Introduction

Geogrid is one among the major forms of geosynthetics which has been used successfully for reinforcing unbound granular materials. Uniaxial, biaxial, and triaxial are the three common types of geogrid. Uniaxial geogrid has rectangular apertures having tensile strength in the longitudinal direction and are commonly used to reinforce slopes, embankments and retaining walls. Biaxial geogrid has square or rectangular apertures having tensile strength in both longitudinal as well as transverse directions. Triaxial geogrid has triangular apertures having tensile strength in all the three directions. Biaxial and triaxial geogrids are used for reinforcing pavements. The geogrid can be used to improve the pavement performance and it is possible to reduce the thickness of pavement resulting in cost reduction [1]. Over the years, several studies have been conducted to study the behavior of geogrid reinforced pavements [2–6], and these studies reported that the geogrid is a viable alternative solution in terms of economy and durability compared to other methods. Geogrid interlocks with aggregate through its apertures. The interlocking between geogrid and aggregate forms a confined zone below and above the geogrid [4, 7,8,9,10]. The confinement effect offered by the geogrid increases the modulus of the geogrid reinforced aggregate layer resulting in improved resistance to vertical stress over the subgrade and consequent reduction in vertical subgrade deformation [4, 10, 11]. The studies showed that the geogrids can extend the service life of a pavement [2, 11–15], reduce base course thickness [13, 16–18] and delay rutting development [15, 19–23]. The improved pavement performance of geogrid reinforced base depends on various factors such as physical properties of geogrid [19–24], mechanical properties of geogrid [5, 13, 15, 21, 25], location of geogrid in pavement [4, 5, 21, 26, 27], subgrade strength [24], thickness of base layer [19, 28], and aggregate base residual stresses [29]. The geogrid improves the subgrade performance through four mechanisms (i.e., separation, confinement, improved load distribution, and tension membrane) which results in reduced stress transfer to the subgrade [21, 23, 24, 26, 30, 31]. Considerable tension membrane effect can be realized when the pavement section is constructed over a weak subgrade (i.e., subgrade with a CBR of less than 3) [30]. The location of geogrid within the base course layer in the pavement system is a crucial factor influencing the effectiveness of reinforcement [5, 27]. Research work in the past have identified the geogrid should be placed near the load, i.e., upper one-third [1, 4, 21, 26], near the bottom, i.e., bottom one-third [3, 32–34], at mid-height [1, 3] and interface of subgrade and base course [2–5, 13, 26, 29, 35, 36]. Ibrahim et al. [37] reported that geogrid placed underneath the asphalt layer greatly reduces the tensile strain at the bottom of asphalt layer. Also, when the geogrid is placed at the interface of subgrade and granular base, the reinforcement reduces the compressive stains at the top of subgrade level but not the tensile strain at the bottom of asphalt layer. As there is no general opinion regarding the optimal position for geogrid placement within a pavement section, there is a need to explore further.

The present study is focused on evaluating the effectiveness of geogrids in pavements and re-examining the optimum location for geogrid placement by carrying out a series of repeated load tests on the model pavement sections. Also, the possibility of reduction in designed pavement thickness is verified through repeated load tests.

