JOURNAL OF APPLIED ENGINEERING SCIENCES VOL.

14(27), ISSUE 1/2024

ISSN: 2247-3769 / e-ISSN: 2284-7197 ART.NO. 395 pp. 55-64

PARAMETRIC ANALYSIS FOR A LOW-VOLUME CONCRETE ROAD

WITH REINFORCED SUBGRADE SUBJECTED TO SWELLING
PRESSURE
Ashwini Chavana, *, S. Bhosale a
a
Civil Engineering Department, COEP Technological University, Shivajinagar, Pune,
Maharashtra, India; E-mails: * [email protected] / [email protected], [email protected]

Received: 11.10.2023 / Accepted: 28.12.2023 / Revised: 31.01.2024 / Available online: 31.05.2024

DOI: 10.2478/jaes-2024-0007

KEY WORDS: expansive soils, geosynthetics, pavement design, soil-structure interaction, swelling pressure, concrete pavement.

ABSTRACT:

This paper presents the parametric analysis of a combined system of a low-volume concrete pavement, base layer and reinforced
expansive soil, subjected to the stresses due to wheel loading and swelling of the soil. A soil-structure Interaction model is
considered for the analysis, in which the pavement and the reinforcement are idealized as a finite beam, with plane-strain conditions,
whereas a base layer and the expansive soil are modelled as Winkler Springs of different stiffnesses. During the analysis, it was
taken into account that the geosynthetic reinforcement, which could be in the form of a geogrid, geocell, or a combination of both,
possesses bending stiffness and is placed at the interface between the base layer and the expansive subgrade. The reinforcement layer
is subjected to stresses due to wheel load, self-weight of pavement slab and surcharge load of the base layer at the top and the
swelling pressure from the subgrade at its bottom. The governing differential equations for the flexural response of the model are
derived and a closed-form solution is presented in a non-dimensional form. The outcomes of the parametric analysis highlight that
the flexural response of the pavement is predominantly affected by the relative stiffness of the base layer and the expansive subgrade.
In contrast, the relative flexural rigidity of the upper and lower beams has a comparatively minor impact on the model's response.
Additionally, parameters like the depth of placement of the lower beam, unit weight of the upper soil layer, and the self-weight of the
upper beam also contribute to influencing the response of the pavement model. The study suggests that, while designing for the
concrete pavement, opting for an increase in the modulus of subgrade of expansive soil may be a preferable choice over reinforcing
the foundation of the pavement.

1. INTRODUCTION tion of excess swell pressure from expansive soil

(Chittoori, Mishra, and Islam 2018; Md Ashrafuzzaman
Low-volume roads are those which carry average daily Khan et al. 2020; Tamim 2017; Kief, Ph, and
traffic of less than 450 Commercial Vehicles Per Day Geotechnical 2016; Vessely and Wu 2002; Al-Omari and
(CVPD). The Indian design code of low-volume concrete Hamodi 1991; Zornberg and Gupta 2009; Zornberg,
pavements recommends the use of a non-expansive Gupta, and Ferreira 2010). However, most of these
subgrade with a minimum CBR of 4% to 8%, to rest the studies focus on flexible pavements or unpaved roads.
pavement system (IRC:SP62 2014). However, local soils Maheshwari et al. (P Maheshwari, Basudhar, and Chandra
which may be expansive, with low CBR of 2% to 3% in 2004) proposed a double-beam model to represent the
wet conditions, are sometimes used for economic reasons. structure and geosynthetic reinforcement as a beam. Other
These pavements would experience stresses due to studies using this model mainly focused on the analysis of
loading (wheel load and environmental load) and swelling strip footings and rails (Priti Maheshwari 2008; Priti
(from the subgrade). Concerning this, various chemical Maheshwari and Viladkar 2009; Priti Maheshwari,
stabilization methods (with lime, cement, additives, Basudhar, and Chandra 2006). Notably, subgrade soil in
industrial waste etc.) and mechanical or physical these models was non-expansive. Musa (Musa 2018)
stabilization methods (such as compaction, pre-wetting, proposed an analytical solution for a long beam subjected
geosynthetic reinforcement etc.) have evolved. to swelling pressure using the superposition technique.
This solution has practical implications for the design of a
The inclusion of geosynthetic reinforcement in the base load-bearing masonry wall. Khan and Wang (Md Adnan
layer or subgrade is found to be successful and cost- Khan and Wang 2017) and Khan et al. (Md Adnan Khan,
effective up to some extent, in solving a dual problem of Wang, and Sarker 2020) developed a closed-form
distributing the stresses from wheel loading and dissipa- solution using Winkler Model for the asphalt concrete
pavement resting on expansive soil.
* Corresponding author. Ashwini Vikas Chavan, Research Scholar, Civil Engineering Department, COEP Technological
University, [email protected], India.

