap).
For the case of a fully flexible pile cap (loading types 3 and 4), the individual pile loads or deflections
are specified for the corresponding load case in the Pile Group data block (see later).
The switches may be set to corresponding negative values (–1, –2, –3 or –4 respectively) to indicate
a non-linear analysis, with limiting axial loads specified for each such load case in the Pile Group
data block (see later).
For laterally loaded pile groups (Scope = 2 or 3), the corresponding switch values are:
Loading type 1: loads applied to a rigid pile cap
Loading type 2: deformations imposed on rigid pile cap
Loading type 3: vertical, horizontal and torsional loads applied to the rigid pile cap, but with zero
rotation of the cap permitted (so-called ‘fixed-head’ condition); the fixing moments
to ensure zero rotation are calculated by the program.
Loading type 4: vertical and horizontal deflections and twist applied to the rigid pile cap, but with
zero rotation of the cap permitted (so-called ‘fixed-head’ condition); the fixing
moments to ensure zero rotation of the pile cap, and associated rotation of each pile
head (consistent with the specified fixity condition) are calculated by the program.
These switches may be set to –1, –2, –3 or –4 respectively, to indicate a non-linear analysis, with
limiting axial loads specified for each such load case in the Pile Group data block (see later).
Where a rigid pile cap is specified (or assumed for problems involving horizontal loading), the total
loads (or deflections) acting on the pile cap are specified, with all loads assumed to act at pile cap
level (z = 0), through the origin (x = y = 0). Horizontal loads are taken as positive in the direction of
the positive x and y axes, and moments are taken as positive in the sense of rotating the x axis towards
the z axis (for loading in the x:z plane) and rotating the y axis towards the z axis (for loading in the
y:z plane). The sign convention in PIGLET differs from the usual right-handed axis rule, as shown
in Figure 15.
The software automatically detects the number of load cases by searching for the first row in the
Loading Cases block where the loading type (Excel column I) is blank or set to zero. A check is also
made that each specified load case has non-zero loads or deformations specified, otherwise it again
curtails the analysis at the preceding loading case.
20
Pile Group Geometry
For each pile, values of shaft diameter, base diameter, embedded length and (x, y) co-ordinates must
be input. In addition, where lateral loading is involved, angles of rake must be specified in radians,
either in the x:z plane (where loading is restricted to one plane only), or in both x:z and y:z planes.
The program assumes a right-handed set of coordinate axes (x, y, z), with the z axis pointing vertically
downwards. Angles of rake should be input in radians measured from the z axis, positive values
indicating a pile lying between the x and z (or y and z) axes. Figure 16 shows this sign convention.
The maximum angle of rake that is permitted is 1 radian.
For convenience, if pile groups of differing numbers of piles are to be analysed, including the
response of a single pile, then setting the pile diameter (in Excel sheet column C) to zero in the row
immediately following the pile(s) of interest will automatically limit the analysis to the intended
number of piles.
Profiles of Bending Moment, Shear and Lateral Deflection
For analyses that involve lateral loading, profiles of bending moments and lateral deflection relative
to the immediately surrounding soil may be output for specified piles. A switch is specified for each
pile as to whether (a) no profiles (switch zero or blank), (b) profiles of bending moment and shear
only (switch = 1), or (c) profiles of bending moment and lateral deflection (switch = 2), are required.
For three-dimensional loading, separate choices are given for the x:z plane and the y:z plane.
It should be emphasised that, since the free field soil deflections (due to interaction between piles)
are not included in the relative lateral deflection profile, the deflection output for the pile head will
not correspond with the total lateral deflection for the pile head (except for analyses with only one
pile in the group). The profiles extend down to the critical depth for the pile, below which
deformations and bending moments die out rapidly.
It should be noted that, since PIGLET is based on solutions for the pile head response, the
profiles of bending moment, shear and lateral deflection down the pile are indicative in nature
and should not be taken as having great accuracy than 10 to 15%.
Pile Group Loads (Deflections) or Limiting Loads
For vertical loading through a fully flexible pile cap (Scope = 1, Loading Type = 3 or 4), the loads
or deflections applied to each pile are specified for the corresponding load case to the right of the pile
group geometry, in the column for the corresponding load case number. For the case of a non-linear
analysis and a rigid pile cap, the limiting axial loads for each pile are specified for the corresponding
load case to the right of the pile group geometry. However, for non-linear analysis with a flexible
pile cap (i.e. vertical loads or deflections applied to each pile head) the limiting axial loads are
specified in the column for the x:z pile rake (Excel column H). For any non-linear analysis, specifying
zero limiting load (or a blank entry), will force elastic response of that particular pile.
For problems involving lateral loading (Scope = 2 or 3), and negative loading types, the limiting axial
loads for each pile are specified for the corresponding load case to the right of the pile group
geometry, in the column for the corresponding load case number.
21
11 PROGRAM OUTPUT
Output from the program is provided on three separate worksheets, ‘Outputparameters’, ‘Loadcases’
and ‘Summary’. The first of these provides a brief summary of the problem to be analysed and then
key solutions parameters for the different deformation modes, followed by:
(1) Response of pile group to unit deformations of the pile cap, giving loads and moments at the
head of each pile, in local coordinates.
(2) Overall stiffness and flexibility matrices for the group.
The second output sheet (‘Loadcases’) then gives the response of the pile group to the different load
cases specified by the user. This section includes loads and resulting deformations of the pile cap
(incrementally, for non-linear analysis) and, at the end of all loading increments for that load case,
loads and moments at the head of each pile, and (optionally) profiles of bending moment and lateral
deflection down specified piles.
Finally, where more than a single load case is analysed, summary tables are included on the third
output worksheet (‘Summary’), where the loads and deflections at the head of each pile are
summarised for each load case.
The sign convention for lateral loads and moments for each pile follows that for specifying the
applied loads, with lateral load being taken as positive in the direction of the positive x and y axes,
and moments taken as positive in the sense of rotating the x axis towards the z axis (for loading in
the x:z plane) and rotating the y axis towards the z axis (for loading in the y:z plane).
A text file, piglet.out, is also created giving the input data and the above output. The default directory
for the output file is where the DLL resides, but optionally a path may be specified directly on the
data sheet.
12 REFERENCES
1. Baguelin F., Bustamante M., Frank R. and Jezequel J.F. (1975). La capacite portante des pieux.
Annales de l'Institut Technique du Batiment et des Travaux Publics, Suppl. 330, Serie SF116,
pp 1-22.
2. Banerjee P.K. and Davies T.G. (1978). The behaviour of axially and laterally loaded piles
embedded in non-homogeneous soils. Géotechnique, 28(3), 309-326.
3. Banerjee P.K., Driscoll R.M.C. and Davies T. (1978). Program For The Analysis Of Pile
Groups Of Any Geometry Subjected To Horizontal And Vertical Loads And Moments, Pgroup,
(3.0). HECB/B/7, Department of Transport, HECB, London.
4. Butterfield R. and Douglas R.A. (1981). Flexibility coefficients for the design of piles and pile
groups. CIRIA Technical Note 108.
5. Caputo, V. and Viggiani, C. (1984). Pile foundation analysis: a simple approach to non linearity
effects. Rivista Italiana di Geotecnica, 18(2), 32-51.
6. Cooke R.W. (1974). Settlement of friction pile foundations. Proc. Conf. on Tall Buildings,
22
Kuala Lumpur, 7-19.

