International Journal of Civil Engineering and Technology (IJCIET), ISSN 0976 – 6308 (Print),
ISSN 0976 – 6316(Online) Volume 4, Issue 2, March - April (2013), © IAEME
258
THE EFFECT OF SOIL IMPROVEMENT ON FOUNDATION & SUPER
STRUCTURE DESIGN
Thulaseedharan V
1, Narayanan S.P
2
1(Department of Civil Engineering, Government Engineering College,
Thrissur, Kerala, India) 2 (Department of Civil Engineering, University Technology PETRONAS, Malaysia)
ABSTRACT
For a particular arrangement of superstructure, soil strength is the most important factor
affecting the design of its supporting raft and hence the cost of construction. Folding of raft
or folded plate foundations can be used to reduce the material consumption in a raft design.
The beneficial effects to foundation and superstructure design by the selective soil
improvement below a raft or folded plate foundation were studied in this paper with the help
of Winkler and continuum methods of soil modelling.
Key words: Raft foundation, folded plate foundation, Winkler model, coefficient of subgrade
reaction, continuum analysis, Mohr-Coulomb model, soil improvement.
1.0 INTRODUCTION
1.1 Overview
Raft foundations are provided in situations where soil is not that weak to need pile
foundations but isolated footings cannot be recommended due to the possible higher
differential settlements. The design of a raft depends on the stiffness of superstructure (SS),
column spacing, and projection of raft beyond the outer lines of boundary columns (PR), raft
thickness, soil stiffness, strength of concrete and yield strength of steel. Among the various
parameters listed above, strength of concrete is decided considering the site specific needs
and difficulties in achieving the maximum strength of concrete. Column spacing is decided
by the architectural needs. The SS and substructure are provided with sufficient sizes to give
a safe and economic structure. The soil related term is the least controllable among the above.
In the conventional method of raft design, a constant bearing pressure is assumed below the
foundation which is possible only in the case of very soft soils almost in a fluid state. When
analyzed using commercial software, the soil media may be represented by a system of
INTERNATIONAL JOURNAL OF CIVIL ENGINEERING AND
TECHNOLOGY (IJCIET)
ISSN 0976 – 6308 (Print)
ISSN 0976 – 6316(Online)
Volume 4, Issue 2, March - April (2013), pp. 258-269 © IAEME: www.iaeme.com/ijciet.asp
Journal Impact Factor (2013): 5.3277 (Calculated by GISI) www.jifactor.com
IJCIET
© IAEME
International Journal of Civil Engineering and Technology (IJCIET), ISSN 0976 – 6308 (Print),
ISSN 0976 – 6316(Online) Volume 4, Issue 2, March - April (2013), © IAEME
259
identical but mutually independent, closely spaced, discrete and linearly elastic springs and
this method is based on Winkler’s hypothesis (1867). The elastic constant of the springs is
obtained from the modulus of subgrade reaction, ks and is defined as
ks = �
� (1)
Where q is the load per unit area (or contact pressure) and is the settlement under the
loaded area. The base pressure below a raft may vary from point to point depending on the
load and moment, rigidity of SS and type of soil. As load is applied, the raft settles and the
contact pressure under the raft is re-distributed depending upon the stiffness of foundation
and SS. The settlement of the raft also depends on factors such as the increase in stress-strain
Modulus (E) with depth of soil below raft and the consolidation of soil. Since ks is a variable
depending on the several factors listed above, its computation is very difficult. The common
methods used for the determination of ks are Plate load test, Triaxial Test, Consolidation Test,
CBR test and Empirical relations (Bowles, 1997). Plate load tests are conducted using small
plates of size varying from 30 cm to 76 cm. The stress increase in the soil due to loading on
the plate is felt over a small area whereas that under a raft influences a large area and hence
Terzhagi (1955) suggested an expression for correction for size effect of foundation. Bowles
(1997) suggested the ranges for ks for sandy soil as given in Table 1. Scott (1981) proposed
empirical expression connecting ks with standard penetration resistance (N) for sandy soils.