55
JOURNAL OF APPLIED ENGINEERING SCIENCES VOL. 14(27), ISSUE 1/2024
ISSN: 2247-3769 / e-ISSN: 2284-7197 ART.NO. 395 pp. 55-64

They adopted a virtual load concept to solve the problem using will be the predominant mode for the response of the
inverse theory for the analysis of stresses, However, they did reinforcement (Priti Maheshwari, Basudhar, and Chandra 2006).
not consider any reinforcement mechanism in the pavement
foundation system. Wang et al. (Wang, Sarker, and Ikra 2018) Wheel load (Q)
developed the model considering reinforced asphalt concrete 2l
pavement as a Timoshenko beam with expansive soil as a Concrete Pavement
Pasternak model. Sarker et al. (Sarker, Wang, and Khan 2019) (E1I1)
proposed a solution for the pavement with geosynthetic h Granular base layer (k1, 1)
reinforcement by applying the theory of inverse analysis and the
virtual load method. Geosynthetic
reinforcement
Presently, Lytton's method has gained widespread acceptance Expansive soil (k2, 2)
for the design of pavement slabs on expansive soil from both (E2I2)
the Texas Department of Transportation (DOT) and the Federal
Highway Administration (FHWA). Lytton et al.(Robert Lytton, Figure 1. Definition sketch of the problem
Aubeny, and Bulut 2004; R. Lytton, Aubeny, and Bulut 2006)
proposed a suction-based approach for the design of rigid
pavement resting on expansive soil, which was the compilation 2.2 Governing Differential Equations
of the results of several studies (McKeen 1978; Mckeen and
Mckeen 1984; Aitchison and Holmes 1965; Covar and Lytton Fig. 2 shows a mechanical model (plane strain condition) for the
2001; R. L. Lytton 1978; McDowell, Herner, and Wooltorton problem mentioned in Fig.1. A concrete pavement and
1956; McKeen and Hamberg 1981). This approach entails the geosynthetic reinforcement are idealized as a finite beam (upper
computation of a suction profile within the moisture active and lower beam respectively) of length 2l and flexural rigidity
zone, utilizing an equation put forth by Mitchell (1980). E1I1 and E2I2 respectively. The moduli of subgrade reactions of
Furthermore, the index properties of the soil are utilized to the base layer and subgrade soil are idealized as Winkler
determine the Suction Compression Index (γh), a parameter that Springs of stiffnesses k1 and k2 respectively. A concentrated
establishes a correlation between matric suction and volume wheel load Q is acting in the middle of the upper beam, whereas
changes in the soil (Covar and Lytton 2001). The method also a lower beam is subjected to a surcharge load ( where
investigates different stabilization techniques for expansive soil is the unit weight of the base layer soil, is the depth of
and explores mitigation methods for both swell and shrinkage. placement of the lower beam and w is the self-weight of the
However, the primary focus of the method is to assess reliability pavement beam from the top and swelling pressure from the
and serviceability indices for analyzing pavement performance. compacted expansive soil at the bottom.
The approach involves the use of specialized software to
generate results. Other methods, like the Post-tensioning Q
Institute method (Post-Tensioning 2003), which is based on
Lytton’s approach, are commonly used for designing slabs on 2l
the ground on expansive soil. E1I
1 h
The review of the literature shows that a simplified mechanical y1 h k1
model for a low-volume concrete pavement with reinforced E2I2
subgrade, considering swelling pressure as a design parameter
has not been studied so far. Concerning this, the objectives of
the present study are as follows. y2 Psc
k2
 To develop an analytical solution for a mechanical
model considering a pavement- geosynthetics - x
expansive soil interaction, subjected to loading and
swelling stresses for a low-volume traffic road. y
 To generalise the model by presenting the analytical Figure 2. Idealization of the problem into a Mechanical Model
solution in the non-dimensional form.
 To explore the influence of various parameters on the When the width of the pavement is wide enough to model it as
performance of the flexural response of pavement
a long beam, the swelling pressure at the centre of the beam and
and the reinforcement through a parametric analysis.
the edge of the beam will differ due to the difference in suction
at these locations (Perera et al. 2004). However, for smaller
2. MATERIALS AND METHODS widths (half-width is 2m), the change in suction across the
length of the lower beam will be uniform. Therefore, swelling
2.1 Statement of problem pressure at the bottom of the lower beam is assumed to be
Fig.1 shows a concrete pavement resting on a granular base uniform.
layer (thickness h) and subjected to a concentrated wheel load in
the middle of the pavement. Reinforcement is placed at the The analysis is to be carried out to evaluate the flexural
interface of the granular base layer and the compacted response of the upper and lower beam considering the wheel
expansive subgrade. As the geosynthetic reinforcement is loading and the range of swelling pressure. The effect of various
subjected to a surcharge load at the top and the swelling parameters such as the normalized depth of placement of
pressure at the bottom from the soil, it is assumed that bending reinforcement ( , the relative flexural rigidity of the upper and

56
JOURNAL OF APPLIED ENGINEERING SCIENCES VOL. 14(27), ISSUE 1/2024
ISSN: 2247-3769 / e-ISSN: 2284-7197 ART.NO. 395 pp. 55-64

lower beam (R), the relative stiffness of base-layer and subgrade

soil (r) and the normalized self-weight of the upper beam (w′)
on the response of the model are to be studied. (8)
Eq. (8) can be rewritten as:
The governing differential equations are derived in this section, (9)
for the proposed model and the closed-form solution is
presented. Owing to the symmetry, only one-half of the model
(x ≥ 0) is considered for the analysis. The weight of the upper Where:
beam w is considered uniformly distributed on the base layer and
below. The deflection ordinates of the upper and lower beam
are denoted as y1 and y2 respectively. The reaction force for the Then, eight roots of the auxiliary Eq. (9) will be:
upper and lower beams is expressed as ) and
respectively. Sign conventions are taken as positive and
for downward forces and negative for upward forces.
Where and
For the region (
The governing differential equation for the upper beam:
Considering and
(1)
The non-dimensional deflection of the upper beam is:
Solving Eq. (1) for , we get Eq. (2):
(2)
4 1 + 2 5 2 + 6 2 + 2 7
2 + 8 2 +( 1 + )
The differential equation for the deflection of the lower beam is (10)
The non-dimensional deflection of the lower beam is:
(3)