7. Davisson M.T. and Salley J.R. (1970). Model study of laterally loaded piles. J. of Soil Mech.
and Found. Engg Div., ASCE, 96(SM5).
8. Doherty, J.P. (2018). LAP – software for lateral pile analysis. www.geocalcs.com/LAP.
9. Ensoft (2016). LPILE –software for lateral pile analysis. Ensoft Inc., Austin, USA.
10. Fleming W.G.K., Weltman A.J., Randolph M.F. and Elson W.K. (1985). Piling Engineering.
Surrey University Press, Glasgow.
11. Frank R. (1974). Etude Theorique Du Comportement Des Pieux Sous Charge Verticale;
Introduction De La Dilatance. Dr-Eng. Thesis, University Paris VI (Pierre et Marie Curie
University).
12. Lee C.Y. (1991). Discrete layer analysis of axially loaded piles and pile groups. Computers and
Geotechnics, 11, 295-313.
13. Mandolini, A. & Viggiani, C. (1997). Settlement of piled foundations. Géotechnique, 47(4),
791-816.
14. Matlock H. and Reese L.C. (1960). Generalised solutions for laterally loaded piles. J. Soil
Mech. and Found. Engng Div., ASCE, 86(SM5).
15. Mylonakis, G. & Gazetas, G. (1998). Settlement and additional internal forces of grouped piles
in layered soil. Géotechnique, 48(1), 55-72.
16. O'Neill M.W., Ghazzaly O.I. and Ha H.B. (1977). Analysis of three-dimensional pile groups
with non-linear soil response and pile-soil pile interaction. Proc. 9th Offshore Technology Conf.,
2, 245-256.
17. Poulos H.G. (1971). Behaviour of laterally loaded piles, I - Single piles, II - Pile groups. J. Soil
Mech and Found. Engng Div., ASCE, 97(SM5).
18. Poulos H.G. (1973). Load-deflection prediction for laterally loaded piles. Australian
Geomechanics Journal, 3(1).
19. Poulos H.G. (1975). Torsional response of piles. J. Geot. Engng Div., ASCE, 101(GT10).
20. Poulos H.G. (1979). An approach for the analysis of offshore pile groups. Proc. Conf. on
Numerical Methods in Offshore Piling, ICE, London, 119-126.
21. Poulos H.G. (1979). Settlement of single piles in non-homogeneous soil. J. Geot. Engng Div.,
ASCE, 105(GT5).
22. Poulos H.G. (1980). Users' Guide To Program DEFPIG - Deformation Analysis Of Pile
Groups. School of Civil Engineering, University of Sydney.
23. Poulos H.G. and Davis E.H. (1980). Pile foundation analysis and design. John Wiley & Sons,
New York.
24. Poulos H.G. and Randolph M.F. (1983). Pile group analysis: a study of two methods. J. of Geot.
23
Eng., ASCE, 109(3), 355-372.

