Fragility functions for non-ductile infilled reinforced concrete buildings using next-generation intensity measures based on analytical models and empirical data from past earthquakes

Abstract

The regional seismic risk assessment of reinforced concrete (RC) building portfolios is a critical issue in earthquake engineering due to their high vulnerability and widespread distribution in seismic prone areas. A pertinent aspect in regional seismic risk applications is the ability to accurately quantify the exceedance of any damage state, generally via fragility functions. To this end, this study derives analytical fragility functions for large-scale seismic risk applications of non-ductile RC buildings with masonry infills characteristic of the Italian peninsula and Southern Europe in general. These were derived using a large database of archetype buildings developed to represent the temporal evolution in construction practice in Italy based on an extensive literature review and interviews with practising engineers and architects. Fragility functions for several infilled RC taxonomy classes were derived for multiple damage states using state-of-the-art analysis on detailed numerical models. Average spectral acceleration was adopted as the intensity measure throughout, since it has been shown to notably reduce dispersion and bias in quantifying the response, and subsequently refine the seismic risk estimates, for these typologies. The fragility functions are compared against empirical data collected following past earthquakes in Italy, namely L’Aquila 2009 and Umbria-Marche 1997. The development of empirical fragility functions was carried out using a novel derivation of average spectral acceleration-based ground-motion fields considering spatial and cross-period correlation models, which is a key component and development in this study. This paper shows how recent advances in analytical fragility function development can be integrated with past empirical observations to give more accurate and representative damage estimates for regional assessment.