Dynamic cone penetration testing has been extensively used in the past for subgrade performance assessment and quality control in road construction practice. However, the method is not commonly employed on high-speed railways. This is due to lack of field data that prove its feasibility as an alternative method for assessing subgrade quality. To mitigate this gap, a series of in situ tests was performed on existing subgrades built with coarse-grained soils at five different sections along the Tehran-Isfahan high-speed railway in Iran. At each subgrade section tested, four parameters for compaction quality control-blow count, degree of compaction, subgrade reaction modulus and dynamic deformation modulus-were determined at nine different depths from subgrade surface. On the basis of the results obtained, a correlation model was developed to relate the traditional quality control parameters of compacted subgrade fill materials with the blow counts. Finally, a simple method using the correlation models established was proposed for assessment of subgrade compaction quality. The method proposed proved to be an alternative approach for evaluating the state of subgrade compaction and also for assessing the subgrade performance of existing railway subgrades.

Some newly built high-speed railways inevitably pass through the qanat region, which may potentially cause the failure of the qanat tunnels upon embankment and traffic loadings. Thus, it is of great importance to evaluate the stability of the qanat tunnels under surcharge for the design of railways. In the context of the Tehran–Isfahan railway construction in Iran, the soil properties obtained from a series of site investigation and laboratory tests show an important spatial variability. To better consider the spatial variability of soil strength parameters, a series of random adaptive finite element limit analysis (RAFELA) is performed in this study to investigate the stability of the qanat tunnels in the cohesive-frictional soils. The emphasis is placed on the effects of the coefficient of variation and vertical spatial correlation length of the strength parameters (i.e. soil cohesion c and effective fiction angle φ) and the qanat tunnel depth (H/D). Results show that considering the variabilities of both c and φis more unfavourable to the qanat stability compared to the cases with only one strength parameter (c or φ) variability being considered. Detailed stability evaluation charts are obtained to assess the probability of failure of the qanat tunnels under different conditions.