The simplifying assumptions of Winkler model itself cause some errors (Terzhagi, 1955).
The springs are neither elastic nor independent. The settlement due to the applied load at one
point in the raft is felt at the adjacent areas and hence a uniformly loaded raft may exhibit a
dish shaped settlement, unlike the uniform settlement predicted by Winkler (Coduto, 2001).
Hence efforts were made to couple the springs so that the effect of vertical load is transmitted
in the lateral direction also (Horvath, 2011). In continuum methods the soil media is
represented by 3D finite elements. However continuum analysis is time consuming and
getting the representative soil properties for assigning to the model is very difficult. In
contrast, the Winkler foundation is very simple and large numbers of software are available
based on this method, capable of analysis and design of rafts meeting different country
specific codes of practices. The difficulty in computing ks led ACI committee 336 (1988) to
recommend that the raft designs be carried out varying the value of ks over a range of one
half to 5 or 10 times the furnished value. The furnished value in a soil report is hereafter
called the designated value of ks.
Table 1 Values of ks for sandy soils. Table 2 Stress –Strain Modulus (Bowles, 1997)
Thulaseedharan and Narayanan (2013a) studied the impact of varying ks in the design
of raft foundations. The maximum settlement and the maximum values of top and bottom
bending moments (BM) in a raft generally reduce as ks increases. The comparisons of
Type of Soil ks in kN/m2/m
Loose Sand 4800-16000
Medium Dense Sand 9600-80000
Dense Sand 64000-128000
Clayey Medium Dense
Sand
32000-80000
Type of soil E in kN/m2
Sand-Silty 5000 to 20000
Sand-Loose 10000 to 25000
Sand Dense 50000 to 81000
International Journal of Civil Engineering and Technology (IJCIET), ISSN 0976 – 6308 (Print),
ISSN 0976 – 6316(Online) Volume 4, Issue 2, March - April (2013), © IAEME
260
Winkler and Continuum methods showed slightly lower maximum top moments and more
maximum bottom moments in the later method. It was also concluded that ks based methods
are sufficient for design purpose of flat rafts. For unsymmetrical rafts the design for a range
of ks varying from half to four times the reported value for site resulted in an increase of 2 to
2.5 % in the total reinforcement required. This was further validated using continuum
method. For symmetrical rafts, the increase in reinforcement was around 1% at bottom with
slight reduction in top reinforcement. Designing the raft for such a range of ks or E values
gives a conservative design at a small additional expense. The raft models incorporating
superstructure showed lower settlement values compared to the models without
superstructure implying that the superstructure stiffness helps to re-distribute base pressure.
PR is a possible variable which may be restricted depending on the site conditions.
Calculations were carried out on rafts with varying PR from zero to 1.5 m. The settlement
pattern changes with PR and a dish like settlement (Central part of raft settling more) was
observed for nearly 1.5 m PR in Winkler method. With continuum modeling using Plaxis 3D,
a dish shaped settlement of raft was observed for low values of PR of nearly 0.25 m. As soil
strength is reduced, the settlement pattern changed with more deflection at the edges and
corners of the raft for the same PR. An increase in PR reduces the maximum top moments in
the raft- which usually occurs in the outer-spans of raft. The effect of top moments in a raft is
felt over a large area and hence requires reinforcement to be given for a greater area.
Generally the design bending moments at bottom at the face of columns are much more than
the top moments. However the bottom moments reduces to minimum value within a small
area around the load transfer area of column and rest of the raft in the bottom portion is given
only nominal reinforcement for satisfying crack width requirements. In general the settlement
computations by Winkler method are not comparable to continuum methods, especially when
the soil is soft. Hassan (2011) studied the variation of raft deflection with ks at various
locations in a raft and the influence of column spacing and raft thickness on settlement of
raft. In the current research, column positions are fixed and raft thickness is to be reduced as
much as possible. Hence these aspects are not given much importance. Gupta (1997)
compared the analysis results of rafts using conventional method and Winkler foundations
and concluded that the former is generally on the conservative side. He also reported that
bending moment in a raft may vary several times depending upon the raft size and soil
properties under the raft. This variation increases further as the deviation from symmetry of
the shape or loading of the raft increases. However a variation from 10 kNm to 70 kNm in
Bending Moment (BM) in a raft is an increase by 7 times which may not require an increase
in member size or reinforcement. We are more interested in finding out the change required
in the size of member or area of steel.
