The performance assessment of structures is subjected to various sources of uncertainties. One of the most challenging aspects of the seismic assessment of existing buildings is the characterization of structural modeling uncertainties since it measures the incomplete knowledge of the structural properties in as-built conditions.

In the structural reliability assessment, the uncertainties are propagated in order to assess the structural performance in terms of the probability of exceeding a specified limit state. The Bayesian framework seems to be particularly suitable for reliability assessment of existing RC structures. It allows for adaptive updating of both the structural reliability and the uncertainties based on the results of tests and inspections available. The application of the Bayesian framework for the performance assessment of existing buildings requires the characterization of the uncertainties (including structural modeling uncertainties) by prescribing prior probability distributions, before taking into account the output of the tests and inspections. Considerable care should be taken when characterizing the prior probability distributions; especially when there is little information available from tests and inspections.

It is possible to distinguish between two different sources of structural modeling uncertainties in the assessment of existing buildings: uncertainty in the mechanical properties of materials used in construction and the uncertainty in construction details that affect the component capacities. It is important to take into account the uncertainties in the construction details since the variations in structural detailing parameters (a.k.a. structural defects) can be quite significant; to the extent that they may change the eventual structural collapse mechanism. The uncertainties in construction details can be mainly attributed to the incomplete knowledge of: (a) the structural original design, and/or (b) the as-built conditions of the structure.

The focus of this work is to provide a basis for the characterization of the uncertainties related to the construction details attributed to the as-built conditions of the structure. This is achieved by interpreting and quantifying the common pool of knowledge created by professional experience into prior probability distributions. These prior probability distributions can be used in the Bayesian framework in order to take into account the uncertainties in the construction details before implementing the data provided by the tests and inspections. A comprehensive survey addressed to practicing engineers has been conducted. The basic idea was to identify and to characterize the most dominant types of construction defects which may be found in the existing RC buildings in Campania, Italy built after the second world war. The preliminary results obtained by interviewing a group of practicing structural engineers are presented. It is demonstrated how the results of such survey can be processed in order to construct prior probability distributions for selected structural detailing parameters.