Simplifying the above equation to Eq. (4)

4 24 7 2 + 8 2 + 1 + ;
(4) (11)
Substituting from Eq. (2) to Eq. (4), we get
The expression for the normalized bending moment of the upper
and the lower beam can be obtained by differentiating Eq. (10)
(5) and Eq. (11). (5)

Arranging Eq. (5) 2.3 Boundary and continuity conditions

There are eight constants of integration, which need to be

(6) evaluated using appropriate boundary and continuity conditions.
For the upper beam, at the point of application of load, i.e., at x
= 0, the slope of the deflected shape of the beam is zero and the
Eq. (2) and Eq. (6) are the governing differential equations for
shear force is Q/2. At the edge of the upper beam, i.e., at x = l,
the proposed model. These equations are converted into a non-
the bending moment and shear force are zero for free ends. For
dimensional form using non-dimensional parameters as below.
the lower beam, which is within the foundation soil, at point x =
(Matlock and Reese 1960). R1 and R2 are the characteristic
0, the slope of the deflected shape of the beam and shear force is
lengths of the upper and lower beams respectively and are
zero and at x = l, the bending moment and shear force are zero.
defined as:
From the above-mentioned boundary conditions, Eq.(12) is
presented in the non-dimensional form.
and For upper beam:

At and
Let , , , , ,
At and
, ,
For lower beam
The governing differential Eq. (2) and Eq. (6) in non-
dimensional form can be written as follows: At and

At and (12)
(7)
Using the above boundary conditions eight equations are
formed. Solving these equations simultaneously using

57
JOURNAL OF APPLIED ENGINEERING SCIENCES VOL. 14(27), ISSUE 1/2024
ISSN: 2247-3769 / e-ISSN: 2284-7197 ART.NO. 395 pp. 55-64

MATLAB software eight unknown constants (C1 to C8) are pressure for the values of initial matric suction. Tu and
evaluated. The present mechanical model is based on Hetenyi’s Vanapalli (2016) proposed two different equations to estimate
solution (Hetényi 1946) of the beam on an elastic foundation the swelling pressure exerted by natural and compacted
and applies the well-established procedure for the development expansive soil, based on the SWCC. The equations are helpful
of governing differential equations for the equilibrium of forces to estimate the maximum swell pressure soil can experience, for
and its solution. The only difference is in Hetenyi’s model the all possible initial suction ranges and degree of saturation,
lower beam is imaginary whereas in the proposed model it is a knowing the conditions of soil at the time of construction of
physical object (Priti Maheshwari, Basudhar, and Chandra, pavement. Though the proposed model can handle any range of
2006). The validation of the model with Hetenyi’s solution is swelling pressure, for the present analysis, the range of the
already presented by (Priti Maheshwari, Basudhar, and Chandr, swelling pressure of compacted expansive soil is taken as
2006). 50 to 500 kPa.

To ensure the computational reliability and accuracy of the 2.4.2 Stiffness of upper and lower soil layer soil
proposed solution using the MATLAB Program, the algorithm The base layer (upper soil layer in the proposed model) is
is first compared with the solution presented by Maheshwari et generally compacted granular soil, but any non-swelling local
al. (Priti Maheshwari, Basudhar, and Chandra, 2006). This soil can be used to support the pavement. The range of values of
comparison is made without considering the swelling pressure stiffness for an upper soil layer (k1) is considered from 10000
and surcharge load on the lower beam. It is observed that the kPa/m to 50000 kPa/m for the present analysis.
results show a consistent match as shown in Fig. (3) for the
deflection of the beam considering non-dimensional parameters The modulus of subgrade reaction or stiffness of soil is an
R=1, r=5 and xm=8. However, to gain confidence in the important parameter in pavement design. It is generally
accuracy of the response of the model, it needs to be verified determined by carrying out a Plate Load Test or can be
with laboratory or field experimentation for known swelling correlated using equations based on the soaked CBR value or
pressure and other parameters. Dynamic Cone Penetrometer Test of soil (IRC:SP62 2014).
However, these correlations are not rational for the expansive
soils as the behaviour of expansive soil depends on the amount
and type of clay content (montmorillonite, kaolinite, illite etc) in
the soil. Payne and Cameron (Payne and Cameron 2014)
presented ‘The Walsh method of the swell-stiffness’ in the form
of a bi-linear curve for swelling clay soil. Where swell stiffness
is the pressure required to suppress a meter of heave in units
kPa/m. For beams in contact with the soil, the swell stiffness
can be calculated as 100q kPa/m, where q is the average loading
pressure on the soil including the self-weight of the structure
(Standards Australia 2011). The stiffness of soil (k2 ) can also
be evaluated by knowing the value of the modulus of elasticity
(E2) of clay from Eq.(12)

(13)

