Increasingly engineers are confronted with the need to perform predictive structural assessment based on limited or incomplete data set. This may include damage up to failure assessment (in the context of so-called performance based engineering), or round robin benchmarks. As such deterministic analyses are of limited predictive values, and a probabilistic-based methodology is necessary.
This paper focuses on the development of a methodology for such assessment, and is believed to be the first such contribution in the context of structural failure following alkali silica reaction (ASR) induced expansion. As a vehicle for such an application, the round-robin benchmark surrounding the shear failure of a reinforced concrete wall having suffered from ASR expansion is used. However, for reasons explained later, the authors believe that the reported experimental data may not be of sufficient reliability, and hence the predictive nature of the analysis is reported but deemphasized.