Abstract
Recent research indicates that small changes in input parameters of reliability models significantly affect outcomes of reliability computations. This is a considerable concern especially in reliability based condition assessments of old bridges which have lower reliabilities. To overcome such errors of input parameters, structural reliability can be combined with interval analysis (i.e., interval reliability). This paper presents an interval reliability based condition assessment procedure of bridges. Interval reliability of a component is introduced when resistance and load behave as: (1) normal variables; (2) log normal variables. Similarly, interval reliabilities of other components can be estimated. Finally, system interval reliability is estimated. Bridge condition is assessed by comparing with the acceptable reliability of the bridge. A case study, a steel girder bridge, is selected, and its interval reliability indices are estimated for possible errors of statistical parameters. The obtained results are then compared with an existing method without considering interval reliability. Results indicate that there are significant variations in reliability when interval reliability is considered.
Get full access to this article
View all access options for this article.
