International Journal of GEOMATE, Jan., 2022, Vol.22, Issue 89, pp.

80-86
ISSN: 2186-2982 (P), 2186-2990 (O), Japan, DOI: https://doi.org/10.21660/2022.89.gxi361
Geotechnique, Construction Materials and Environment

PERFORMANCE OF REINFORCED BASE COURSES OF FLEXIBLE

PAVEMENTS OVERLYING SOFT SUBGRADES: INSIGHTS FROM
LARGE-SCALE MODEL EXPERIMENTS

Ramu Baadiga1, *Umashankar Balunaini2, and Sireesh Saride3

1
Doctoral Student, Indian Institute of Technology Hyderabad, India; 2,3 Faculty, Indian Institute of
Technology Hyderabad, India

*Corresponding Author, Received: 06 Oct. 2021, Revised: 01 Dec. 2021, Accepted: 09 Jan. 2022

ABSTRACT: In this study, large-scale model pavement experiments (LSMPEs) were performed to assess the
performance of geosynthetic-reinforced bases of flexible pavements overlying soft subgrades (with California
bearing ratio equal to 1% and 5%). Two configurations of reinforcement were considered – (1) base layers
reinforced with geocell (GC) alone, and (2) base layers reinforced with geocell and geogrid (GG) combinations
(GC+GG). Model pavements were constructed inside the test chamber measuring 1.5m (length) × 1.5m (width) ×
1.0m (height). Wheel loads on the pavements were simulated by applying repetitive load equivalent to the tire
contact pressure of 550 kPa through an actuator on a rigid circular plate (diameter equal to 300 mm). The
reinforcement combination with GC+GG was found to perform better than GC alone due to additional support
offered by geogrid when used as a base or basal reinforcement underneath the geocell mattress. The performance
was evaluated in terms of cumulative permanent deformations and Traffic Benefit Ratio (TBR) over 100,000
repetitive load cycles. TBRs of the GC+GG combination showed values greater than 50 for test conditions
considered. Furthermore, test results showed that the GC+GG combination reduced permanent deformation by up
to 78%, while GC alone reduced it by about 52% compared to the unreinforced section. TBRs of geocell alone
and geocell with basal geogrid combinations indicated extended service life of the pavement.

Keywords: Soft subgrade, Geocell, Geogrid, Permanent deformations, Traffic benefit ratio

1. INTRODUCTION placed to arrest downward deflections and provide

extra support [8, 12, 13].
Pavements overlying soft subgrades cause The geogrid basal reinforcement is used to
excessive permanent deformation under repetitive increase the pavement's service life, decreasing base
loads. The subgrades associated with wetlands and course thickness or arresting longitudinal cracks in
heavy rainfall regions are a practical concern for the pavements [14]. The published studies suggest
improvement. Available conventional methods such significant improvement in load-bearing capacity due
as cement or lime stabilization or fiber reinforcement to additional support with basal geogrid placed
used to treat soft subgrades yield a good platform for underneath geocell [8, 12]. The overall performance
the construction of pavement layers but consume of the basal reinforcement accounts due to the tensile
enormous time and money. Alternatively, strength mobilization and membrane support [8, 13].
geosynthetics are highly preferred due to their cost- Previous researchers indicated the necessity of
effectiveness and ease of construction. Geosynthetic additional support with basal geogrid underneath the
reinforcement materials, viz., geogrid, and geocell geocell mattress to increase the overall performance
offer various benefits in flexible pavements such as of the pavement [12, 13]. On the contrary, its benefit
lateral restraint, load-carrying capacity, and was reported to be marginal with an increase in the
membrane support [1-4]. Similarly, a three- height of the geocell and was found to give superior
dimensional honeycomb structure called geocell performance when the limited size of geocell and
offers overall confinement and higher load carrying appropriate geogrid is used [15]. The increase in
capacity [1, 5-7]. However, their benefit in terms of height was reported to show buckling of geocell walls
reduced deformations and improved pavement leading to a decrease in the load-bearing capacity [15].
performance over soft subgrades is observed when The studies mentioned herewith have focused
reinforcement is placed near the loading region [8-11]. solely on evaluating the bearing capacity of
In most cases, when soft subgrades are encountered, reinforced foundations or pavements. However,
single reinforcement such as geogrid or geocell studies have not focused on quantifying the overall
placement in the base layers may not be adequate to pavement performance of geocell and geogrid basal
control the pavement deformations. Hence, to reinforcement, such as improved service life,
overcome such problems generally, a geogrid especially when soft subgrades are encountered. The
reinforcement layer underlying a geocell mattress is improved service life of the flexible pavement is