25. Randolph M.F. (1977). A Theoretical Study Of The Performance Of Piles. PhD Thesis,
University of Cambridge.
26. Randolph M.F. (1981). Analysis of the behaviour of piles subjected to torsion. J. of Geot.
Engng Div., ASCE, 107(GT8), 1095-1111.
27. Randolph M.F. (1981). The response of flexible piles to lateral loading. Géotechnique, 31(2),
247-259.
28. Randolph M.F. (1994). Design methods for pile groups and piled rafts. Proc. 13th Int. Conf. on
Soil Mech. and Found. Eng., New Delhi, 5, 61-82.
29. Randolph, M.F. (2003). Science and empiricism in pile foundation design: 43rd Rankine
Lecture, Géotechnique, 53(10), 847-875.
30. Randolph, M.F. (2003). RATZ: Load transfer analysis of axially loaded piles. Users’ Manual,
Version 4 2, Perth.
31. Randolph, M.F. and Reul, O. (2019). K. Rainer Massarsch Lecture: Practical approaches for
design of pile groups and piled rafts. Proc. 4th Bolivian Int. Conf. on Deep Foundations.
32. Randolph M.F. and Wroth C.P. (1978). Analysis of deformation of vertically loaded piles. J.
of the Geot. Eng. Div., ASCE, 104(GT12), 1465-1488.
33. Randolph M.F. and Wroth C.P. (1978). A simple approach to pile design and the analysis of
pile tests. Proc. Symp. on Behaviour of Deep Foundations, ASTM STP 470, 484-499.
34. Randolph M.F. and Wroth C.P. (1979). An analysis of the vertical deformation of pile groups.
Géotechnique, 29(4), 423-439.
35. Reese L.C. and Matlock H. (1956). Non-dimensional solutions for laterally loaded piles. Proc.
8th Texas Conf. on Soil Mech.
36. Rollins, K.M., Olsen, K.G., Jensen, D.H., Garrett, B.H., Olsen, R.J. and Egbert, J.J., 2006. Pile
spacing effects on lateral pile group behaviour: Analysis. J. Geotech. and Geoenv. Eng., ASCE,
132(10), 1272-1283.
37. Rollins, K.M., Peterson, K.T. and Weaver, T.J., 1998. Lateral load behaviour of full-scale pile
group in clay. J. Geotech. and Geoenv. Eng., ASCE, 124(6), 468-478.
38. Wakeling T.R.M. (1970). A comparison of the results of standard site investigation methods
against the results of a detailed geotechnical investigation in Middle Chalk at Mundford,
Norfolk. Proc. Conf. on In Situ Investigations in Soils and Rocks, British Geotechnical Society,
London.
39. Wroth C.P., Randolph M.F., Houlsby G.T. and Fahey M. (1979). A review of the engineering
properties of soils with particular reference to the shear modulus. Cambridge University
Engineering Department Research Report, CUED/D - Soils TR 75.
24
FIGURE TITLES
Figure 1 Assumed variation of soil shear modulus with depth