Folded plates are widely used in SS for spanning large areas. Due to its folding,
bending moments are reduced, which reduces the required concrete and reinforcement.
Folding is done in straight lines and form work can be placed very easily. In foundations, if
the folding is done in such a way that steep slopes are not provided, then form work can be
avoided. The construction of folded rafts are easy compared to beam and slab rafts and the
additional space created at the basement level by folding a raft can be used for storage of
water or for using as cable trenches. Hanna and Rahman (1990) investigated on the
geotechnical aspects of triangular strip footings and concluded that there is 40 % increase in
bearing capacity when such structures are used as foundations with consequent reduction in
settlement. Thulaseedharan and Narayanan (2013b), compared raft and folded plate
foundations, giving importance to material savings that could achived by using the later. The
International Journal of Civil Engineering and Technology (IJCIET), ISSN 0976 – 6308 (Print),
ISSN 0976 – 6316(Online) Volume 4, Issue 2, March - April (2013), © IAEME
261
design was arrived after conducting analysis for a wide range of soil properties in order to get
a conservative design of both raft and folded plate foundation. Nearly 40% savings in
concrete, 30 % savings in steel and overall 35% savings in cost of construction could be
achieved by folding the raft. Between folded and flat rafts, folded raft structure requires
slightly lower reinforcement for columns and beams. In continuum method, the settlements
under corner and edge columns were more in folded rafts compared to flat rafts, especially
when the raft projections were low. This is found to be due to the heavy lateral loads at those
supports and due to which folded raft was deflecting in the fold direction. This increases the
reinforcement in the outer spans of folded raft and in the corner and edge columns at the
ground floor of the building. BM in the outer beams of ground floor also showed some
increase compared to flat raft. As E of soil increases, the BM in raft, column reinforcement
and BM in beams reduces. The increase in PR brings significant changes to top moment
values in the raft. Column and beam reinforcement were slightly reduced as PR increases. It
also reduces reinforcement required and settlement of both flat and folded rafts. In general
Winkler method is sufficient for the design of flat rafts and in the case of folded plates, lateral
stiffness of soil need to be considered in the analysis. The column designs obtained by giving
fixed supports to the columns and sometimes even that obtained from Winkler method may
not give a conservative design pointing the need for elaborate soil structure interaction
studies including continuum analysis for important structures. The designs of folded plates
are affected by the central rise or fall of fold.(In foundations, the folded part is going below
ground level and hence the term fall is used). With increase in the fall, reinforcement required
can be further reduced. As the fall of the folded raft increases, settlement reduces due to the
increase in stiffness of foundation. Reinforcement required in folded raft also reduces.
Similarly column and beam reinforcement in the superstructure was slightly lower compared
to one with less fold height.
1.2 Objectives of study The present paper compares the performance of folded plate and raft foundation under
identical SS stiffness and loading when soil strength parameters are varied below it at
different locations. The positions of maximum settlement in a raft were identified using
Winkler and continuum methods and soil strength was then increased at those places.
Numerically, this strengthening was accounted with increase in ks and E values. Winkler
method is very popular and easy to use for soil modeling and its inherent disadvantages were
taken care in this study with the help of continuum methods. There is uncertainty in the
determination of ks and E values and hence the impact of its variation over a range in
foundation and SS design were also studied.
