Unlike in conventional bridges, the backfill and the roadway pavement have a major bearing on the load capacity of buried corrugated metal structures. In the soil-steel structure model one can distinguish two structural subsystems: the shell made of corrugated steel plates and the soil backfill with the road pavement. The interaction between them is modelled as a contact (interfacial) interaction, i.e. forces normal and tangent to the surface of the shell. The normal interactions are variable during both construction and service life. Two algorithms are presented. In the first algorithm on the basis of unit strains the internal forces in the shell are determined and consequently the contact interactions are calculated. A large number of measuring points distributed on the circumferential section of the shell is needed for the calculations. In the second algorithm the collocation condition, according to which the result obtained from the shell geometry model must agree with the measured displacement of the structure’s collocation point, is used. When there are more such points, the estimated result is more precise. The advantage of both algorithms is that they take into account the physical characteristics of the soil in the backfill layers, but above all the backfill laying and compacting technology. The results of such analyses can be the basis for comparing the effectiveness of conventional geotechnical models.

The problem of the arch barrel deformation in railway backfilled arch bridges caused by their typical service loads is analysed. The main attention is paid to vertical or radial displacements of characteristic points of the arch barrel. In the study results of deflection measurements carried out on single and multi-span backfilled arch bridges made of bricks or plain concrete during passages of various typical railway vehicles are used. On the basis of such results empirical influence functions of displacements are being created. In the next step, the results are utilised to estimate bending effects within the arch. The paper includes different procedures based on measurements of displacements in various points and directions. Using empirical influence functions arbitrary virtual load cases may be also considered. In this manner the proposed methodology shows a potential to be an effective tool of comprehensive calibration of numerical models of backfilled arch bridges on the basic of field tests carried out under any live loads.

In this paper, existing knowledge on the behaviour of soil-steel composite structures (SSCSs) has been reviewed. In particular, the response of buried corrugated steel plates (CSPs) to static, semistatic, and dynamic loads has been covered. Furthermore, the performance of SSCS under extreme loading, i.e., loading until failure, has been studied. To investigate the behaviour of the type of composite structures considered, numerous full-scale tests and numerical simulations have been conducted for both arched and box shapes of the shell. In addition, researchers have examined different span lengths and cover depths. Furthermore, to enhance the load-bearing capacity of the composite structures, various stiffening elements have been applied and tested. The reviewshows that the mechanical features of SSCSs are mainly based on the interaction of the shell with the soil backfill. The structures, as a composite system, become appropriately stiff when completely backfilled. For this reason, the construction phase corresponds to the highest values of shell displacement and stress. Moreover, the method of laying and compacting the backfill, as well as the thickness of the cover, has a significant impact on the behaviour of the structure at the stage of operation in both the quantitative and qualitative sense. Finally, a limited number of studies are conducted on the ultimate bearing capacity of large-span SSCS and various reinforcing methods. Considerably more works will need to be done on this topic. It applies to both full scale tests and numerical analysis.