80
 International Journal of GEOMATE, Jan., 2022, Vol.22, Issue 89, pp.80-86

generally quantified in terms of Traffic Benefit Ratio prepared at compacted dry unit weights and molding
(TBR) [1]. TBR is defined as the ratio of several load water contents of 16.55 kN/m3 and 18.5%, and 17.40
cycles sustained by base reinforced pavement at the kN/m3 and 17.2%, respectively. The base and subbase
defined permanent deformation to the number of load layers were prepared following the gradation of wet-
cycles sustained by the unreinforced pavement mix macadam with crushed stone aggregates, meeting
section at the same defined permanent deformation the standard specifications mentioned in the
with the same material constitutions and thicknesses. American Society for Testing and Materials [16] and
Permanent deformation is also called cumulative Ministry of Road Transportation and Highways [17],
permanent deformation or rutting. Thus, in this study, Government of India. Aggregates used for base and
the influence of bases reinforced with geocell alone subbase layers had a maximum dry unit weight of
and combination of geocell and geogrid is evaluated 22.7 kN/m3 and optimum water content of 5.5%. The
in terms of cumulative permanent deformation and asphalt material used in the study as a binder course
TBR of reinforced pavements overlying soft to had a compacted unit weight of 23.4 kN/m3.
relatively fair subgrades (with California bearing
ratio (CBR) equal to 1% and 5%). The overall 3.2 Geocell and Geogrid
purpose of this study is to quantify the enhanced
pavement performance in terms of increased service High-density polyethylene (HDPE) geocell (GC)
life by the inclusion of geosynthetics when soft of height equal to 100 mm with weld spacing of 356
subgrades with a CBR ≤ 5% are encountered. mm and a seam peel strength of 1420 N was used. The
geocell used in the study had an expanded cell length
2. RESEARCH SIGNIFICANCE of 224 mm and an expanded cell width of 259 mm.
Geogrid (GG), made of polypropylene with peak
This article details the large-scale model tensile strength of 30 kN/m in biaxial directions and
pavement experiments (LSMPEs) conducted on aperture size equal to 40 mm × 40 mm, was used as
geogrid, and a combination of geogrid and geocell basal reinforcement placed right below the geocell
(basal reinforcement application) reinforced flexible mattress. The average rib and bar width of geogrid
pavements overlying soft (CBR=1%) to relatively fair was 3.0 and 3.77 mm, respectively. The average node
(CBR=5%) subgrades. The use of geocell and basal thickness of geogrid was 4.0 mm. The geogrid had a
reinforcement application in base layers overlying stiffness of 503 kN/m × 509 kN/m at 2% tensile strain
soft subgrade is examined. The prepared test sections in the biaxial directions. The geogrid and geocell used
were subjected to repetitive load action by applying a in the study are the typical ones used in the field for
typical tire contact pressure of 550 kPa through a reinforcing base layers. Figures 1 (a) and (b) show the
double-acting linear dynamic actuator. The benefit of geocell and geogrid used in the study.
reinforcement in the base layers was quantified in
terms of traffic benefit ratios (TBRs) for overlaying
soft to relatively fair subgrade conditions. The input
range of TBRs for the design of flexible pavements is
recommended to extend the pavement's service life.

3. MATERIALS AND METHODOLOGY

3.1 Subgrade Soil, Granular Base and Subbase,

and Asphalt Layer

In this testing program, crushed aggregate, dense

bituminous macadam, and clayey sand were used to
prepare granular layers, asphalt layer, and subgrade,
respectively. Similarly, geocell and geogrid made up
of high-density polyethylene and polypropylene are
selected for base reinforcement.
Locally available clayey sand was used as the
subgrade material. The Standard Proctor test results
showed a maximum dry unit weight of 19 kN/m3 and
an optimum water content of 14.5%. Soft and
relatively fair subgrade conditions (CBR=1% and
5%) were simulated in the test chamber by varying
water content and the compacted unit weight. Based
on extensive calibration studies, targeted CBR =1%
and 5% were achieved for the chosen subgrade when