Figure 3. Comparison of deflection of the upper beam with Where: b is the width of the loaded area in m and is the
Maheshwari et al. (2006) Poisson's ratio of clay. Vanapalli and Oh (Vanapalli and Oh
2010) presented an equation for the modulus of elasticity
considering the matric suction for the soils with a plasticity
2.4 Selection of range of values of design parameters index <16 %. This model is then extended by Adem and
Vanapalli (Adem and Vanapalli 2015) for the modulus of
2.4.1 Soil-suction approach for the computation of swelling elasticity of unsaturated expansive soil based on SWCC and the
pressure Expansive soil tends to swell when it comes in contact modulus of elasticity of soil in a saturated condition as shown in
with water. Swelling pressure gets exerted on the pavement Eq. (14).
when the swelling of expansive soil is restricted by it. Reliable
prediction of swelling pressure and its impact on the pavement (14)
structure is a major challenge to design engineers. Swelling
pressure can be computed in the laboratory through different
types of oedometer tests or using empirical or semi-empirical Where is the modulus of elasticity under unsaturated
models (Mou and Chu 1981; Pedarla et al. 2012; Tu and conditions, is the modulus of elasticity under saturated
Vanapalli 2016; Erzin and Erol 2007; Fredlund, Rahardjo, and conditions, is the matric suction, is the
Fredlund 2012; Singhal, Houston, and Houston 2015). These atmospheric pressure (101.3 kPa), and S is the degree of
models have shown that the matric suction (negative pore water saturation. and are the fitting parameters. Based on the
pressure) is the most relevant parameter to characterize the literature, the range of parameter (k2) is taken as 1000 kPa/m to
swelling behaviour of soil and the maximum potential swelling 20000 kPa/m for the analysis of the proposed model. However,
pressure depends on the initial condition of the soil such as the the value of k2 can be further increased by various stabilization
initial matric suction, dry density, degree of saturation etc. methods
(Mallik, Chandra, and Singh 2006) Tu (2015) recognized that
there is a strong relationship between the Soil-Water
Characteristic Curve (SWCC) and the variation of swelling

58
JOURNAL OF APPLIED ENGINEERING SCIENCES VOL. 14(27), ISSUE 1/2024
ISSN: 2247-3769 / e-ISSN: 2284-7197 ART.NO. 395 pp. 55-64

is minimal for saturated soil in its initial state. This concept also
2.4.3 Other design parameters. Other design parameters supports the pre-wetting of soil approach to mitigate swelling
considered for the analysis are mentioned below: pressure after post-construction.
Design wheel load (Q) = 50 kN
Length of the upper beam (L) = 1.875 m (Half-width of Sr. Soil Properties Unit Value Reference
pavement) No.
Width of the beam (b) =1 m (Plane strain condition) 1. Liquid Limit (LL) % 85
Modulus of Elasticity of upper beam (E1) = 30000 MPa (Chavan
2. Plastic Limit (PL) % 36
Modulus of Elasticity of lower beam (E2) = 60 MPa - 200 MPa and
3. Plasticity Index (PI) 49
Thickness of upper beam (t) = 0.16m, 0.2 m, 0.25m Bhosale
( The range of thickness of the upper beam 0.16m, 0.2 m and 4. Sand content % 15
5. % clay <0.002mm % 30.16 2022)
0.25 m, is based on the design of a low-volume road for daily
traffic of 50 CVPD, 150 CVPD and 450 CVPD respectively, 6. Fines content % % 83.26
considering wheel loading and temperature stresses (IRC:SP62 passing #No.200
2014). ASTM sieve (IS 75µ
Unit weight of base layer (γ) = 14 kN/m3, 16 kN/m3, 18 kN/m3 sieve)
Depth of placement of lower beam = 0.1m, 0.3m, 0.5m, 1.0 m, 7 Void Ratio (e0) 1.13
1.5m Table 2. Index properties of expansive soil at village Ravet,
Pune, India
Table 1 shows the range of values for various non-dimensional
parameters based on the above design parameters, considered
for the analysis. An algorithm is developed in MATLAB to (a)
perform a parametric study.

Sr. Range of
Parameter Symbol
No. values
1 Relative flexural rigidity of
R= E1I1/ E2I2 150-500
upper and lower beams
Relative modulus of upper
2 r= k1 / k2 1- 50
and lower soil layer
Normalised unit weight of
3 γ′ 0.28- 0.36
granular soil
Normalised depth of (b)
4 placement of the h′ 0.1-1.5
geosynthetic layer
5 Normalised swelling
P′s 1-10
Pressure
Normalised self-weight of
6 w′ .077-0.12
the pavement slab

Table 1. The range of values of non-dimensional parameters

considered for the analysis

3. CASE STUDY Figure 4 (a) Soil-Water Characteristic Curve and (b) Swelling
Pressure Vs Initial suction Curve for expansive soil at village
In this section, a case study of expansive soil at the village Ravet, India
Ravet, Pune, India is discussed, illustrating the impact of initial
soil suction on soil swelling pressure. Table 2. presents grain-
size distribution data and index properties of soil, which are 4. RESULTS AND DISCUSSION
utilized in the generation of the Soil-Water Characteristic Curve
(SWCC), depicted in Fig. 4(a), (Fredlund and Xing 1994). The Though the Winkler model shows limitations such as
Swelling pressure Initial Suction Curve is then derived as discontinuity in the response of the model at the edge of the
shown in Fig. 4(b), considering the computed SWCC. (Tu and beam, it is still used in the design of rigid pavement because of
Vanapalli (2016). its simplicity. Despite the limitations, the present research is
more general and takes care of the practical values of
In Figure 4(b), it is observed that the swelling pressure reaches a parameters. Some typical results have been presented in this
maximum value of 190 kPa when the soil is initially in its driest section, to understand the influence of various parameters on the
state, corresponding to the maximum initial soil suction and response of the pavement system.
then gets fully saturated progressively (until suction approaches
zero). Conversely, if the initial soil suction is 1000 kPa, as Note that the word ‘deflection’ used in the discussion below
depicted in Fig. 4(b) for an initial volumetric water content of indicates the resultant positive deflection (heave) unless
38%, referring to Fig 4(a), the maximum value of swelling specified.
pressure decreases to 24.5 kPa. Similarly, the swelling pressure

