Abstract
In the context of life-cycle assessment of bridges, risk-based techniques have addressed, to a varying extent and resolution, the modelling of failure consequences. It is generally accepted that consequences fall into four major categories, namely, human, economic, environmental and social. It is also acknowledged that the modelling approach should be appropriate for the decision-making context in which consequences need to be evaluated, for example distinguishing between inspection planning and disaster impact assessment. However, recent bridge failures have raised doubts regarding the apparent lack of consistency in consequence modelling and the implications that this may have in prioritising maintenance and upgrading activities.. The aim of this paper is to review recent work on this topic and to put forward a framework for consequence modelling within a context-dependent decision-making process. The principles involved in setting up system boundaries and in specifying consequence classes, depending on bridge exposure and functionality, are briefly discussed, together with suitable modelling options, depending on the resource and information level available. Emphasis is given to the utilisation of historical failure data in refining fatality models and appraising consequence class definitions, both of which are essential in modelling bridge failure consequences.