81
 International Journal of GEOMATE, Jan., 2022, Vol.22, Issue 89, pp.80-86

was ensured for proper consistency in unit weight and

water content for all the test cases; the maximum
deviation in the measured properties was within 3%.
As shown in Fig. 2, a repetitive load of haversine load
pattern with 5 Hz frequency was simulated and
applied over the prepared bed with a typical tire
contact pressure of 550 kPa using 100 kN actuator on
a 300 mm rigid circular plate placed on the center of
the testbed [2, 3]. Figure 3 presents the typical load
pattern applied on the prepared testbed. The
permanent deformation data was measured using an
inline displacement transducer inbuilt with the
actuator system (refer to Fig. 2).

Fig.1 Photo snaps of geocell and geogrid used in the

study (a) geocell and (b) basal geogrid used
underneath the geocell mattress

4. EXPERIMENTAL PROGRAM

Pavement layers were prepared using a vibratory

plate compactor inside the test chamber measuring
1.5 m × 1.5 m × 1.0 m [3]. The total thickness of
granular layers (base plus subbase) and subgrade
were maintained as 600 mm and 400 mm,
respectively, overlying a soft subgrade of CBR=1%.
In the case of pavements underlying relatively fair
subgrade with CBR=5%, a total thickness of granular
and subgrade layer was selected as 440 mm and 500
mm, respectively. These thicknesses were finalized Fig. 2 Test setup showing prepared testbed system
based on Indian Road Congress guidelines for the with actuator mounting
design of flexible pavements [18]. The base layer
thickness of 225 mm was maintained commonly for
all the pavement sections. Over the prepared granular
layer, a 50 mm thick asphalt layer was placed.
Pavement layers were prepared to achieve 98% of
their targeted densities. The expanded geocell
mattress size equal to 1.45 m × 1.45 m in length and
width was placed at 0.15D from the top of the base
layer, where D is the diameter of the circular plate
(i.e., 300 mm). In the case of basal reinforcement
application, geogrid of size equal to 1.45 m × 1.45 m
in length and width was placed right below the
geocell mattress. Prepared testbed with complete
instrumentation is shown in Fig. 2. In total, six large-
scale model pavements were conducted. Two sections
unreinforced, two test sections with geocell alone,
and two test sections with geocell and geogrid as a Fig. 3 Repetitive load pattern applied on the testbed
basal reinforcement were tested. The prepared testbed