Figure 2 Uncoupling of effects due to pile shaft and base
Figure 3 Load settlement ratios for compressible piles
Figure 4 Comparison of pile group and equivalent pier stiffnesses
Figure 5 Torsional stiffness factor for piles in homogeneous soil
Figure 6 Notation for analysis of laterally loaded piles
Figure 7 Plan view of two piles subjected to lateral loading
Figure 8 Choice of modes for interaction between pairs of non parallel piles
Figure 9 Allowance for free standing length of piles
Figure 10 Non-linear axial pile response
Figure 11 Non-linear lateral pile response
Figure 12 Model pile test arrangement (Davisson and Salley, 1970)
Figure 13 Layout and lateral load response of nine pile group (Rollins et al. 2006)
Figure 14 Flow chart for PIGLET
Figure 15 Sign convention for loading
Figure 16 Sign convention for pile rake
25
P (w) Shear modulus
Gavg GL Gb
Gavg GL
 
GL Gb
d
d Ep
L L/2  b 
Ep d GL
db  EI  p
Ep 
 d 2 / 4
L
Depth, z
Figure 1 Assumed variation of soil shear modulus with depth
26
Pt
Pt = Ps + Pb
Ps
L Shaft response
A B A B
Pb
A' B'
Base response
Figure 2 Uncoupling of effects due to pile shaft and base
27
30
3000
25
20
1000
P/(Gwd)
15
300
10
100
5 30
 = 10
0
1 10 100
Pile slenderness ratio, L/d
(a)  = 0.5
45
40 3000
35
30
1000
P/(Gwd)
25
20
15 300
10 100
5 30
 = 10
0
1 10 100
(b)  = 0.75
50
45 3000
40
35
30
P/(Gwd)
1000
25
20
15 300
10 100
5 30
 = 10
0
1 10 100
(c)  = 1
Figure 3 Load settlement ratios for compressible piles
28
20
Group
L/d = 25
stiffness 18
Ep/GL = 1000
K g/GLB 16
s/d = 2  = 0.75
14  = 0.3
12
10 s/d = 3
8
s/d = 5
6
4 s/d = 10
2
Raft stiffness
0
0.1 1 10
Normalised width of pile group, B/L
(a) Normalised stiffness of pile groups
Group 20
stiffness 18
K g/GLB 16
14 Stiffness of
incompressible pier
12
10 Equivalent pier
8 (same area and
length as pile group)
6
4 80 % of stiffness of
2 incompressible pier
0
0.1 1 10
Normalised width of pile group, B/L
(b) Comparison of equivalent pier stiffness
Figure 4 Comparison of pile group and equivalent pier stiffnesses
29
Figure 5 Torsional stiffness factor for piles in homogeneous soil
30
H (u0) M (0) cGc Gc Modified shear

