Soil–geosynthetic interface shear strength is an essential parameter for the design and stability analysis of geosynthetic-reinforced soil structures. Economic and environmental reasons have led to increasing use of locally available residual soils with a significant percentage of fines and lower draining capacity, when compared with the traditional good-quality backfill materials. This paper describes an extensive laboratory study carried out using a large-scale direct shear test device, in which the influence of soil moisture content, soil density and geosynthetic type on the direct shear behaviour of the soil–geosynthetic interface was evaluated. The study involved a locally available granite residual soil and four geosynthetics: two geogrids (one uniaxial and the other biaxial), one geocomposite reinforcement (high-strength geotextile) and one geotextile. Test results have revealed that the increase in soil moisture content can measurably reduce the soil–geosynthetic interface shear strength. Regardless of soil moisture content, soil density proved to have a remarkable influence on interface shear strength, particularly when geogrids were used. Among the different geosynthetics tested, the biaxial geogrid was found to be the most effective reinforcement for this particular type of soil, concerning the direct shear mechanism. For soil–geogrid interfaces, the coefficients of interaction ranged from 0.71 to 0.99. For soil–geotextile interfaces, the coefficients of interaction varied from 0.54 to 0.85.

Railway tracks are conventionally built on compacted ballast and structural fill layers placed above the natural (subgrade) foundation. However, during train operations, track deteriorations occur progressively due to ballast degradation. The associated track deformation is usually accompanied by a reduction in both load bearing capacity and drainage, apart from imposing frequent track maintenance. Suitable ground improvement techniques involving plastic inclusions (e.g., geogrids) and energy absorbing materials (e.g., rubber products) to enhance the stability and longevity of tracks have become increasingly popular. This paper presents the outcomes from innovative research and development measures into the use of plastic and rubber elements in rail tracks undertaken at the University of Wollongong, Australia, over the past twenty years. The results obtained from laboratory tests, mathematical modelling and numerical modelling reveal that track performance can be improved significantly by using geogrid and energy absorbing rubber products (e.g., rubber crumbs, waste tire-cell and rubber mats). Test results show that the addition of rubber materials can efficiently improve the energy absorption of the structural layer and also reduce ballast breakage. Furthermore, by incorporating the work input parameters, the energy absorbing property of the newly developed synthetic capping layer is captured by correct modelling of dilatancy. In addition, the laboratory behavior of tire cells and geogrids has been validated by numerical modelling (i.e., Finite Element Modelling-FEM, Discrete Element—DEM), and a coupled DEM-FEM modelling approach is also introduced to simulate ballast deformation.

Geosynthetics, including geogrids and geotextiles, have been extensively used for stabilisation and soil reinforcement in several geotechnical structures, such as foundations, abutments, walls and slopes. In these applications, soil–geosynthetic interaction plays a determinant role. This paper describes an experimental study carried out using a large-scale pullout test apparatus, aiming to investigate the pullout behaviour of different geosynthetics embedded in a granite residual soil. The study involved two geogrids (one biaxial and the other uniaxial), one geocomposite reinforcement (high-strength geotextile) and one geotextile. The soil was compacted to different relative densities. Test results have revealed that soil–geosynthetic interaction under pullout loading conditions is highly influenced by the geosynthetic properties and soil density. Regardless of soil density, the biaxial geogrid exhibited higher pullout resistance than the other geosynthetics. At maximum pullout force, the deformation along the length of the geotextiles was considerably more pronounced than that along the geogrids. For the conditions adopted in this study, the soil–geosynthetic pullout interaction coefficients ranged from .25 to .52. In general, the values of the scale effect correction factor obtained for the geotextiles were slightly lower than the value recommended by the Federal Highway Administration in the absence of test data.

The understanding of soil-geosynthetic interaction under cyclic loading conditions is essential for the safe design of geosynthetic-reinforced soil structures subjected to repeated loads, such as those induced by road and railway traffic and earthquakes. This paper describes a series of large-scale monotonic and multistage pullout tests carried out to investigate the behaviour of an HDPE uniaxial geogrid embedded in a locally available granite residual soil under monotonic and cyclic pullout loading. The effects of the pullout load level at the start of the cyclic stage, cyclic load frequency and amplitude, number of cycles and soil density on the load-strain-displacement response of the reinforcement are evaluated and discussed. Test results show that the cumulative displacements measured along the length of the geogrid during cyclic loading increased significantly with the pre-cyclic pullout load level and the load amplitude. In contrast, the cumulative cyclic displacements were found to decrease with increasing frequency and soil density. In medium dense soil conditions, the geogrid post-cyclic pullout resistance decreased by up to 20%, with respect to the value obtained in the comparable monotonic test. However, for dense soil, the effect of cyclic loading on the peak pullout forces recorded during the tests was almost negligible.

In slope stability analysis, the high degree of uncertainty associated with design parameters has led to increasing use of reliability-based approaches as a means of evaluating the combined effects of such uncertainties on the structure performance. In this study, the reliability level of geosynthetic-reinforced steep slopes designed according to Eurocode 7 (EC7), without any additional margin of safety, was assessed using the commercially available Slide 6.0 software based on Monte Carlo simulation. To validate the EC7 partial factor design method regarding structural reliability, the estimated reliability indexes were compared with the minimum value recommended by Eurocode 0 (EC0). Additionally, through a probabilistic sensitivity analysis, the effect of variability in design parameters on slope reliability was evaluated and discussed. The results have shown that the geosynthetic-reinforced slopes designed to EC7 specifications exhibit generally an adequate reliability level according to EC0. The soil friction angle and the friction angle of the soil–geosynthetic interface and, secondly, the surcharge load, were found to be the most significant parameters for the reliability of the analysed slopes. For typical coefficients of variation of design parameters, the EC7 partial factor method tends to be conservative in terms of structural reliability. However, in situations of abnormal high variability, the partial factor methodology may lead to unsafe design, and thus reliability analyses should be implemented.