82
 International Journal of GEOMATE, Jan., 2022, Vol.22, Issue 89, pp.80-86

5. RESULTS AND DISCUSSION combination of geocell and basal geogrid. Based on

the permanent deformations presented in Fig. 4(b), it
5.1 Influence of Single Geocell and Combination can be inferred that the reductions were in the order
of Geocell and Basal Geogrid On Permanent of 51% and 66% for GC and GC+GG, respectively,
Deformations over 100,000 load cycles. In the case of pavements
encountered with relatively fair subgrade conditions,
Figures 4(a) and (b) present the permanent geocell alone might be used to reinforce the base
deformation vs. load cycles for geocell alone and layers as no significant improvement in reducing the
geocell plus geogrid reinforced pavement bases deformations was observed for GC+GG.
overlying soft (CBR=1%) and relatively fair subgrade
(CBR=5%), respectively. In all test cases, repetitive
loads were applied up to 100,000 load cycles (N) due
to test constraints related to the enormous amount of
time it takes to apply repetitive loads. Figure 4(a)
shows that the unreinforced section exhibits more
than 20 mm of permanent deformation; thus
exceeding the failure criterion in rutting. The
relatively high permanent deformation of the
unreinforced pavement section might have occurred
to large vertical and lateral deformations of granular
material overlying soft subgrade conditions under
repetitive loads. During initial repetitive load cycles,
reinforcement mobilization is low; however, as the
load cycles increase further, deformations were
drastically reduced for GC alone and GC+GG, as
shown in Fig. 4(a). The superior benefit is witnessed
for the GC+GG combination. For example, at
100,000 load cycles, unreinforced section, GC, and
GC + GG showed deformations of 24 mm, 11 mm,
and 5 mm, respectively. Compared to an unreinforced
section, a reduction in measured deformations was
found to be about 54% and 73% corresponding to GC
and GC+GG at the end of load cycles, respectively.
In the case of pavement section reinforced with
geocell alone, reduced deformation might have
resulted due to overall confinement. In the case of the
GC+GG reinforcement combination, it is evident that
due to the extra structural support offered through
membrane and interlock action of basal geogrid,
further reduction in permanent deformation was
observed. Beyond the 30,000 load cycles, a near-
constant permanent deformation was seen with the
GC+GG reinforcement combination. Therefore,
Fig. 4 Permanent deformation versus repetitive load
geocell with basal geogrid combination might be a
cycles of unreinforced, GC reinforced base layer and
viable solution to control deformation over very soft
GC+GG reinforced base layer for (a) pavement
subgrades.
overlying CBR = 1%, and (b) pavement overlying
On the other hand, permanent deformations
CBR = 5%
observed for pavements overlying relatively fair
subgrade condition is presented in Fig. 4(b). In
5.2 Traffic Benefit Ratios (TBR) of Reinforced
pavements underlying fair subgrade condition
Base Layers
(CBR=5%), the resulting deformations were lower
than those for a soft subgrade (CBR=1%) for a given
TBR is also called traffic improvement life, which
applied load cycle. Between GC and GC+GG, the
indicates the extended life of the pavement. Figures
reinforcement benefit was found to be only marginal.
5(a) and (b) show the TBRs of reinforced pavements
Nevertheless, the GC+GG reinforcement
overlying soft and relatively fair subgrade conditions.
combination performed better showing lower
In this study, about 11 mm maximum permanent
deformations than the geocell alone. Almost
deformations were observed for reinforced sections.
consistent deformations were observed beyond
Hence, the TBRs were calculated within specified
30,000 load cycles for geocell alone and a
permanent deformation; however, to witness TBRs at

83
 International Journal of GEOMATE, Jan., 2022, Vol.22, Issue 89, pp.80-86

higher deformation, might require higher repetitive 6. RECOMMENDATIONS

load cycles than those applied in the study. From Figs. The combination of geocell and geogrid placed
5(a) and (b), it can be noticed that pavements below the geocell was found to be a viable solution
overlying soft and relatively fair subgrade conditions, over soft subgrades. The reinforcement benefit can be
GC reinforced base sections resulted in TBR as high observed majorly in two ways. Firstly, the reduced
as 12 and 8.3, respectively. Likewise, TBRs of bases cost due to a decrease in thickness of a granular layer,
reinforced with GC+GG combination was found to be and secondly, increased service life and reduced
as high as 150 and 38 over soft and relatively fair maintenance cost if the same thickness of the
subgrade condition, respectively (refer to Fig. 5(b)). pavement is maintained. On the contrary, the
Initially, TBRs increase, and then with an increase in unreinforced pavement section frequently requires
further permanent deformations, TBRs attenuates for rehabilitation and asphalt overlays leading to huge
GC reinforced section alone. However, this trend is cost incurrences; this activity might be minimal in the
expected as TBR is calculated as a ratio of reinforced case of reinforced pavements. Over relatively fair
cycles to unreinforced cycles at the same defined subgrade conditions, geocell alone might help in
permanent deformation. In the case of the GC+GG improving pavement performance.
reinforced section, resulting load-carrying cycles It can be noted that in the present study, reported
were more, thereby leading to higher TBR than the TBR values ranged as high as 150 corresponding to
GC alone. Higher TBRs were observed in the case of low permanent deformation values (i.e., below 10
pavement overlying soft subgrades, and lower values mm). However, this value may decrease further at
were observed for fair subgrades. In the available higher permanent deformations (rutting) or
literature, TBRs ranged as high as 670 for designated rutting failure criteria. For example, at a
geosynthetic reinforced pavement bases [1]. However, standard rutting criterion of 20 mm [18] of the
TBRs recommended for design at failure pavement overlying subgrade with CBR=1%, the
deformations were reported between 1 to 4 [1]. computed load cycles of unreinforced pavement and
extrapolated load cycles at the same rut depth for GC
[refer to Fig. 4(a)] alone resulted in about 76,593 and
1,014,862 cycles, respectively. Thus, the
corresponding TBR is 13.25. Hence, it is
recommended to adopt the TBR values about the
rutting failure criterion. The design-related aspects
that involve reducing base course and improving the
pavement's service life can be found in Berg et al. [1]
and Holtz et al. [19] studies.