59
JOURNAL OF APPLIED ENGINEERING SCIENCES VOL. 14(27), ISSUE 1/2024
ISSN: 2247-3769 / e-ISSN: 2284-7197 ART.NO. 395 pp. 55-64

4.1 Effect of the depth of placement of a lower beam on the 4.2 Effect of relative flexural rigidity of the upper and
deflection of beams lower beam on the deflection of beams

Fig. 5(a) and 5(b), show the influence of the depth of placement Fig. 6(a) and 6(b), show the influence of the relative flexural
of the lower beam , on the normalized deformation of the rigidity of the upper and lower beams, on the normalized
upper and lower beam respectively, for the non-dimensional deformation of the upper and lower beam respectively, for the
parameters =2, = 150, = 2, = 0.32, = 0.096. non-dimensional parameters =2, = 0.5, = 2, = 0.32,
 As the parameter increases from 0.1 to 1.5, the = 0.096.
normalised deflection of the upper beam decreases  It is observed that the decrease in the value of
around 29 % at the edge and 32 % at the centre of the parameter from 500 to 150, does not affect the
beam. normalised deflection of the upper beam. For the
 In the case of the lower beam for the same parameters, lower beam as the parameter decreases from 150 to
when increases from 0.1 to 1.5, the maximum 500 deflection increases by 26 % at the centre and the
normalised deflection for the lower beam is found to edge of the beam. The increase in the swelling
decrease around 27 % at the edge and 29 % at the centre. pressure =1 to 10, show a similar trend of
 It is observed through analysis that when swelling pressure deflection of the upper and lower beam.
is equal to 1, the decrease in the deflection due to the  Results indicate that the decrease in the value of ,
increase in from 0.1 to 1.5 is 112 % at the centre and 79 which means increasing the rigidity of the lower beam
% at the edge of the upper beam. Whereas, for the increase while keeping the rigidity of the upper beam constant,
in swelling pressure to =10, the decrease in the is not found significant to reduce the deflection of the
deflection due to an increase in is about 4.7 % at the upper beam. This may be due to the presence of a
centre and edge of the upper and lower beam. more rigid upper beam, the ‘beam effect’ of
 The above result suggests that as the depth of placement of geosynthetic reinforcement does not get mobilised
the lower beam increases, deflection decreases due to the enough under loading and swelling stresses. This
increase in the weight of the surcharge. means the lower beam just adds the surcharge weight
over the swelling pressure without providing the
bending stiffness. However, this needs to be verified
with experimental studies.

Figure 5. (a) Effect of depth of placement of lower beam

on the deflection of upper beam (b) Effect of depth of
placement of lower beam on the deflection of lower beam

Figure 6. (a) Effect of relative flexural rigidity of beams on the

deflection of the upper beam (b) Effect of relative flexural
rigidity of beams on the deflection of lower beam

60
JOURNAL OF APPLIED ENGINEERING SCIENCES VOL. 14(27), ISSUE 1/2024
ISSN: 2247-3769 / e-ISSN: 2284-7197 ART.NO. 395 pp. 55-64

4.3 Effect of the relative stiffness of base layer and  For the upper beam, the maximum positive bending
subgrade on the deflection and bending moment of the moment is seen at the centre for =50 and gradually
beams
decreases to zero at the edge of the beam. As the ratio
The effect of variation of the relative stiffness of the base layer increases from 1 to 50, the maximum normalised
and subgrade soil , on the normalized deflection of the upper bending moment at the centre of the upper beam
beam and the lower beam , is shown in Fig. 7(a) and 7(b) increases by 11 %.
respectively, for non-dimensional parameters =2, = 0.5,  For the Lower beam, the maximum normalised
=150, = 0.32, = 0.096. Fig. 8(a) and 8(b) show the effect
of relative stiffness r, on the normalized bending moment of the bending mment is observed at the centre for =2
upper and lower beam respectively. and gradually decreases and becomes zero at the edge
of the beams. As the ratio increases from 2 to 50,
the maximum normalised bending moment at the
centre of the lower beam decreases by 78 %.
 The results show that the relative stiffness of the base
layer and subgrade soil significantly affects the
response of upper and lower beams.

Figure 7. (a) Effect of the relative stiffness of upper and lower

soil on the deflection of upper beam (b) Effect of the relative
stiffness of upper and lower soil on the deflection of lower
beam

 As the value of r decreases from 50 to 1, the

normalised deflection of the upper beam decreases
by 98 % at the centre and edge of the beam whereas Figure 8. (a) Effect of the relative stiffness of upper and lower
soil on the bending moment of the upper beam (b) Effect of the
the normalised deflection of the lower beam
relative stiffness of upper and lower soil on the bending moment
decreases around 69% at the centre and 66% at the of the lower beam
edge of the lower beam. The decrease in percentage
remains almost the same with an increase in swelling 4.4 Effect of self-weight of the upper beam on the
pressure from 1 to 10. deflection of beams
 It is also found that the percentage variation in the
relative stiffness of the base layer soil and subgrade The increase in the thickness of the upper beam (pavement slab)
is related to an increase in the daily CVPD and temperature
soil is directly proportional to the percentage variation
differentials. The change in the self-weight of the upper beam
in the normalised deflection of the upper beam with can be achieved by changing the thickness of the upper beam
the constant of proportionality approximately equal to from 0.16 m to 0. 25 m (IRC:SP62 2014), which also changes
1. This means that if the value of parameter the rigidity and characteristic length of the upper beam which
increases by 30%, the deflection of the upper beam affects all the other non-dimensional parameters. Therefore the
will increase by around 30 %. analysis is performed while keeping the characteristic length R1
equal to 1. Fig. 9(a) and 9(b) show the influence of the self-