1.3 Importance of the study The increased soil stiffness reduces settlement of raft and with several other benefits
to substructure and superstructure. The localized improvement in soil strength (LISS) can
reduce differential settlements along with considerable reduction in material requirements of
steel and concrete. Folded plate foundations are provided to reduce material consumption and
localized strengthening of soil may further improve the advantages.
International Journal of Civil Engineering and Technology (IJCIET), ISSN 0976 – 6308 (Print),
ISSN 0976 – 6316(Online) Volume 4, Issue 2, March - April (2013), © IAEME
262
2.0 MODELING AND STRUCTURAL IDEALIZATION
2.1 Winkler and continuum modeling: StaadPro (2008) and SAFE (2009) were used to model the Winkler Foundation. The
ks of the proposed site is given as 25000 kN/m2/m and this value is hereafter referred to as
designated ks. The analysis was carried out with and without the SS. The SS was modeled
using line and plate elements and the substructure with quadrilateral plates. Several analyses
were carried out taking typical situations. All designs were carried out using BS 8110 and
BS 8007. Continuum modeling was carried out using Abaqus (2011), Plaxis 3D (2004) and
StaadPro (2008) . Solid elements were used to model the soil mass. In Abaqus and Plaxis,
interface option is possible, where as in StaadPro, rough contact is assumed. In StaadPro
modelling, only E is varied. Mohr- Coulomb model was used for the study purpose in Plaxis
and Abaqus as the present comparison is made for a site containing sandy strata. Mohr-
Coulomb model in Plaxis requires 5 parameters as input namely the Cohesion (C), the angle
of Internal Friction ( θ), the Modulus of Elasticity (E), the dilatancy and the Poisson’s ratio.
Here the exact evaluation of E is very difficult and hence a range of values are taken
consistent with the known soil properties. In this study the E values were varied from 15000
to 60000 kN/m2 and from 30 to 43 degrees. A very small value of cohesion is given to aid
computation as recommended in Plaxis 3D foundation user manual (2004). The
recommended range of variation of E values is given in Table 2 (Bowles, 1997). Dilatancy
value is given as zero and the Poisson’s ratio was varied from 0.35 to 0.4. Abaqus was used
only for comparing the results obtained from the structural software Staad for the folded
plate.The comparison of designs were done between flat slab raft and folded raft. The rafts
were analyzed for 43 service load combinations and 53 ultimate load combinations. M40
concrete and Steel of grade 460 was used. The comparison was made on a raft consisting of 4
equal spans of 8 m in the X direction and 7m in the Y direction. Two projections of raft in X
direction, 0.3m and 1m were considered in the present studies. Raft projection in the Y
direction is kept at 0.3 m. A 3D view of the folded raft is given in Fig. 1. Folding is
introduced in the X direction in such a way that the inclination of the surface is less than 32
degrees. The building is seven storied with column sizes of 600x600 mm. Seismic forces are
generated as per UBC (1997) for zone 2A and soil classification Sc as per the site
requirements.
3.0 ANALYSIS AND DESIGN RESULTS
The flat and folded raft models were analyzed using Winkler soil model with 3 values
of ks ( 12500, 25000 50000 kN/m2/m). The continuum analysis was carried out in two steps.
In one step, E is the only variable with values of 15000, 30000 and 60000 kN/ m2. In the
second case, E and were varied for analysis with Mohr-Coulomb model. To study the
effect of soil improvement, ks or E values are varied at select locations below the raft; the
increase was 4 times in ks or E values compared to the adjacent area. The impact of the same
on structural design was worked out. Analysis and design results are given in eight sub-parts.
Variation of settlement below raft, base pressure, BM in rafts, Shear and Impact of raft
projections are covered in the first five parts. Part 6, contains the design of raft and folded
plates. Part 7 deals with the impact of soil stiffness on SS design and the eighth part is a
comparison of material and cost savings. In each section, a comparison on the performance of
flat and folded plate foundations are given for both Winkler and continuum modeling.