Materials

Black cotton soil, crushed aggregate, and silty sand were used for the preparation of model pavement section. The black cotton soil was collected from Bagalkot, Karnataka, India, and used for the preparation of subgrade. Crushed aggregate was collected from a quarry located in Mydala, Tumkur, Karnataka, India, and used in the preparation of sub-base course layer. Silty sand was collected from a local site in Tumkur, Karnataka, India, and used in the preparation of sacrificial layer at the top of model pavement section. The collected materials were tested for their engineering properties, and all the tests were carried out as per the relevant Indian Standards (IS). The properties of the soils used in this study are listed in Table 1. Based on the grain size distribution and plasticity characteristics, the black cotton soil is classified as highly compressible clay with group symbol CH according to IS and unified soil classification systems [38]. It is classified as A-7-C according to HRB classification system and A-7-6 as per AASHTO [39] classification system. Based on the plasticity characteristics, dry unit weight, CBR and swell index, the soil forms a weak subgrade material as per Ministry of Road Transport and Highways, Government of India [40]. Silty sand is classified as well-graded silty sand with group symbol SW-SM according to IS and unified soil classification systems [38]. It is classified as A-2-4 according to HRB and AASHTO [39] classification system. Crushed aggregate was separated into three fractions, i.e., 40 mm down, 12.5 mm down, and stone dust of size 4.75 mm down. The properties of the crushed aggregate along with MoRT&H [40] guidelines are tabulated in Table 2. From the table, it is observed that the aggregate satisfies the strength requirements specified by MoRT&H [40], the three fractions of the crushed aggregate were blended into a mix with different proportions so as achieve the grading II sub-base course material in accordance with IRC: SP: 72 [41]. The designed sub-base mix possesses a maximum dry unit weight of 21.2 kN/m3 and optimum water content of 4% under modified proctor condition. Under this condition, the sub-base material showed a CBR of 35% and 30%, under unsoaked and soaked conditions respectively. Figure 1 shows the particle size distribution curves of black cotton soil, silty sand, and crushed aggregate.

Table 1 Engineering properties of soils
Full size table
Table 2 Properties of crushed aggregate  along with MoRT&H [40] guidelines
Full size table
Fig. 1
figure 1

Particle size distribution curves

Full size image

Two biaxial geogrids were used to reinforce the model pavement sections and are referred as BX1 and BX2. The geogrids were made up of polyester yarns with square aperture and were procured from Strata Geosystems India Pvt. Ltd., Mumbai. BX1 and BX2 had a tensile strength of 30 kN/m and 40 kN/m, respectively, in both longitudinal and transverse directions. In this study, four locations were identified for the geogrid placement, i.e., one-third from top, middle, two-third from the top of GSB and interface of subgrade and GSB to investigate the effect of position of geogrid on the pavement performance.

Construction of model pavement sections

Stage construction was adopted for the construction of model pavement sections. The test sections were built in a steel tank of size 2 m × 2 m × 2 m. The tank was initially filled with sand up to a depth of 1 m with an average relative density of greater than 85%. The model pavement of size 1 m × 1 m was built using a steel frame of size 1 m × 1 m × 0.3 m. Over the prepared sand deposit, subgrade, sub-base, and sacrificial layers were constructed sequentially. The subgrade and surface layer thicknesses were 300 mm and 50 mm, respectively. Three sub-base course layer thicknesses were considered in the present investigation, i.e., 0.25 m, 0.30 m and 0.35 m in order to study the effect of pavement thickness on the effectiveness of geogrid. The subgrade was compacted manually using a rammer to standard proctor conditions. The sub-base course and sacrificial layers were compacted using vibratory plate compactor to the respective modified proctor conditions. The compactor consists of a base plate of size 0.25 m × 0.25 m and 0.01 m thick. The compactor facilitates compaction at four different frequencies, i.e., 20 Hz, 40 Hz, 60 Hz, and 80 Hz. The vibratory compactor used in this study is shown in Fig. 2. The GSB and sacrificial layers were compacted under a frequency of 80 Hz. In case of geogrid reinforced test sections, the geogrid was placed at the desired location followed by pouring of the material and compaction. All three layers of the model pavement section were compacted in two lifts confirming to MoRT&H [40] guidelines. At the end of compaction of each layer, the density of the compacted layer was determined by suitable methods at a minimum of three points. A relative compaction of  greater than 95% for subgrade and 97% sub-base and sacrificial layers was maintained in all the model pavement sections. Figure 3 shows the pictorial view of construction of model pavement sections. At the end of construction of the test section, the surroundings of the test section were filled with sand which acts as earthen shoulders in pavements. Figures 4 and 5 show the typical pavement section under unreinforced and geogrid reinforced conditions, respectively.