61
JOURNAL OF APPLIED ENGINEERING SCIENCES VOL. 14(27), ISSUE 1/2024
ISSN: 2247-3769 / e-ISSN: 2284-7197 ART.NO. 395 pp. 55-64

weight of the upper beam w′, on the normalized deflection of and 2.8 % at the edge of the beam. Whereas for the
the upper and lower beam respectively for the non-dimensional lower beam the deflection is about 2.8% at the
parameters =2, = 2, = 0.5, =150, = 0.32, centre and 2.4% at the edge of the beam.
 Figures show that with the increase of normalised  With the increase in swelling pressure to 10, the
self-weight of the upper beam w′ from 0.077 to 0.12, decrease in the deflection of the upper and lower
the normalised deflection of the upper beam beam is around 0.5 % at the centre and the edge of
decreased by around 4 % at the centre and 3.5 % at both the upper and lower beam.
the edge of the beam whereas, the normalized
deflection of the lower beam is decreased by 3.6% The results for the higher swelling pressure, increasing the unit
at the centre and 3.3% at the edge of the beam. weight of base course soil have a negligible impact on the
 It is observed that as the swelling pressure deflection of upper and lower beams
increases to 10, the decrease in the deflection due to
the increase in the self-weight of the upper beam from
0.077 to 0.12 is found to be around 1 % at the centre
and edge of the upper and lower beam.

Figure 10. (a) Effect of unit weight of base-layer soil on the

deflection of the upper beam (b) Effect of unit weight of base-
layer soil on the deflection of the upper beam

Figure 9. (a) Effect of self-weight of the upper beam on the

deflection of the upper beam (b) Effect of self-weight of the 5 CONCLUDING REMARKS
upper beam on the deflection of the lower beam
Based on the results and discussion presented in the earlier
It can be concluded from the result that in the case of higher section following generalized conclusions have been drawn.
swelling pressure increasing the self-weight of the  Maximum deflection (heave) in the upper and lower
upper beam is not very significant. beam is seen at the edge of the upper and lower beam.
 The relative stiffness of the base layer soil and
4.5 Effect of unit weight of base layer soil on the deflection subgrade soil ( ), is found to be the most significant
of beams parameter affecting the response of the model. It is
observed that the relative stiffness of the base layer
Fig. 10(a) and 10(b) show the effect of the unit weight of base
soil and subgrade soil is directly proportional to the
course soil, on the normalized deflection of the upper and lower
beam for the parameters =2, = 0.5, = 150, = 2, = percentage variation in the normalised deflection of
0.096 respectively. the upper beam with the constant of proportionality
 Figures show that with the increase in the value of approximately equal to 1.
from 0.28 to 0.36, the normalised deflection of the  The relative flexural rigidity of the upper and lower
upper beam decreases by around 3 % at the centre beam ( does not affect the flexural response of the

62
JOURNAL OF APPLIED ENGINEERING SCIENCES VOL. 14(27), ISSUE 1/2024
ISSN: 2247-3769 / e-ISSN: 2284-7197 ART.NO. 395 pp. 55-64

upper beam. This may be due to the presence of a by Suction Methods.” Engineering Geology 92 (3–4): 133–45.
more rigid upper beam, the ‘beam effect’ of https://doi.org/10.1016/j.enggeo.2007.04.002.
geosynthetic reinforcement does not get mobilised Fredlund, Delwyn G., Harianto Rahardjo, and Murray D.
enough under loading and swelling stresses. Fredlund. 2012. Unsaturated Soil Mechanics in Engineering
 The significance of the parameters to decrease the Practice. Unsaturated Soil Mechanics in Engineering Practice.
deflection of the upper beam is recognized as >> https://doi.org/10.1002/9781118280492.
>> >> >> , i.e. is the most significant
Fredlund, Delwyn G, and Anqing Xing. 1994. “Equations for
parameter whereas is the least significant
the Soil-Water Characteristic Curve.” Canadian Geotechnical
parameter. This suggests that improving the stiffness Journal 31 (4): 521–32.
of the soil is a better option than the inclusion of
reinforcement at the interface of the base layer and Hetényi, M. 1946. Beams on Elastic Foundation: Theory with
expansive soil. Applications in the Fields of Civil and Mechanical Engineering.
University of Michigan Studies: Scientific Series. University of
 The non-dimensional solution presented herein can
Michigan Press, Ann Arbor, MI.
help design low-volume concrete roads in a wet or https://books.google.co.in/books?id=FitSAAAAMAAJ.
temperate climate zone, where swelling is a more
predominant mode of failure than the shrinkage of IRC:SP62. 2014. “Guidelines for the Design and Construction
soil. of Cement Concrete Pavement for Low Volume Roads.” In
Indian Road Congress, New Delhi.