International Journal of Civil Engineering and Technology (IJCIET), ISSN 0976 – 6308 (Print),
ISSN 0976 – 6316(Online) Volume 4, Issue 2, March - April (2013), © IAEME
263
Fig.1 View of the raft folded in X- direction.
3.1 Comparison of settlement The maximum settlement of the foundation was reduced as ks or E increased in both
Winkler and continuum methods. When PR is considerable, the raft may settle in a dish
shape with more deflection at centre. With LISS, the settlement pattern is changing in
addition to reduction in settlement. Without any increase in PR, the location of maximum
settlement of raft shifts to centre from the edges.
International Journal of Civil Engineering and Technology (IJCIET), ISSN 0976 – 6308 (Print),
ISSN 0976 – 6316(Online) Volume 4, Issue 2, March - April (2013), © IAEME
264
Fig.2 Variation of maximum settlement below flat and folded rafts in continuum method with
change in E for LISS
In Fig 2, the maximum settlement is more below folded raft in continuum method compared
to flat raft. This is due to the very high lateral forces applied and consequent lateral
displacement at the supports of corner and edge columns in X direction (Fig.1). The LISS
changes the settlement pattern of the folded raft. Even for low PR values, dish like settlement
pattern is obtained. The maximum settlement is reduced as shown in fig.2.
3.2. Variation of base pressure below the raft and the folded plates Base pressure below the raft was more under the column load transfer area. For
smaller PR values, the maximum base pressure occurred under corner and edge columns. As
PR increases, the area of raft increases and maximum base pressure below the raft is reduced
significantly. At interior columns, the base pressure under column load transfer point and
adjacent areas were almost uniform for low values of ks. As ks increases, the difference
between maximum and minimum base pressure adjacent to the load transfer area also
increased. The increase in ks leads to load transfer through a small concentrated area below
load point. The same trend was observed below the folded rafts with much less variation
between the maximum and minimum base pressure anywhere. This may be due to the higher
stiffness of folded raft re-distributing base pressure. Continuum modeling also gave similar
results for both flat and folded rafts. LISS results in very high base pressure at the soil
improved area and the values go on reducing as the ks or E in the central area increases.
3.3 Variation of maximum design moments
The maximum value of bottom moments at the face of columns in the raft is taken as
the design bottom moment in a raft. As ks or E increases, the bottom and top moments are
reduced (Fig. 3a and 3b). With LISS, for the flat raft, considerable reduction in maximum
top moment occurs along with increase in bottom moments for low E values. For folded rafts
with LISS at the boundary area, maximum top moments were slightly increased in the outer
most fold portion compared to the case without LISS as shown in fig.3 (a). Continuum
methods showed a reduction of 67% for the maximum top moment in X-direction compared
to flat raft. Bottom moment (Mx) was reduced by 50%.
0
2
4
6
8
10
12
14
16
18
20
0 10000 20000 30000 40000 50000 60000 70000
Se
ttle
me
nt
in m
m
Young's Modulus E in kN/m2
Flat Raft
Folded Raft
Flat Raft with Soil
improvement
Folded Raft with
Soil improvement
International Journal of Civil Engineering and Technology (IJCIET), ISSN 0976 – 6308 (Print),
ISSN 0976 – 6316(Online) Volume 4, Issue 2, March - April (2013), © IAEME
265
Fig.3 a and 3b Variation of top and bottom maximum BM in folded and flat rafts using
continuum methods
The top bending moment Mx in the fold portion was reduced by 80% and goes on
reducing with increase in fall of the folded portion. BM in y direction is concentrated in a
narrow width of folded raft supporting the columns and for the rest of the area BM was
reduced by 90%.The increase in fall of the folded portion increases the stiffness and reduces
settlement and which in turn reduces BM and hence the reinforcement needed in a raft.
3.4 Influence of shear The thickness of the folded raft is increased for a small cross-section supporting the
columns as shown in Fig. 1. No other special care was required in comparison to the flat rafts.