Fig. 2
figure 2

Vibratory compactor

Full size image
Fig. 3
figure 3

Pictorial view of construction of model pavement section. a compacted subgrade, b partially compacted GSB, c geogrid with sakes, d geogrid layer with overlying GSB material, e compacted GSB, f compacted sacrificial layer, g model pavement section surrounded by sand

Full size image
Fig. 4
figure 4

Typical unreinforced model pavement section

Full size image
Fig. 5
figure 5

Typical geogrid reinforced model pavement section

Full size image

Test setup and procedure

A square steel plate of size  0.30 m x 0.30 m and 0.03m thick was used as test plate to transfer load to the model pavement sections. The size of the loading plate was approximately equal to one-sixth of the width of the test tank, and with this setup, the boundary effects were avoided [42]. On the prepared test section, the test plate was placed exactly at the center of the section. The assembly forms a reaction frame for loading through hydraulic jack of capacity 100 kN with a minimum count of 1 kN. A cylindrical column of 0.135 m diameter and 0.01 m thickness was used as vertical support to transfer the load from the jack to the test plate. A precalibrated proving ring of 200 kN capacity with a minimum count of 0.2 kN was attached between the hydraulic jack and the support to read the applied load. Four dial gauges were mounted one at each corner of the test plate to record the settlement of the test plate. Three dial gauges were mounted along the centerline of the test section on one side of the plate at a distance of 0.83 B, 1.16 B, and 1.5 B from the edge of the plate to record the deformation of the surface. Haversine loading was adopted to simulate the repeated applications of vehicular loading. Initially, the dial gauges were set to zero and the test plate was preloaded to a pressure of 0.7 kg/cm2 [43]. When no further settlement occurs, the dial gauges were set to zero again. A peak load of 68 kN was considered in order to simulate the tire pressure exerted by a truck with single axle and single wheel on pavements (i.e., 760 kPa). The load was then increased from seating pressure to the peak and unloaded, followed by a rest period. The dial gauge readings were recorded on loading, unloading and at the end of rest period. A total of 500 loading cycles were applied on each test section. At the end of every 10 loading cycles, the dial gauge readings were recorded. Figure 6 shows the experimental setup.

Fig. 6
figure 6

Experimental setup for repeated load test

Full size image

Results and discussions

A series of repeated load tests were carried out on unreinforced and geogrid reinforced model pavement sections. The effect of provision of geogrid within the pavement structure on the pavement performance was investigated. The influence of tensile strength of geogrid and its position and the thickness of granular layer on the pavement performance was studied. The deformation characteristics of the model pavements under repetitive loading and the influence of geogrid and their properties on the performance of the model pavements are presented and discussed in the following paragraphs.

Figures 7a–c, 8a–c, and 9a–c show the variation of total, plastic, and elastic settlement with the number of load cycles of unreinforced and geogrid reinforced (with varied geogrid position) model pavement sections with sub-base thicknesses of (a) 0.25 m, (b) 0.30 m and (c) 0.35 m, respectively. In both unreinforced and geogrid reinforced pavement sections, plastic settlement accounts for 86% to 92% of the total settlement and elastic settlement accounts for about 8% to 14% of the total settlement. This clearly indicates the predominant rutting behavior of pavements. It is observed that the total and plastic settlements increased with increase in the number of load cycles in the unreinforced model pavement section. The total and plastic settlements of geogrid reinforced test sections follow the same trend as in case of unreinforced test section with the reduced magnitude depending on the stiffness of geogrid and its position. Also, the pavement section with thicker sub-base course layer showed lesser settlement irrespective of the test condition and this is attributed to the increased durability of the thicker layer. At 50 load cycles which correspond to a traffic volume of 50CVPD (commercial vehicles per day), about 45% to 73% of the total settlement and plastic settlements occur in both unreinforced and geogrid reinforced (with varied geogrid stiffness, position and sub-base thickness) test sections when compared with the respective total and plastic settlements at the end of 500 load cycles which correspond to a traffic volume of 500CVPD. The geogrid reinforcement does not show any significant influence on the elastic behavior of the test section, and the elastic settlement remains almost same irrespective of the test condition.

