Probabilistic earthquake loss estimation and loss disaggregation in buildings

This study develops a probabilistic loss estimation methodology that explicitly accounts and propagates many sources of uncertainty with the objective of providing quantitative measures of seismic performance in terms of economic losses. The methodology involves four main steps: (1) estimation of the ground motion intensity at the site; (2) estimation of the response of the building at different levels of ground motion intensity; (3) estimation of damage in structural and nonstructural components in the building at increasing levels of structural response; and (4) estimation of economic losses at increasing levels of damage. Results from these four separate analyses are combined using the total probability theorem to provide seven different measures of earthquake losses in a building. Sources of uncertainty in each of these four separate analyses are identified and rationally treated by considering the ground motion intensity at the site, the structural response at all story and floor levels of the building, the damage state in all structural and nonstructural components, and the repair and replacement costs as random variables. Epistemic uncertainties involved in each of these analyses is explicitly considered and propagated. Economic losses in a building are computed as a combination of losses resulting from repair and replacements of structural and nonstructural components when the building does not collapse and of losses associated with demolition and reconstruction when the building collapses. Fragility functions for various types of structural components are developed based on results from experimental research. The probability of collapse at different levels of ground motion is explicitly considered in the estimation of earthquake losses. Correlation between losses in individual elements is considered and its effects are evaluated. Loss disaggregation is proposed as a way to identify the ground motion intensities, levels of structural response and components that primarily contribute to damage and direct economic losses. Results of a loss disaggregation analysis can then provide valuable information to engineers and project stakeholders in making efficient risk management decisions. The building loss estimation methodology is illustrated by applying it to an existing non-ductile reinforced concrete building.