3.5. Impact of projections of the raft (PR) beyond the outer line of boundary columns By providing projections to the raft, some other advantages were also observed. There
was a reduction in the total and differential settlements with change in deflection pattern and
reduction in reinforcement in the substructure and SS. The studies on folded rafts with
continuum model also gave similar results. If PR could be increased, then LISS is not
0
100
200
300
400
500
600
700
800
900
0 10000 20000 30000 40000 50000 60000 70000
To
p m
om
en
t in
kN
m
Young's Modulus E in kN/m2
Flat Raft
Folded Raft
Flat Raft with Soil
improvement
Folded Raft with Soil
improvement
0
200
400
600
800
1000
1200
1400
0 10000 20000 30000 40000 50000 60000 70000
Bo
tto
m m
om
en
t in
kN
m
Young's Modulus E in kN/m2
Flat Raft
Folded Raft
Flat Raft with
Soil
improvementFolded Raft
with Soil
improvement
International Journal of Civil Engineering and Technology (IJCIET), ISSN 0976 – 6308 (Print),
ISSN 0976 – 6316(Online) Volume 4, Issue 2, March - April (2013), © IAEME
266
required at the boundary area. However if the central area of the raft is subjected to more
settlement, then LISS at that area is beneficial.
3.6 Design of Flat and folded Rafts Minimum reinforcement was decided considering the crack width limitations. For 900
mm raft the value of this moment is 375 KNm for Service load in main direction and 325
kNm for secondary moment for a reinforcement of 20mm@ 200 mm c/c, for a crack width of
0.3mm. Similarly the other ranges are worked out for different diameters of extra bars to be
provided like 20@200 mm c/c, 25@ 200 mm c/c etc. After finding out ranges of BM for
different combinations of reinforcements, the raft BM at different locations were grouped in
to these ranges and reinforcement was provided accordingly. Then the raft was investigated
for one way shear and punching shear. At few locations, the bottom reinforcement is
increased to give additional shear capacity for avoiding shear links. After completing the
reinforcement design, the raft was further analyzed changing the ks values to 12500
KN/sqm/m and locations where reinforcement requires modification were identified. Then
analysis was repeated after varying the ks to 50000 Kn/m 2/m and design was reviewed and
reinforcement detailing done incorporating all the cases of maximum moments. After
working out the total quantity the design was checked for continuum modeling. There was no
increase in reinforcement required for variation in ks, may be due to the symmetry of the
structure. Checking using Continuum method resulted in an increase of 1% reinforcement in
the bottom area. There was a decrease in reinforcement in the top area and was neglected.
For folded rafts, the same design procedure was followed, though the reinforcement required
and ranges of moments were different. The thickness of folded portion is 350mm and it is
more at column supports. The folded portion was subjected to much less moments and hence
much less thickness and reinforcement were needed. The impact of varying ks values from
half to two times the designated value was found to be insignificant as far as structural design
was concerned. However continuum modeling required more reinforcement upto 9%. This
showed the need for modifying the Winkler Method giving stiffness to soil in the lateral
directions also. With LISS, there was a small increase in reinforcement required at the bottom
and reduction in reinforcement required for the top in the case of flat raft, the later being
much more. In the case of folded raft, the top reinforcement needed increased in the outer
spans of fold area and reduced at several other places at the bottom. The net effect was that
there was no significant savings in reinforcement for the folded raft but other advantages
were there in superstructure design along with reduction in raft settlement.
3.7 Impact on SS design.
For Winkler model, it is observed that the column reinforcement decreases with
increasing ks values for both flat and folded rafts. Continuum analysis of flat rafts also
showed similar results. With soil improvement there is further reduction in column
reinforcements as shown in Fig. 4a. Buildings supported on folded raft required more
reinforcement for edge and corner columns at Ground floor (Fig.4b). The figures 4 and 5 also
contain comparisons with cases involving fixed end condition to columns. For the interior
columns, reinforcement required was slightly lower compared to building with flat raft
foundation, for all floors. LISS at the boundary reduces column reinforcements as shown in
Fig. 4a and 4b for both flat and folded plate foundations. In general beam bending moments
are also reduced with soil improvement with occasional local variations as shown in Fig 5a
and 5b.