Fig. 7
figure 7

Total settlement versus the number of load cycles of unreinforced and geogrid reinforced model pavement sections with GSB of thickness a 0.25 m, b 0.30 m, c 0.35 m

Full size image
Fig. 8
figure 8

Plastic settlement versus the number of load cycles of unreinforced and geogrid reinforced model pavement sections with GSB of thickness a 0.25 m, b 0.30 m, c 0.35 m

Full size image
Fig. 9
figure 9

Elastic settlement versus the number of load cycles of unreinforced and geogrid reinforced model pavement sections with GSB of thickness a 0.25 m, b 0.30 m, c 0.35 m

Full size image

The provision of geogrid provides resistance to lateral movement of aggregates in the granular sub-base and the degree of resistance depends on the properties of geogrid such as stiffness of geogrid and its position. Increased geogrid stiffness results in improved performance for all pavement thicknesses and positions of geogrid. The position of geogrid is found to have a significant influence on the pavement performance. When the geogrid is placed near to the top surface (i.e., at one-third of sub-base from its top surface), the geogrid was subjected to compression under the application of repetitive loading and the tension stress developed at the interface of geogrid and sub-base course material is very less leading to an insignificant reduction in the settlement. In this position of geogrid, the reduced settlement is due to the particle interlocking mechanism. When the position of geogrid is shifted downwards (i.e., mid-height of sub-base), the geogrid is still subjected to compression and the lateral stresses developed are higher than that of one-third position leading to an increased reduction in the settlement. The improved performance is attributed to particle interlocking and partial tension membrane effect. When the geogrid is placed at two-third of sub-base thickness from its top surface, the tensile stresses developed at the interface of geogrid and the sub-base course material are higher and therefore, the degree of resistance offered by the geogrid against lateral flow is maximum at this location. The maximum reduction in the settlement at this location is attributed to both particle interlocking and tension membrane effect. When the geogrid is shifted further downwards (i.e., at the interface of subgrade and sub-base), the geogrid offers resistance to deformation but it is found to be lesser when compared with that of two-third position. This is due to the reduced particle interlocking and reduced interface friction due to the presence of friction material on the top and cohesive material at the bottom. Based on the above discussions, it is concluded that the geogrid placed at two-third of sub-base thickness from the top (i.e., two-third) yields maximum reduction in settlement and improved pavement performance in terms of rutting.

To compare the effectiveness of geogrid reinforcement numerically, a parameter called reduction in plastic or permanent settlement (RPS) is used [1, 44]. RPS is defined as a ratio of difference between permanent settlement of unreinforced pavement section and permanent settlement of geogrid reinforced pavement section to permanent settlement of unreinforced pavement section with the number of load cycles being constant. It is expressed in terms of percentage and is given by Eq. (1).

$${\text{RPS}} = \frac{{P_{\text{UR}} - P_{\text{RE}} }}{{P_{\text{UR}} }} \times 100$$
(1)

where RPS = percentage reduction in plastic/permanent settlement, PUR = permanent settlement of unreinforced pavement section at the end of N number of load cycles, PRE = permanent deformation of geogrid reinforced pavement section at the end of N number of load cycles.

