Earthquake damage assessment of 1-story precast industrial buildings using damage probability matrices

Abstract

In this study, risk assessment of 1-story precast building is conducted in probabilistic manner by using damage probability matrices (DPMs). Peak ground velocity (PGV) is used as reference ground motion parameter to associate with structural damages and more than three hundred real earthquake records is used for nonlinear dynamic analysis of precast buildings. Three different probabilistic calculation approaches mainly based on analytical methods have been used to obtain DPMs and utilized methods are compared. Comparisons have shown that analytical methods presented herein can be used to describe the upper and lower bound of damage probabilities of precast buildings, respectively. Evaluations on the damage probabilities have also pointed out that the structural differences among the buildings, use of different probabilistic approaches and damage ratios related with structural damage costs cause remarkable differences between the damage predictions and they can be considered as primary sources of uncertainties. Logic tree method, which has ability to gather involved uncertainties, is used to carry out reliable damage assessment. Results have shown that damage ratios of precast buildings determined after logic tree analysis have a good agreement with site investigation studies of 1999 Kocaeli earthquake occurred in Turkey and presented methods can be used in risk assessment of precast industrial buildings.