7. CONCLUSIONS

In this study, essential insights obtained from large-

scale model pavement experiments were discussed.
The flexible pavement bases overlying soft
(CBR=1%) to relatively fair subgrade (CBR=5%)
conditions were reinforced with geocell alone (GC)
and a combination of geocell plus geogrid (GC+GG)
to witness the overall pavement performance. The
repetitive load tests were conducted to evaluate the
load-response of geosynthetic reinforced flexible
pavements. The permanent deformation behavior
versus the applied repetitive load cycles was
evaluated for geocell alone, and geocell plus geogrid
combination placed in the base layer of the flexible
pavement overlying soft and relatively fair subgrades.
Further, the extended life of the pavement due to
reinforcement is quantified in terms of traffic benefit
ratio (TBR) by considering permanent deformation
observed over 100,000 load cycles. The main
Fig. 5 TBR versus permanent deformations for (a) conclusions obtained from the present study are
geocell alone for pavement overlying subgrades with discussed below.
CBR=1% and 5%, (b) combination of geocell plus
basal geogrid for pavement overlying subgrades with • The reduction in pavement deformation due to
CBR = 1% and 5% geocell (GC) alone and geocell plus basal

84
 International Journal of GEOMATE, Jan., 2022, Vol.22, Issue 89, pp.80-86

geogrid (GC+GG) ranged as high as 54% and Eng., 2021, 10.1061/(ASCE)MT.1943-

73%, respectively, over soft subgrades 5533.0003972
(CBR=1%). Likewise, the corresponding [4] Saride S. and Baadiga R., New Layer
reductions were 51% and 66% for pavements Coefficients for Geogrid-reinforced
over relatively fair subgrades (CBR=5%). Pavement bases. Indian Geotech. J., Vol.51,
• Geogrid placed right underneath the geocell was No.1, 2021, pp.182 – 196. DOI:
found to arrest lateral movement of granular 10.1007/s40098-020-00484-6
material effectively leading to higher reductions [5] Pokharel S.K., Han J., Manandhar C., Yang
in deformations. In other words, the additional X., Leshchinsky D., Halahmi I. and Parsons
basal reinforcement right underneath the geocell R., Accelerated Pavement Testing of
mattress leads to sustain higher load cycles than Geocell-Reinforced Unpaved Roads over
the pavement base system supported with geocell Weak Subgrade. Transp. Res. Rec., J. Transp.
alone for the same rut deformation. Res. Board., Vol.2204, No.1, 2011, pp.67–75.
• GC+GG combination was more effective than DOI:10.3141/2204-09
GC alone for pavements overlying soft subgrades [6] Han J. and Thakur J. K., Sustainable
(CBR=1%). However, no significant Roadway Construction using Recycled
improvement of GC+GG compared to GC alone Aggregates with Geosynthetics. Sustainable.
was observed for pavements overlying fair Cities Soc., 14, 2014, pp.342–350.
subgrades (CBR=5%). DOI:10.1016/j.scs.2013.11.011
• Higher TBRs were witnessed for a GC+GG due [7] Hegde A., Geocell Reinforced Foundation
to extra support offered by geogrid underneath Beds-Past Findings, Present Trends, and
the geocell. Future Prospects: A State-of-the-Art Review.
• Traffic Benefit Ratio (TBR) of GC alone ranged Constr. Build. Mater., Vol.154, 2017,
as high as 12 and 8.3, respectively, for pavements pp.658–674.
overlying soft and relatively fair subgrades. DOI:10.1016/j.conbuildmat.2017.07.230
• TBRs of geogrid placed underneath a geocell
mattress (GC+GG) showed as high as 150 and 38 [8] Dash S. K., Sireesh S., and Sitharam T. G.,