modulus, G*
Lc/4
Lc Lc/2 G* = G(1 + 3/4)
d Gc = G* at z = Lc/2
Lc  EI  p
Ep 
Depth, z  d 4 / 64
Critical pile length:
Lc = d[Ep/Gc]2/7
Figure 6 Notation for analysis of laterally loaded piles
31
s

Figure 7 Plan view of two piles subjected to lateral loading
32
uA uA
wA wA
Pile A Pile B
(a) Interactive displacements of Pile B parallel to Pile A
uA
uA
wA
wA
Pile A Pile B
(a) Interactive displacements of Pile B axial and lateral
Figure 8 Choice of modes for interaction between pairs of non parallel piles
33
Pile cap
Depth of free-standing
section of piles
Level of
bearing strata
Piles Penetration of piles into

bearing strata
Figure 9 Allowance for free standing length of piles
34
1.1
f = 0 f = 0.7, g = 0.9 f = 0.9
1
Normalised axial load, P/Plim
0.9
0.8
0.7
0.6 g = 0.3, 0.7, 1.5
0.5
0.4
0.3
0.2 Axial response
0.1 Generalised hyperbola
0
0 2 4 6 8 10 12
Normalised axial displacement, w0kv,0/Plim
Figure 10 Non-linear axial pile response
35
6
Lateral response
Generalised hyperbola
5
Normalised lateral load, H/kh,0u0,50
Linear p = 0.5
4
3 p = 2
2
p = 100
0
0 2 4 6 8 10 12 14 16 18 20
Normalised lateral displacement, (u0/d)/(u0/d)50
Figure 11 Non-linear lateral pile response
36
0.222 kN
0.138 kN
76 mm
fine, dry sand
piles 0.533 m
3 long
127 mm
1 Plan View
12.7 mm
Figure 12 Model pile test arrangement (Davisson and Salley, 1970)
37
0.97 m 0.97 m
0.97 m 0.324 m
0.97 m
Piles:
steel tubes: 0.324 m diam.
wall thickness: 9.5 mm
embedded 9.1 m
(a) Layout of pile group and pile geometry
200 Measured - single pile

180 Piglet output
160
Lateral load, H (kN)
Measured - average group pile

140
Piglet output - group
120
100
80
60
40
20
0
0 10 20 30 40 50 60 70
Lateral displacement, u0 (mm)
(b) Responses of single pile and pile group
Figure 13 Layout and lateral load response of nine pile group (Rollins et al. 2006)
38
INDATA
Start Read (new) data
VSTIF
Axial load-deformation
response of pile group
VERTLD
1 Flexibility and stiffness of
NSCOPE
group, response to load cases
2, 3 (vertical loading only)
HSTIF
Lateral load-deformation
response (x:z plane)
2
NSCOPE
3
HSTIF
Lateral load-deformation
response (y:z plane)
TSTIF
Torsional load-deformation
response of pile group
FORMGS
Form terms in overall group
stiffness matrix
GENLD
Flexibility and stiffness of
group; response to load cases
Yes
No
Modify data? Stop
Figure 14 Flow chart for PIGLET
39
x x
Px Px
Mx My to z
Mx to z
Tx to y
Mz
My Py Py
Pz y Pz y
Note: Mx to z = -My
z z
(a) Conventional right-hand notation (b) PIGLET notation
Figure 15 Sign convention for loading
40
x
y
Plan view
y
z
Negative Positive
rake rake
Elevation - x:z plane

z
Negative Positive
rake rake
Elevation - y:z plane
Figure 16 Sign convention for pile rake
41