Figure 10a–c shows the variation of reduction in plastic settlement (RPS) with the number of load cycles of geogrid reinforced pavement sections with sub-bases of thickness (a) 0.25 m, (b) 0.30 m, and (C) 0.35 m, respectively. For the model pavement section with 0.25 m sub-base thickness, geogrid reinforcement positioned at one-third of sub-base from its top surface was found to perform similar to that of and/or slightly better than the respective unreinforced pavement section and did not show any appreciable increase in the RPS values irrespective of the geogrid stiffness. Reinforcing the geogrid at mid-height of sub-base resulted in increased RPS values when compared with that of the pavement section reinforced at one-third the height of sub-base from its top surface. The pavement section with geogrid reinforcement at two-third of sub-base from its top surface showed increased RPS values when compared with that of the pavement section reinforced with geogrid at one-third and mid-height of sub-base from its top surface and this position is the optimum location. When the geogrid is placed in the tension zone (i.e., near to the bottom), the geogrid layer develops tensile stresses at the interface of sub-base and geogrid due to lateral flow of granular layer resulting in the development of tension membrane effect yielding higher RPS values. At the interface of subgrade and sub-base, though the geogrid layer develops tension membrane effect yields lower RPS values when compared with that of two-third of sub-base from its top. The reduced RPS value is attributed to the particle interlocking mechanism. When the geogrid is placed within the sub-base (i.e., at two-third of sub-base from its top), the aggregate particles present above and below the geogrid forms a stiffer zone when compared to the geogrid placed at the interface where the aggregate particles present above the geogrid and soil below the geogrid. At the interface, non-homogeneous materials at the top and bottom result in unequal friction mobilization. Similar observations were found for pavement sections with a sub-base of 0.30 m and 0.35 m thick. At 50 load cycles (i.e., 50CVPD), the highest benefit (RPS) generated by the geogrid BX1 is 5% to 6%, 17% to 23%, 34% to 40% and 28% to 37%, respectively, when the geogrid is positioned at one-third, middle, two-third of sub-base from its top and interface of subgrade and sub-base with varied sub-base thicknesses. Similarly, the highest benefit (RPS) offered by the geogrid BX2 is 10% to 12%, 24% to 31%, 44% to 52% and 33% to 43%, respectively, when the geogrid is positioned at one-third, middle, two-third of sub-base from its top and interface of subgrade and sub-base with varied sub-base thicknesses. At the end of 500 load cycles (i.e., 500CVPD), the RPS values were found to be reduced and this is attributed to the degradation of aggregate particles. The reduced RPS value ranges from 3 to 37% at the end of 500 load cycles with varied geogrid stiffness, its position and sub-base thickness.

Fig. 10
figure 10

Reduction in plastic settlement (RPS) versus the number of load cycles of geogrid reinforced pavement sections with GSB of thickness of a 0.25 m, b 0.30 m, c 0.35 m

Full size image

Figure 11a–c shows the surface profiles of total settlement of unreinforced and geogrid reinforced model pavement sections with sub-base of thickness (a) 0.25 m, (b) 0.30 m and (c) 0.35 m at the end of 50 and 500 load cycles. The number of load cycles is given in the parenthesis. It is observed that the increased number of load cycles showed increased heaving at the edge of the loading plate. However, thicker pavement section showed lesser heaving. The surface heaving was found to be reduced with the provision of geogrid irrespective of the geogrid stiffness and sub-base thickness. The geogrid reinforced at two-third of sub-base from its top showed minimum heave at the surface when compared with the other positions considered irrespective of the stiffness of geogrid and sub-base thickness. This is attributed to the tension membrane effect and particle interlocking mechanism. At a particular geogrid position, BX2 was found to be effective in reducing the surface heave when compared with BX1 which indicates the effect of geogrid stiffness in minimizing surface heave and consequently reduced rutting.

Fig. 11
figure 11

Surface profiles of settlement of unreinforced and geogrid reinforced (at two-third of GSB from its top surface) pavement sections with GSB of thickness a 0.25 m, b 0.30 m, c 0.35 m at the end of 50 and 500 load cycles

Full size image

As per IRC: 115 [45], rut depth is defined as the maximum perpendicular distance measured between the bottom surface of a 3 m straight edge and the contact area of the gauge with the pavement surface at a specific location and the rut depth was determined from the test results. Figure 12 shows the schematic representation of rut depth as per IRC: 115 [45]. IRC: SP: 20 [46] specifies a maximum permissible rut depth of 50 mm for unpaved roads or gravel roads. If a pavement structure develops a rut depth beyond the maximum permissible limit, the pavement is considered unserviceable and necessary maintenance or rehabilitation must be resorted to (2014).