International Journal of Civil Engineering and Technology (IJCIET), ISSN 0976 – 6308 (Print),
ISSN 0976 – 6316(Online) Volume 4, Issue 2, March - April (2013), © IAEME
267
(a) Winkler Method
(b) Continuum Method
Fig. 4. Variation in corner column reinforcement with soil strength
(a) Winkler Method
1500
2000
2500
3000
3500
4000
4500
0 20000 40000 60000Are
a o
f S
tee
l in
mm
2
Modulus of Subgrade reaction in kN/m2/m
Flat Raft without
soil improvement
Flat Raft with soil
improvement
Fixed Support
1500
2000
2500
3000
3500
4000
4500
0 20000 40000 60000 80000Are
a o
f S
tee
l in
mm
2
Young's Modulus in kN/m2
Flat Raft without Soil
improvement
Folded Raft without soil
improvement
Flat Raft with Soil
improvement
Folded Raft with soil
improvement
540550560570580590
0 20000 40000 60000Ma
x.
Mo
me
nt
in
be
am
in k
Nm
Modulus of subgrade reaction in kN/m2/m
Flat Raft without soil
improvementFlat Raft with soil
improvementFixed Support
International Journal of Civil Engineering and Technology (IJCIET), ISSN 0976 – 6308 (Print),
ISSN 0976 – 6316(Online) Volume 4, Issue 2, March - April (2013), © IAEME
268
(b) Continuum Method
Fig.5. the variations in a GF beam BM in Winkler and continuum method.
3.8. Comparison of material consumption and cost of execution. For a flat raft, LISS reduces the maximum top moment and hence reinforcement
considerably and it more than compensates the small increase in bottom reinforcement. There is a
small reduction in BM of the beams for all floors of the building with LISS and the variation was
very small and may not affect design. For folded rafts, the top BM in the outer spans were
slightly increased in fold direction. In superstructure design, there was an overall reduction in
beam moments and column reinforcements. If the cost of LISS is not substantial, considering the
overall advantages- especially in settlement reduction- it is a better option.
4.0 CONCLUSIONS
A raft and folded plate foundation were designed for varying ks values from half to
two times the designated values. Soil improvement was done at the boundary of the raft
below columns. For flat rafts, the bottom maximum BM increased and the top maximum BM
reduced in the case of LISS compared to the case without soil improvement. Continuum
methods showed similarity in the BM values obtained with those computed using ks with
slightly higher bottom moments and lower top moments. The top reinforcement required
reduces and the decrease is much more than the increase in bottom reinforcement. Compared
to the design made for the superstructure of flat raft without soil improvement below it,
column reinforcement required reduces with LISS. Similarly the bending moments are
reduced for the beams throughout the building due to LISS, the variation being very small.
LISS also helps to reduce the total and differential settlements. Between folded and flat rafts,
folded raft structure requires slightly lower reinforcement for columns and beams. With
LISS, the reinforcement required further reduces in both cases. In continuum method, the
settlements under corner and edge columns were more in folded rafts compared to flat rafts,
especially when the raft projections were low. This is found to be due to the heavy lateral
loads at those supports and due to which folded raft was deflecting in the fold direction. This
increases the reinforcement in the outer spans of folded raft and in the corner and edge
columns of ground floor. Hence Winkler methods may not give conservative results as far as
column designs are concerned. The effect of LISS reduces such local high values of
settlement and the increase in column reinforcement. In folded rafts, slight increase in top and
bottom reinforcement was needed with LISS at certain areas and there was a reduction in
some other areas. Top reinforcement increases for the outer spans and bottom reinforcement
525
530
535
540
545
550
555
560
0 20000 40000 60000 80000Ma
x.