5.1 Acknowledgements Khan, Md Adnan, and Jay X. Wang. 2017. “Application of

Euler-Bernoulli Beam on Winkler Foundation for Highway
The authors would like to acknowledge the Director, COEP Pavement on Expansive Soils,” 228–39.
Technological University and Technical Education Quality https://doi.org/10.1061/9780784481691.023.
Improvement Program [TEQIP]-Phase III (The World Bank
Assisted Project) for providing research facilities for the Khan, Md Adnan, Jay X. Wang, and Debojit Sarker. 2020.
research paper “Development of Analytic Method for Computing Expansive
Soil–Induced Stresses in Highway Pavement.” International
References Journal of Geomechanics 20 (2): 04019160.
https://doi.org/10.1061/(asce)gm.1943-5622.0001511.
Adem, Hana H., and Sai K. Vanapalli. 2015. “Prediction of the
Modulus of Elasticity of Compacted Unsaturated Expansive Khan, Md Ashrafuzzaman, Nripojyoti Biswas, Aritra Banerjee,
Soils.” International Journal of Geotechnical Engineering 9 and Anand J. Puppala. 2020. “Field Performance of Geocell
(2): 163–75. Reinforced Recycled Asphalt Pavement Base Layer.”
https://doi.org/10.1179/1939787914Y.0000000050. Transportation Research Record 2674 (3): 69–80.
https://doi.org/10.1177/0361198120908861.
Aitchison, G.D., and J.W. Holmes. 1965. “Suction Profiles in
Soils Beneath Covered and Uncovered Areas.” Kief, Ofer, D Ph, and Double K Geotechnical. 2016. “Rail
INTERNATIONAL SOCIETY FOR SOIL MECHANICS AND Track Pavements on Expansive Clay Restrained by Hybrid
GEOTECHNICAL ENGINEERING, 187–91. Geosynthetic Solution.”
https://doi.org/10.1007/978-3-319-73568-9_174.
Lytton, R., C. Aubeny, and R. Bulut. 2006. “Design Procedure
Al-Omari, Raid R, and Faris J Hamodi. 1991. “Swelling for Pavements on Expansive Soils. Volume II” 7 (2).
Resistant Geogrid—A New Approach for the Treatment of
Lytton, R L. 1978. “CHARACTERIZATION OF EXPANSIVE
Expansive Soils.” Geotextiles and Geomembranes 10 (4): 295–
SOILS IN ENGINEERING.” In TRANSACTIONS-AMERICAN
317.
GEOPHYSICAL UNION, 59:243. AMER GEOPHYSICAL
Chavan, Ashwini, and Sukhanand Bhosale. 2022. “Estimation UNION 2000 FLORIDA AVE NW, WASHINGTON, DC
of Suction Profile and Vertical Deformation in the Indian 20009.
Expansive Soil Using Thornthwaite Moisture Index.”
Lytton, Robert, Charles Aubeny, and Rifat Bulut. 2004. “Design
Proceedings of the 7th World Congress on Civil, Structural, and
Procedure for Pavements on Expansive Soils. Volume I.” Texas
Environmental Engineering.
Deparment of Transportation 7 (FHWA Report No. 0-4518-1
Chittoori, Bhaskar C.S., Debakanta Mishra, and Kazi Moinul Vol. 2): 232.
Islam. 2018. “Forensic Investigations into Recurrent Pavement
Maheshwari, P, P K Basudhar, and S Chandra. 2004. “Analysis
Heave from Underlying Expansive Soil Deposits.”
of Beams on Reinforced Granular Beds.” Geosynthetics
Transportation Research Record 2672 (52): 118–28.
International, no. 6: 470–80.
https://doi.org/10.1177/0361198118758625.
Maheshwari, Priti. 2008. “Analysis of Beams on Tensionless
Covar, Andrew P., and Robert L. Lytton. 2001. “Estimating Soil
Reinforced Granular Fill-Soil.” International Journal for
Swelling Behavior Using Soil Classification Properties.”
Numerical and Analytical Methods in Geomechanics 32 (March
Expansive Clay Soils and Vegetative Influence on Shallow
2007): 189–213. https://doi.org/10.1002/nag.
Foundations, 44–63. https://doi.org/10.1061/40592(270)3.
Maheshwari, Priti, P. K. Basudhar, and S. Chandra. 2006.
Erzin, Yusuf, and Orhan Erol. 2007. “Swell Pressure Prediction

63
JOURNAL OF APPLIED ENGINEERING SCIENCES VOL. 14(27), ISSUE 1/2024
ISSN: 2247-3769 / e-ISSN: 2284-7197 ART.NO. 395 pp. 55-64

