Seismic assessment of masonry buildings accounting for limited knowledge on materials by Bayesian updating

Abstract

Several building codes propose methodologies to account for epistemic uncertainties in the seismic assessment of masonry buildings by selecting a knowledge level and reducing material strengths by means of the associated value of the confidence factor. Previous works showed that, in the case of masonry structures, this approach has various limitations, such as the lack of proper consideration of experimental tests performed. This article focuses on the issue of imperfect knowledge on material properties of existing masonry buildings and proposes a probabilistic methodology for the assessment, based on Bayesian updating of mechanical properties. The use of a Bayesian approach allows to update the values of the material properties assumed a priori as knowledge on the building increases, by taking into account all the experimental information gathered during the assessment process. A large number of simulated assessments is carried out and the values of the confidence factors on material properties are defined through the comparison between the obtained results and those of the reference structure, assumed to be perfectly known. These factors are useful in a more general framework for the assessment of masonry buildings accounting for different sources of uncertainty.

Probabilistic Seismic Assessment of Existing Masonry Buildings

“…Following an approach similar to that proposed in [6], a normal distribution has been associated to the shear strength τ k . The mean value of τ k , τ k , has been estimated combining in situ test results with the indications reported in [29], taking into account the masonry class and its qualitative features, while its standard deviation σ τk has been derived in such a way that 90% of the experimental data lies within the range of values given in [29] for the identified masonry class.…”

“…A first attempt to consider the uncertainty of masonry properties in the analysis of existing masonry buildings is discussed in [6] where a Bayesian updating procedure is proposed for the calibration of material properties to be used in the structural analysis.…”

A Decision Support System for Multi-Criteria Assessment of Large Building Stocks

“…In order to tackle this issue, a Bayesian modelling approach to learn the thermal preferences of individual occupants even in those cases in which the model structure is complex and the amount of data collected from each occupant is relatively limited was presented by Lee, Bilionis, Karava, and Tzempelikos (2017). Moreover, an approach for considering the incomplete knowledge about material properties in the assessment of existing masonry buildings based on a Bayesian framework was presented by Bracchi, Rota, Magenes, and Penna (2016). The use of a Bayesian approach makes it possible to update the values of the building material properties assumed as prior knowledge by taking into account all the experimental information gathered during the assessment process.…”

Fragility Functions for Tall URM Buildings around Early 20th Century in Lisbon. Part 1: Methodology and Application at Building Level

“…Group 1 and 2 (rubble stone masonry), Group 3 and 4 (solid brick masonry) and Group 6 and 7 (hollow brick masonry) define the mechanical properties of masonry. The interval of values is defined from the application of the Bayesian update approach (Bracchi et al, 2016;Franchin and Pagnoni, 2018). The Bayes' Theorem is used to update the probability of a priori distribution given that new evidences are available.…”