In this work the main assumptions to be made in the seismic assessment of existing RC buildings have been assessed, in order to evaluate how much each of them can affect the obtained evaluation of the building safety, with reference to a case-study structure. Eurocode 8 provides all the information needed to perform the seismic assessment of existing buildings. Even if all the provided prescriptions are followed, in several assumptions a large subjectivity is left to the engineer in charge of the analysis. In this work the scatter related to such choices is evaluated with reference to a case-study building: a real RC Italian building currently used as a hospital. The contemplated assumptions concern the material characterization, the seismic input definition, the type of performed analysis and the set-up of the numerical model adopted to be representative of the building behaviour. The effects of each of these assumptions have been assessed in terms of scatter induced in the assumed response parameters, in order to evaluate the role of the discretionary choices in the seismic assessment. The results presented in this paper are representative of the analysed case-study building only, but aim at putting a spotlight to the effect that some too discretional prescriptions of the technical codes can have on the seismic assessment of existing structures.

This work investigates the effects of the choice of different ensembles of ground motions on the seismic assessment of existing RC buildings through nonlinear dynamic analysis. Nowadays indeed, all the main International Seismic Codes provide a soil classification which is based on the shear wave velocity, the soil morphology and the assumed distance from the fault source. Depending on the soil properties, a suitable elastic spectrum is provided as target, defined on the basis of average properties assumed for the soil. An ensemble of ground motions, compatible to the target one, must be selected to perform a nonlinear dynamic analysis. The ensemble can be made by artificial or natural ground motions, compatible with the Code spectrum for the assumed soil-type. Alternatively, the set of ground motions can be assumed as compatible with the bedrock Code spectrum and, subsequently, subjected to site response analysis, i.e. filtered through the specific stratigraphy of the site soil. In this work a comparison among these different approaches, all compatible to the European (Eurocode 8, EC8) and Italian (NTC 2008) Code provisions, has been made on a case-study, i.e. a real RC Italian building. The seismic response of the case-study under the assumed seismic inputs, expressed in terms of chord rotation and shear force, has been found by performing a nonlinear dynamic analysis under the different assumed seismic excitations. The comparison has been made in terms of seismic performance, expressed as the ratio between the seismic response found for each structural element and the corresponding capacity. The comparison among the seismic performance found by the application of the different ground motion ensembles pointed out significant differences, which underline the importance of the seismic input choice in the seismic assessment of RC buildings.

It is well known that the axial load plays an important role in the evaluation of the structural capacity of RC columns. In existing buildings this problem can be even more significant than in new ones, since the material can easily present poor mechanical properties. The paper is aimed at the investigation of the role of the axial load variation on the seismic performance of RC columns of a case-study, i.e. a doubly symmetric 4-storey RC building. The effects of the axial load variation have been checked on the first storey columns, by comparing the seismic response, measured in terms of chord rotation and shear force, with the corresponding capacity. The sensitivity of the seismic performance to the axial load is evaluated with special attention on the type of analysis adopted to determine the seismic response and on considering a wide range of values for the concrete strength. The study points out a non-negligible effect of the axial load variation on the seismic response of the case-study building, especially when combined to concrete strength variability.

Computational models for masonry are briefly reviewed and judged upon their practical performance at the structural scale, i.e. at building level, in a nonlinear pushover or nonlinear time history setting. Particular attention is given to an anisotropic macro model based upon total stress-strain relations in tension, shear and compression with proper unloading/reloading. A multi-level experimental campaign for Groningen masonry delivered material input and validation data at component and structural level. Brief attention is given to the temporal discretization, with a sequentially linear scheme that drives the solution over the peak up to structural softening down to zero, as an alternative to incremental-iterative implicit schemes.