Fig. 12
figure 12

Schematic representation of rut depth as per IRC: 115 [45]

Full size image

Figure 13a–c shows the variation of rut depth with the number of load cycles of unreinforced and geogrid reinforced (at two-third of sub-base from its top) pavement sections with sub-bases of thickness (a) 0.25 m, (b) 0.30 m and (c) 0.35 m. The unreinforced test section develops significant rut depths with 50 load cycles (i.e., 50CVPD) and it further increased significantly with the increased load cycles irrespective of the sub-base thickness. The geogrid reinforced test sections follow a similar trend with the reduced rut depths. At 50 load cycles (i.e., 50CVPD), the measured rut depths are 12 mm to 18 mm, 7 mm to 11 mm, and 6 mm to 11 mm, respectively, in the case of unreinforced, BX1 and BX2 reinforced test sections with sub-base thickness ranging from 0.25 to 0.35 m. At 500 load cycles (i.e., 500CVPD), the measured rut depths are 19 mm to 33 mm, 14 mm to 25 mm and 12 mm to 23 mm, respectively, in the case of unreinforced, BX1 and BX2 reinforced test sections with sub-base thickness ranging from 0.25 to 0.35 m. Between 50 to 500 load cycles, the increase in rut depth is 7 mm to 15 mm in unreinforced section and 6 mm to 14 mm in BX1 reinforced section and 6 mm to 12 mm in BX2 reinforced test section with sub-base thickness ranging from 0.25 to 0.35 m. Among the two geogrids considered, BX2 is better in improving the serviceability conditions of roads over a longer duration, and thereby, the sustainability of the pavement improves.

Fig. 13
figure 13

Rut depth versus the number of load cycles of unreinforced and geogrid reinforced (at two-third of GSB from its top) pavement sections with GSB of thickness a 0.25 m, b 0.30 m, c 0.35 m

Full size image

As the tests were terminated at 500 load cycles, the rut depth accumulation was forecasted by considering the pattern of increase in rut depth with the number of load cycles. Figure 14a–c shows the variation of forecasted rut depth with the number of cycles for unreinforced and geogrid reinforced (at two-third of sub-base from its top) pavement sections with sub-base of thicknesses (a) 0.25 m, (b) 0.30 m and (c) 0.35 m. From the figure it is observed that the unreinforced model pavement sections with sub-base of 0.25 m, 0.30 m and 0.35 m thicknesses are serviceable up to 1000, 1550 and 2250 load cycles, respectively and at this stage the rut depth reaches a value of 50 mm which is the maximum permissible rut depth for unsurfaced roads as per IRC: SP: 72 [41]. The pavement section reinforced with BX1 and BX2 reinforced at two-third of sub-base from its top were found to be serviceable up to 1300 and 1450 load cycles, 1800 and 2150 load cycles and 3300 and 3500 load cycles, respectively, for sections with sub-base of thickness 0.25 m, 0.30 m and 0.35 m, respectively. Based on this forecast, with BX1 and BX2 the serviceability of the pavement section is found to increase by 1.2 to 1.5 times and 1.4 to 1.6 times for the sub-base thicknesses considered. Thus, the provision of geogrid within the pavement structure increases the service life of the pavements and the degree of improvement depends on the geogrid position and stiffness of geogrid.

Fig. 14
figure 14

Forecast rut depth versus the number of load cycles of unreinforced and geogrid reinforced (at two-third of GSB from its top) pavement sections with GSB of thickness a 0.25 m, b 0.30 m, c 0.35 m

Full size image

Two non-dimensional parameters called improvement factor [42, 44] and traffic benefit ratio [44, 47–50] are used to quantify the benefit of geogrid reinforcement. Improvement factor (If) is defined as the ratio of plastic settlement in unreinforced pavement section to the plastic settlement in geogrid reinforced pavement section with the load cycles being constant and is given by Eq. 2. Traffic benefit ratio (TBR) is defined as a ratio of the number of load cycles in geogrid reinforced pavement section to the number of load cycles in unreinforced pavement section to obtain a constant permanent settlement and is given by Eq. 3.

$$I_{\text{f}} = \frac{{P_{\text{UR}} }}{{P_{\text{RE}} }}$$
(2)

where, If = improvement factor, PUR = plastic settlement in unreinforced pavement section at the end of N load cycles, PRE = plastic settlement in geogrid reinforced pavement section at the end of N load cycles.

$${\text{TBR}} = \frac{{N_{\text{RE}} }}{{N_{\text{UR}} }}$$
(3)

where, TBR = traffic benefit ratio, NUR = number of load cycles in unreinforced pavement section corresponding to ‘X’ settlement, NRE = number of load cycles in geogrid reinforced pavement section corresponding to ‘X’ settlement.