for pavement overlying soft and relatively fair Model Studies on Circular Footing Supported
subgrades, respectively. However, it may be on Geocell Reinforced Sand underlain by
noted that the reported TBR values correspond to Soft Clay. Geotext. and Geomembr., Vol. 21,
relatively low permanent deformation values (i.e., No.3, 2003, pp.197–219.
within 12 mm). DOI:10.1016/S0266-1144(03)00017-7
[9] Sitharam T. G., Sireesh S., and Dash S. K.,
8. ACKNOWLEDGEMENT Model Studies of a Circular Footing
Supported on Geocell-Reinforced Clay. Can.
The authors would like to thank the National Geotech. J., Vol. 42, No.2, 2005, pp.693–703.
Highway Authority of India for the research grant https://doi.org/10.1139/t04-117
under grant number: NHAI/TIC/R&D/108/2016. [10] Sireesh S., Mole P. A. F., Madhav M. R., and
Kumar R. V., Non-linear Response of
9. REFERENCES Geocell Reinforced Dense Granular Layer
over Weak Soil Under Circular Loading. Int.
[1] Berg R.R., Christopher B.R., and Perkins J. Geotech. Eng., Vol.10, No.1, 2016, pp.23 –
S.W., Geosynthetic Reinforcement of the 30. OI:10.1179/1939787915Y.0000000014
Aggregate Base/subbase Courses of [11] Thakur J. K., Han J., and Parsons R. L.,
Pavement Structures. Geosynthetic Materials Factors Influencing Deformations of
Association White Paper-II, 2000, Roseville, Geocell-Reinforced Recycled Asphalt
Minnesota, USA. Pavement Bases Under Cyclic Loading. J.
[2] Baadiga R., Stride S., Balunaini U., and Mater. Civ. Eng., Vol.29, No.3, 2017, pp.1–
Madhira R. M., Influence of Tensile Strength 12.DOI:10.1061/(ASCE)MT.1943-
of Geogrid and Subgrade Modulus on Layer 5533.0001760
Coefficients of Granular Bases. Transport. [12] Saride S., Sitharam T.G., and Puppala A.J.,
Geotech., Vol.29, 2021, pp.100557, The Function of Basal Geogrids in
DOI:10.1016/j.trgeo.2021.100557 Minimizing Rutting of Geocell Reinforced
[3] Baadiga R., Balunaini U., Saride S., and Subgrades. Geo-Frontiers Congress, 2011,
Madhira R. M., Influence of Geogrid pp.4645–4652. DOI:10.1061/41165(397)475
Properties on Rutting and Stress Distribution [13] Hegde A., and Sitharam T. G., 3-Dimensional
in Reinforced Flexible Pavements under Numerical Modelling of Geocell Reinforced
Repetitive Wheel Loading, J. Mater. Civ. Sand Beds. Geotext. and Geomembr., Vol. 43,

85
 International Journal of GEOMATE, Jan., 2022, Vol.22, Issue 89, pp.80-86

No.2, 2015, pp.171–181. [17] MORTH. Specifications for Road and Bridge
DOI:10.1016/j.geotexmem.2014.11.009 Works, (5th Revision), Ministry of Road
[14] Zornberg J. G., Reinforcement of Pavements Transport & Highways, 2013
over Expansive Clay Subgrades. In Proc. [18] IRC 37. Guidelines for the Design of Flexible
17th Int. Conf. Soil Mech. Geotech. Eng., M. Pavements, (3rd Revision), Indian Road
Hamza et al. (Eds.), 2009, pp. 765 – 768, Congress, 2012
doi:10.3233/978-1-60750-031-5-765 [19] Holtz R.D, Christopher B.R. and Berg R.R.,
[15] Sitharam T. G. and Sireesh S., Effects of Base Geosynthetic Design and Construction
Geogrid on Geocell-Reinforced Foundation Guidelines, Participant Notebook, FHWA
Beds. Geomech. Geoeng., Vol. 1, No.3, 2006, Publication No. FHWA HI -07-092, Federal
pp.207-216, Highway Administration, Washington, D.C,
DOI:10.1080/17486020600900596 2008.
[16] ASTM .2020. Standard Classification for
Copyright © Int. J. of GEOMATE All rights reserved,
Graded Aggregate Material for Bases or
including making copies unless permission is obtained
Subbases for Highways or Airports, ASTM
from the copyright proprietors.
D2940M-20, West Conshohocken, PA:
ASTM.

86