2.4.

3 Wheel tracking
Barksdale et al. (1989) performed a comprehensive experimental and
finite element modeling study using one type of geogrid and one type of
geotextile. Indoor test track for pavement test sections were constructed.
The testing track was loaded with one moving wheel. The pavement
surfacing and base course layer were fairly thin, while the applied load was
quite light.
Barksdale et al. (1989) noted that a geogrid with less stiffness than a
geotextile generally led to better performance. This improved performance
was attributed to the interlocking ability of the geogrid and its role in
preventing lateral spreading of the base layer soil. The test results suggested
that the geotextile required significantly higher deformation in order to
mobilize the same reinforcing potential as the geogrid. The geotextile was
superior to the geogrid in preventing mixing of the subgrade with the base
course soil.
The importance of geosynthetic location was noted, with the proper location
being dependent on the quality and thickness of the base layer soil. For the
loading used in this study, the most effective location appeared to be in the
middle of the base layer. It was suggested that higher locations in the base
layer may have produced even better results.
Collin et al. (1996) performed moving wheel load tests in an indoor test
track for the purpose of supplementing the test results of Haas et al. (1988)
and Webster (1993). Two geogrid types and sections with various base
course thicknesses (Figure 5) were examined in the study.
Collin et al. (1996) showed that test section performance improved due to
geogrid reinforcement and increased with increasing base thickness up to a
base thickness of 255mm, and then decreased with continued increasing
base thickness when the geogrid was placed at the bottom of the base. The
number of wheel loads carried by the reinforced sections exceeded that of
the control section by a factor as great as 10 for a 25 mm rut depth. Geogrid
reinforcement decreased the initial pavement deformations that occur during
the first several hundred load cycles before the section stiffens. The
reinforcement caused the deflection versus load cycle curves to flatten and
become approximately linear.

Brown et al. (1982) examined the use of nonwoven geotextile products as

reinforcement. This is one of the first reported studies of reinforcement
applications for flexible pavements. Pavement sections were constructed.
Brown et al. (1982) demonstrated that the two nonwoven geotextiles tested
were ineffective in improving performance, and in most cases resulted in
larger deformations as compared to the control sections. Observed slippage
of the base soil and the geotextile was noted as the cause of this occurrence.
Brown et al. (1982) noted that geotextiles were beneficial for construction on
wet soils where the geotextile functions of separation and filtration were
required.
Moghaddas-Nejad and Small (1996) conducted relatively small-scale
experiments using an indoor test track. The loading and pavement section
thickness were reduced by a scale from that commonly encountered in
practice.
Moghaddas-Nejad and Small (1996) demonstrated that placing Geogrid at
the bottom of a thin base results in a 40% decrease in rut depth for singletrack wheel passes and a relatively light load, whereas geogrid placement in
the middle of the base results in a 70% decrease. These values correspond to
5000 wheel passes. It was further observed that the reinforcement improved
the pressure distribution on the subgrade. Improvement in performance of
the geogrid was not due to a tensioned membrane effect, but was due to
lateral restraint of the base soil.