3D behaviour of masonry arch bridges

Lead Research Organisation: University of Sheffield
Department Name: Civil and Structural Engineering


Approaching half the bridge spans in the Western Europe are currently masonry. In the UK this amounts to >25,000 spans on the railways alone [1], and, with an estimated average masonry reconstruction cost of £800k per underbridge [2], the need to maintain and manage these bridges effectively is obvious.

Whilst in recent years there have been great advances in the analysis tools that can be applied to constructions such as masonry arch bridges, this masks the fact that the decision support tools currently used by bridge owners and their consultants tend to be highly simplified, and incapable of identifying bridges which are likely to deteriorate under traffic loads, and of diagnosing the causes of a range of commonly encountered defects. This can lead to expensive problems, something that led to Network Rail, the largest owner of masonry arch bridges in the UK, to recently raise their freight access charge, at great cost to UK plc. Network Rail also acknowledged that 'the relationship between traffic growth and cost [of maintenance] is a complex one' and that 'there would be considerable merit in undertaking further research in this area, particularly for brick and masonry underbridges' [3].

Early research on the behaviour of masonry arch bridges focussed primarily on behaviour of the arch barrel, using either elastic or plastic idealisations of behaviour (e.g. [4,5,6]). Research activity intensified in the 1980s and early 1990s, with the Transport Research Laboratory (TRL) performing load tests to collapse on redundant masonry arch bridges [7]. These tests served to highlight the complexity of real-world masonry arch bridges, though unfortunately yielded only limited useful information on their fundamental behaviour. This was due to the lack of information about the makeup of the bridges prior to testing and because at the time the sophisticated measuring techniques needed to capture key aspects of their behaviour were not available.

In an effort to capture more useful data, a range of laboratory tests were performed, for example to explore the behaviour of multi-span bridges [8] and the behaviour of soil-filled masonry arch bridges [9, 10]. In parallel with experimental studies, numerical modelling studies were also undertaken, though these have almost all used two-dimensional idealisations of the behaviour (e.g. [11]). This has meant that a wide range of intrinsically three-dimensional modes of response have been little explored, and that bridges exhibiting intrinsically 3D modes of response (e.g. diagonal cracks) are often demolished rather than repaired, at significant expense. A key aim of the present project is to address this, using high quality experimental test results obtained from controlled small-scale 3D bridge models to calibrate numerical models capable both of capturing key modes of response and of being used by practitioners.

The overall aim of the proposed project is to develop an improved understanding of the fundamental mode of response of real-world 3D masonry arch bridge configurations and to develop modelling tools capable of capturing these. This will be achieved via the following work packages:
WP1 Obtain high quality experimental data-set from 3D soil-filled masonry arch bridge tests, using a recently developed small-scale test facility
WP2 Development of practical 3D direct analysis tool
WP3 Application and dissemination of the research methods developed


10 25 50

Studentship Projects

Project Reference Relationship Related To Start End Student Name
EP/N509735/1 01/10/2016 30/09/2021
1979792 Studentship EP/N509735/1 16/10/2017 15/04/2021 Serenella Amodio
Description The overall aim of the research is to investigate the behaviour of masonry arch bridges containing internal spandrel walls, and to develop new serviceability criteria for the assessment of the working load capacity of bridges in service. This will be achieved via numerical limit analysis and laboratory tests.
Numerical limit analysis was used to model bridges containing either soil backfill or internal spandrel walls, allowing comparisons to be drawn. Specifically, the discontinuity layout optimization (DLO) limit analysis method was employed and applied to a range of bridges that have been load tested to collapse together with a case study bridge currently in service. Application of DLO has shown that the collapse behaviour of bridges with soil backfill is likely to be broadly similar to that of bridges with spandrel walls, with both failing in hinged mechanisms; however slight differences in the geometry of the failure mechanism are likely to be observed. Also, a bridge with internal spandrel walls is likely to have a higher load carrying capacity than a comparable bridge with soil backfill, though this is dependent on the extent of the voids in the bridge superstructure, and also on the boundary conditions involved.
In parallel, the permissible limit state (PLS) of masonry arch bridges is investigated, defined as the state which, if not exceeded, the lifespan of the bridge will not be affected by repeated live loading. In particular, the possibility that the PLS may correspond to the elastic limit of masonry bridges is explored. To this end, load-defection data of bridge tests from the literature are analysed and compared with numerical prediction of PLS, computed on the basis of criteria proposed in the literature: limiting displacements and accounting for material degradation.
The experimental work will involve load test to collapse on small-scale masonry arch bridges constructed with either soil fill or internal spandrels. This part of the study is currently undergoing, and outcomes will be available in the future.
Exploitation Route Masonry arch bridges continue to form a vital part of the transport networks of the UK, where most masonry arch bridges have been in service for well over 100 years. To verify that they can safely carry modern traffic they need to be regularly assessed. Furthermore, masonry bridges require lower life-cycle costs compared to other type of structures, proving that carrying out repairs is more cost effective than replacement of bridges.
However, assessment is not always straightforward. Specifically, whereas research in recent years has predominantly focused on the behaviour of arch bridges containing soil backfill, much less attention has been paid to bridges containing internal spandrel walls.
The research will provide useful information on the behaviour of bridges with internal spandrels that will help to identify the causes of a range of commonly encountered defects, feeding into the assessment process. Also, establishing the Permissible Limit State will be particularly useful for the management of bridges in service, enabling bridges which are likely to degrade rapidly under a new loading regime to be identified effectively. This should allow resources to be directed more effectively, as well as helping to preserve heritage infrastructure.
Sectors Construction,Culture, Heritage, Museums and Collections,Transport