“Modeling of Beams on Reinforced Granular Beds.” Sarker, Debojit, Jay X Wang, and Md Adnan Khan. 2019.
Geotechnical and Geological Engineering 24 (2): 313–24. “Development of the Virtual Load Method by Applying the
https://doi.org/10.1007/s10706-004-7548-z. Inverse Theory for the Analysis of Geosynthetic-Reinforced
Pavement on Expansive Soils.” In Geo-Congress 2019:
Maheshwari, Priti, and M. N. Viladkar. 2009. “A Mathematical Geotechnical Materials, Modeling, and Testing, 326–39.
Model for Beams on Geosynthetic Reinforced Earth Beds under American Society of Civil Engineers Reston, VA.
Strip Loading.” Applied Mathematical Modelling 33 (4): 1803–
14. https://doi.org/10.1016/j.apm.2008.03.009. Singhal, Sonal, Sandra L Houston, and William N Houston.
2015. “Swell Pressure, Matric Suction, and Matric Suction
Mallik, A. K., Sarvesh Chandra, and Avinash B. Singh. 2006. Equivalent for Undisturbed Expansive Clays.” Canadian
“Steady-State Response of an Elastically Supported Infinite Geotechnical Journal 52 (3): 356–66.
Beam to a Moving Load.” Journal of Sound and Vibration 291
(3–5): 1148–69. https://doi.org/10.1016/j.jsv.2005.07.031. Standards Australia. 2011. “AS2870-2011 Residential Slabs &
Footings.Pdf.”
Matlock, Hudson, and Lymon C Reese. 1960. “Generalized
Solutions for Laterally Loaded Piles.” Journal of the Soil Tamim, Mir Md. 2017. “Evaluating the Effectiveness of a
Mechanics and Foundations Division 86 (5): 63–92. Hybrid Geosynthetic.” A Thesis Submitted in Partial Fulfillment
of the Requirements for the Degree of Master of Science in Civil
McDowell, Chester, R C Herner, and F L Wooltorton. 1956. Engineering at Boise State University, no. December.
“Interrelationship of Load, Volume Change, and Layer
Thicknesses of Soils to the Behavior of Engineering Tu, Hongyu, and Sai K. Vanapalli. 2016. “Prediction of the
Structures.” In Highway Research Board Proceedings. Vol. 35. Variation of Swelling Pressure and One-Dimensional Heave of
Expansive Soils with Respect to Suction Using the Soil-Water
Mckeen, G R, and G R Mckeen. 1984. “Scholars ’ Mine Retention Curve as a Tool.” Canadian Geotechnical Journal 53
Expansive Soil Pavement Design Using Case Studies,” no. (8): 1213–34. https://doi.org/10.1139/cgj-2015-0222.
May.
Vanapalli, Sai K., and Won Taek Oh. 2010. “A Model for
McKeen, R Gordon. 1978. “Design and Construction of Airport Predicting the Modulus of Elasticity of Unsaturated Soils Using
Pavements on Expansive Soils.” Transportation Research the Soil-Water Characteristic Curve.” International Journal of
Board Special Report, no. 175: 57–59. Geotechnical Engineering 4 (4): 425–33.
https://doi.org/10.3328/IJGE.2010.04.04.425-433.
McKeen, R Gordon, and Debora J Hamberg. 1981.
“Characterization of Expansive Soils.” Transportation Research Vessely, Mark J., and Jonathan T.H. Wu. 2002. “Feasibility of
Record, no. 790. Geosynthetic Inclusion for Reducing Swelling of Expansive
Soils.” Transportation Research Record, no. 1787: 42–52.
Mitchell, Peter W. 1980. “The Structural Analysis of Footings https://doi.org/10.3141/1787-05.
on Expansive Soil.” In Expansive Soils, 438–47. ASCE.
Wang, Jay X, Debojit Sarker, and Berjees Ikra. 2018.
Mou, C H, and T Y Chu. 1981. “Soil-Suction Approach for “Development of a Mechanistic-Based Design Method for
Evaluation of Swelling Potential.” Transportation Research Geosynthetic-Reinforced Pavement on Expansive Soils and
Record 790: 54–60. Prediction of Moisture Content Fluctuations in Subgrades.”
Southern Plain Transportation Center, Norman, OK (2018).
Musa, Abubakr E.S. 2018. “An Analytical Bending Solution for file:///C:/Users/ashwi/Downloads/scholar (6).ris.
Analysis and Design of Long Beams Constructed on the
Expansive Soil.” Arabian Journal for Science and Engineering Zornberg, J. G., and R. Gupta. 2009. “Reinforcement of
43 (10): 5203–14. https://doi.org/10.1007/s13369-017-3041-x. Pavements over Expansive Clay Subgrades.” Proceedings of the
17th International Conference on Soil Mechanics and
Payne, David C., and Don A. Cameron. 2014. “The Walsh Geotechnical Engineering: The Academia and Practice of
Method of Beam-on-Mound Design from Inception to Current Geotechnical Engineering 1: 765–68.
Practice.” Australian Journal of Structural Engineering 15 (2): https://doi.org/10.3233/978-1-60750-031-5-765.
177–88. https://doi.org/10.7158/S13-008.2014.15.2.
Zornberg, J. G., R. Gupta, and J. A.Z. Ferreira. 2010. “Field
Pedarla, Aravind, Anand J. Puppala, Laureano R. Hoyos, Sai K. Performance of Geosynthetic Reinforced Pavements over
Vanapalli, and Claudia Zapata. 2012. “SWRC Modelling Expansive Clay Subgrades.” 9th International Conference on
Framework for Evaluating Volume Change Behavior of Geosynthetics - Geosynthetics: Advanced Solutions for a
Expansive Soils.” Unsaturated Soils: Research and Challenging World, ICG 2010, 1481–84.
Applications C: 221–28. https://doi.org/10.1007/978-3-642-
31116-1_30.

Perera, Yugantha Y., Claudia E. Zapata, William N. Houston,

and Sandra L. Houston. 2004. “Long-Term Moisture Conditions
under Highway Pavements” 40744 (August 2015): 1132–43.
https://doi.org/10.1061/40744(154)102.

Post-Tensioning, Institute. 2003. “Alternate Procedures for

Determining Soil Support Parameters for Shallow Foundations
on Expansive Clay Soil Sites.”

64