The use of economic losses as measure of seismic performance is proposed. A methodology is developed to evaluate the expected annual loss in buildings resulting from the occurrence of earthquake ground motions. The methodology uses a component-based approach to estimate the expected loss in buildings.
A primary step in building loss estimation is a probabilistic evaluation of the structural response. A procedure aimed at computing the probability of exceedance of different types of engineering demand parameters (EDPs) is proposed. Emphasis is given to the estimation of the probability of exceedance of peak interstory drifts and peak floor accelerations at all levels in buildings. The peak interstory drift ratio at each story provides a way to estimate the damage to structural components and some of the nonstructural components. Additionally, the peak floor acceleration provides a basis for estimating damage to acceleration-sensitive moment-resisting components. The proposed procedure is aimed at situations in which economic losses are produced over a wide range of ground motion intensities.
Parameters of the probability distribution of structural response conditioned on ground motion intensity are computed from a relatively small number of deterministic response history structural analyses at three ground motion intensity levels. The proposed procedure explicitly takes into account the variation of central tendency and dispersion with changes in the loading intensity by using efficient curve-fitting procedures. The results are then combined with probabilistic estimates of ground motion intensity, obtained with a conventional probabilistic seismic hazard analysis, in order to estimate the annual probabilities of exceedance of the structural response parameters. The proposed procedure is evaluated when applied to an existing seven-story reinforced concrete building. The results are compared to those obtained with the SAC simplified procedure for estimating building response. The effects of various simplifying assumptions made in the SAC procedure are evaluated and discussed. It is concluded that the proposed procedure provides more accurate results with only a minimum additional computational effort.
Full List of PEER Reports: click here.