Figures 15a–b and 16a–b show the variation of improvement factor (If) and traffic benefit ratio (TBR) of geogrid reinforced pavement sections with thickness of sub-base for BX1 and BX2, respectively. It is observed that the maximum If and TBR values were observed for the pavement sections reinforced with geogrid placed at two-third of sub-base from its top surface. This once again confirms that the location for optimum position of geogrid placement within the pavement structure is two-third of the thickness of granular layer from its top surface. The If and TBR are approximately 1 for pavements with geogrid placed at one-third of sub-base from its top surface. The values of If and TBR increases in the order of middle of sub-base, interface of subgrade and sub-base and two-third of sub-base from its top surface. It is observed that higher the stiffness of geogrid, the corresponding values of If and TBR will be maximum. With BX1 and BX2 (positioned at two-third of sub-base from its top), the maximum If values are in the order of 1.50 to 1.79, 1.52 to 1.81 and 1.80 to 2.07, respectively, for pavement sections with sub-base thicknesses of 0.25 m, 0.30 m and 0.35 m, respectively. Similarly, with BX1 and BX2 (positioned at two-third of sub-base from its top), the maximum TBR values are in the order of 1.57 to 1.89, 1.60 to 1.9 and 1.85 to 2.15, respectively, for pavement sections with 0.25 m, 0.30 m and 0.35 m thick sub-base.

Fig. 15
figure 15

Variation of improvement factor of geogrid reinforced (a) BX1 (b) BX2 pavement sections with thickness of GSB

Full size image
Fig. 16
figure 16

Variation of traffic benefit ratio of geogrid reinforced a BX1, b BX2 pavement sections with thickness of GSB

Full size image

Table 3 shows the summary of repeated load tests carried out on unreinforced and geogrid reinforced model pavement sections with varying sub-base thickness, geogrid stiffness, and geogrid position. It is observed that with the provision of geogrid at the optimal position it is possible to optimize the pavement thickness by 50 mm with equivalent/slightly better performance of that of unreinforced section depending on the geogrid stiffness.

Table 3 Repeated load test results (total and plastic settlement) of unreinforced and geogrid reinforced model pavement sections with varying sub-base thickness, geogrid stiffness, and geogrid position
Full size table

Conclusions

A laboratory study was carried out on model pavement sections built in a steel tank with unreinforced and geogrid reinforced conditions to evaluate the effectiveness of geogrid, its stiffness, and position on the pavement performance by carrying out repeated load tests. Based on the test results, the following conclusions are drawn.

  • The provision of geogrid within the pavement section, its stiffness, and position influence the pavement performance significantly. Providing the geogrid reinforcement at two-third of the thickness of granular layer from the top is very effective in improving the pavement performance and is attributed to tension membrane effect and particle interlocking mechanism. Higher the geogrid stiffness, better will be the pavement performance.

  • The provision of geogrid at optimal position reduced the plastic settlement significantly and therefore leads to improved rutting behavior. The reduction in plastic settlement is in the order of 34% to 52% depending on the geogrid stiffness and granular layer thickness. The rut life of the pavement is found to increase by 1.2 to 1.6 depending on the geogrid stiffness and granular layer thickness. However, the geogrid reinforcement has no significant effect on the elastic behavior of pavements.

  • The provision of geogrid reduced the heaving at the edges of the pavement compared to unreinforced pavement section and this reduction increased with increase in geogrid stiffness thus improving the riding quality and providing comfort to the road users.

  • The improvement factor and traffic benefit ratio were found in the range of 1.50 to 2.07 and 1.57 to 2.15 for varied geogrid stiffness and granular layer thickness.

  • With geogrid reinforcement, it is possible to reduce the thickness of the pavement by 50 mm with equivalent and/or slightly better pavement performance when compared to unreinforced pavement section.

  • However, further studies on the performance of geogrid reinforced sections using accelerated load teats and intelligent sensors are necessary.