Title of propsed paper for ASRANet International Colloquium 2002:
Element and System Risk-Based Evaluation of Highway Bridges
Authors
Wong, S.M., Hobbs, R.E., and Onof, C.
Abstract:
The areas that are currently being emphasised in bridge assessment include element behaviour and criticality, system behaviour, inspection levels, traffic load category and road roughness. In the UK, Highways Agency document BD 21/97 adopts reliability-calibrated live load levels to account for different scenarios involving traffic load category and road roughness but this does not explicitly account for the element and system behaviours. On the other hand, some bridge assessment codes propose a range of values of reliability indices to account for different circumstances related to these areas. However, it is found that the proposed values are tentative. Therefore, further research is needed to establish a rational approach for bridge assessment and design, considering the element and system effects.
In the early probabilistic design codes, the consistency of level of safety/failure probability has been emphasised for bridge design and assessment. However, if proper consideration is not given to the consequences of failure, structural components associated with serious consequences of failure (i.e. total collapse) may be under-designed. Similarly, a frequently damaged bridge element/member may not lead to a total collapse. Such an element may be redundant and have been given too much attention for maintenance while there are higher risks existing in other ‘hidden’ critical elements that lead to a system failure.
Risk assessment based on cost-benefit analysis has been gaining in importance as people recognise that safety and failure consequence involve not only engineering analysis but also social, economic, cultural, ethical and many other factors. Theoretically, it is a general approach for assessing structural safety associated with failure consequences. However, such an approach may pose difficulty when evaluating failure consequences that involve injuries and fatalities.
In this project, a failure-consequence-based approach has been adopted to measure bridge safety in terms of the risk level rather than the conventional failure probability level. In view of the probabilistic approach, a response surface approach coupled with grillage analysis has been implemented to perform the risk analysis. As far as the failure consequence is concerned, it is further defined as the ‘immediate consequence without system failure’ and the ‘total consequence with system failure’. From the structural point of view, these two types of failure consequences will be evaluated in terms of the changes in system strength. As the research progresses, consideration will also be given for the consequences associated with different types of works, traffic and site management during execution, and traffic delays.
Much effort will be directed to the development of this idea and approach so that it can be applied to produce the results of risks at various states (i.e. the system risk due to element(s) failures or multiple failure modes and the risk of progressive collapse). With the risk of a highway bridge evaluated at an element and at a system level, the critical elements and failure paths can be clearly identified. Hence, this will provide an important piece of information for bridge maintenance and also enable a risk-based management framework for highway bridges to be established on a more rational basis.
Contact:
Imperial College of Science, Technology and Medicine, London