A methodology for the prediction of structure level costs based on element condition states

Dilum Fernando*, Bryan T. Adey, Scott Walbridge

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract / Description of output

The determination of work programmes for bridges that maximise benefit to all stakeholders requires consideration of not only the condition of the elements of which a bridge is comprised, but also the performance of the structure as a whole. This is required because although some costs can be related to the condition of the elements, others, such as the costs of travelling over the bridge, cannot. One possible methodology to do this involves linking relevant costs to structure performance states (SPS) that are determined from the element condition states (CSs). However, even a bridge with a moderate number of elements and element CSs results in a large number of possible SPSs and an unwieldy amount of required work to estimate the costs associated with each. This work can be drastically reduced, however, by exploiting the almost linear system behaviour of the bridge that can occur between many combinations of element CSs. This article presents a methodology for relating SPSs to element CSs for the purpose of determining structure level costs. This methodology is demonstrated in several examples, wherein the effects of exploiting the linearity of the system behaviour to reduce the computational effort are also explored.

Original languageEnglish
Pages (from-to)735-748
Number of pages14
JournalStructure and Infrastructure Engineering
Issue number8
Early online date1 Sept 2011
Publication statusPublished - Aug 2013

Keywords / Materials (for Non-textual outputs)

  • bridge maintenance
  • deterioration
  • life-cycle cost analysis
  • management systems
  • Markov chain
  • optimal intervention strategies
  • user cost


Dive into the research topics of 'A methodology for the prediction of structure level costs based on element condition states'. Together they form a unique fingerprint.

Cite this