Mo
me
nt
in b
ea
m i
n
kN
m
Young's Modulus in kN/m2
Flat Raft without soil
improvement
Folded Raft without soil
improvement
Flat Raft with soil
improvement
Folded Raft with soil
improvement
Fixed Support
International Journal of Civil Engineering and Technology (IJCIET), ISSN 0976 – 6308 (Print),
ISSN 0976 – 6316(Online) Volume 4, Issue 2, March - April (2013), © IAEME
269
in the central area. Overall there was not much change in the reinforcement required. BM in the
outer beams of ground floor also showed some increase in BM compared to that with the flat raft.
As E of soil increases, the BM in raft, column reinforcement and BM in beams reduces. At
locations where settlement of the raft is computed to be more, with LISS, there are several
advantages in the overall performance of sub-structure and superstructure. However, the cost
savings depends on the cost of LISS and may not be substantial.
REFERENCES
[1] Winkler. E, (1867), Die Lehre von Elastizitat und Festigkeit (On Elasticity and Fixity),
Prague, p.182.
[2] Bowles, J.E (1997), Foundation Analysis and Design, 5th Edition, Mc GrawHill
International Edition, Singapore.
[3] Terzaghi, K. (1955), “Evaluation of Coefficients of subgrade reaction”, Geotechnique, 4,
297-326.
[4] Scott, R.F. (1981), Foundation analysis, Prentice Hall Inc.
[5] Coduto, D. P (2001), Foundation design: principles and practices, Prentice Hall Pvt Ltd,
pp.352-369.
[6] Horvath, J.S and Colasanti R.J (2011), “Practical Subgrade Model for Improved Soil-
Structure Interaction Analysis: Model Development”, International Journal of Geomechanics, 11,
1, February 1.
[7] ACI Committee 336.2R (1988), Suggested design procedure for Combined Footings and
Mats, Detroit, pp-13.
[8] Thulaseedharan, V. and Narayanan, S.P (2013a), “Impact of coefficient of subgrade
reaction on raft foundation design”, International Journal of Applied Engineering Research,
8, 2, pp. 187-201 © Research India Publication. ISSN 0973-4562
[9] Plaxis 3D Foundation User Manual, 2004 PLAXIS bv, Netherlands.
[10] Hassan, I.M. (2011), “Influence of structural and soil parameters on Mat deflection”,
International Journal of Civil and Structural Engineering, 2, 1.
[11] Gupta, S.C. (1997), Raft Foundation Design and Analysis with A Practical Approach,
New Age International, New Delhi.
[12] Hanna A and Mohamed Abd El- Rahman (1990), “Ultimate bearing capacity of triangular
shell strip footings on sand”, Journal of Geotechnical Engineering, ASCE, 116, 12.
[13] Thulaseedharan, V. and Narayanan, S.P (2013b), “Comparison of Raft and folded Plate
Foundations”, International Journal of Applied Engineering Research, (accepted for
publication) © Research India Publication. ISSN 0973-4562
[14] StaadPro V8i, 2008, Technical references, www.bentley.com
[15] SAFE v12, Analysis reference Manual, 2009, Computers & structures, Inc., Berkeley.
[16] BS 8110 Part 1 (1997), Structural Use of Concrete: Code practice for design and
construction.
[17] BS 8007 (1987), Code of Practice for design of concrete structures for retaining
aqueous liquids.
[18] Abaqus User Manual (2011), Dassault Systems Simulia Corporation, Providence, RI,
USA.
[19] UBC 1997, Uniform Building Code, International Council of Building Officials.
[20] M. Alhassan and I. L. Boiko, “Effect Of Vertical Cross-Sectional Shape Of Foundation
And Soil Reinforcement On Settlement And Bearing Capacity Of Soils” International
Journal Of Civil Engineering & Technology (IJCIET) Volume 4,Issue 2,2013,Pp: 80 –
88, Issn Print : 0976 – 6308, Issn Online : 0976 – 6316.