Piglet 6-2 Manual-June 2021

Uploaded by

Copyright:

Available Formats

Piglet 6-2 Manual-June 2021

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Piglet 6-2 Manual-June 2021

Uploaded by

Copyright:

Available Formats

PIGLET

ANALYSIS AND DESIGN OF

RELEASED JUNE 2021

(MANUAL LAST EDITED JUNE 2021)

VERSION 6.2: RELEASED JUNE 2021

(MANUAL LAST EDITED JUNE 2021)

Technical support in relation to operation of the program, or in respect of engineering

Contact Address: 29/82 King Street

Telephone: +61 421 596 075

PART A: GENERAL DESCRIPTION

2 IDEALISATION OF SOIL PROPERTIES ........................................................................ 1

3 RESPONSE OF PILES TO AXIAL LOADING ................................................................ 3

3.1 Solution for single axially loaded piles ...................................................................... 3

3.2 Extension of solution to pile groups ........................................................................... 5

4 RESPONSE OF PILES TO TORSIONAL LOADING ...................................................... 6

5 RESPONSE OF PILES TO LATERAL LOADING .......................................................... 8

5.1 Deformation of single laterally loaded piles .............................................................. 8

5.2 Interaction between laterally loaded piles .................................................................. 9

6 ANALYSIS OF PILE GROUP ......................................................................................... 10

6.1 Treatment of raking piles ......................................................................................... 11

6.2 Allowance for free-standing length of piles ............................................................. 11

7 NON-LINEAR PILE RESPONSE .................................................................................... 12

7.1 Elastic-plastic analysis (E-P) .................................................................................... 12

7.2 Non-linear axial and lateral response (NL-P) .......................................................... 12

9 STRUCTURE OF PROGRAM......................................................................................... 16

10 PROGRAM INPUT .......................................................................................................... 16

10.1 Data input and editing .............................................................................................. 16

10.2 Data Items ................................................................................................................ 17

11 PROGRAM OUTPUT ...................................................................................................... 22

FIGURE TITLES .................................................................................................................... 25

PART A: GENERAL DESCRIPTION

(1) vertical loading only;

(2) vertical and horizontal loading in a single plane;

(3) general three-dimensional loading, including torsion.

2 IDEALISATION OF SOIL PROPERTIES

(a) the value of shear modulus at the ground surface, G0.

(b) the rate of increase of shear modulus with depth, m = dG/dz;

(c) the value of shear modulus at the pile base, Gb.

3 RESPONSE OF PILES TO AXIAL LOADING

3.1 Solution for single axially loaded piles

where, summarising the various dimensionless parameters:

rm = {0.25 + [2.5(1 - ) - 0.25]}L (7a)

rm = 2.5d + {0.25 + [2.5(1 - ) - 0.25]}L (7a)

3.2 Extension of solution to pile groups

where the diffraction factor, , is a function of  and L, according to

2L  sinh  2L    2 sinh  2L   2L   2 cosh  2L   1

4 RESPONSE OF PILES TO TORSIONAL LOADING

where the parameters are as defined previously for axial loading.

where GL is interpreted as the shear modulus at a depth of z = 0.5d(Gp/ GL)0.5.

5 RESPONSE OF PILES TO LATERAL LOADING

5.1 Deformation of single laterally loaded piles

G* = G(1 + 3/4) (18)

where Ep is the equivalent Young's modulus of the pile, given by

the parameter Gc is given by

Gc = G0* + 0.5m*Lc = (1 + 3/4)( G0 + 0.5mLc) (22)

5.2 Interaction between laterally loaded piles

uM = H  uH2

6 ANALYSIS OF PILE GROUP

6.1 Treatment of raking piles

6.2 Allowance for free-standing length of piles