HR: 17:45h
AN: U22C-08 [PDF]
TI: A Probabilistic Benefit Cost Analysis of Seismic Retrofitting of Apartment Houses in Turkey
AU: Deodatis, G
EM: deodatis@civil.columbia.edu
AF: Columbia University, 500W 120th St.
SW Mudd Bldg. Rm 610
MC 4709, New York, NY 10027 United States
AU: * Franco, G E
EM: franco@civil.columbia.edu
AF: Columbia University, 500W 120th St.
SW Mudd Bldg. Rm 610
MC 4709, New York, NY 10027 United States
AU: Kunreuther, H
EM: Kunreuther@wharton.upenn.edu
AF: University of Pennsylvania, The Wharton School
3730 Walnut St.
500 Jon M. Huntsman Hall, Philadelphia, PA 19104 United States
AU: Lus, H
EM: hilmilus@boun.edu.tr
AF: Bogazici University, School of Engineering
Dept. of Civil Engineering, Istanbul, 34342
Turkey
AU: Seeber, N
EM: nano@ldeo.columbia.edu
AF: Columbia University, Lamont-Doherty Earth Observatory, Palisades, NY 10964 United States
AU: Smyth, A W
EM: smyth@civil.columbia.edu
AF: Columbia University, 500W 120th St.
SW Mudd Bldg. Rm 610
MC 4709, New York, NY 10027 United States
AB:
Three different seismic retrofitting measures for a common and vulnerable type of apartment building in Istanbul are studied
in this paper. A probabilistic cost benefit analysis is carried out by means of obtaining the fragility curves of the
structure in its retrofitted configurations. These curves are calculated using a Monte Carlo simulation, that is subjecting
the three dimensional nonlinear model of the structure to 400 ground motion acceleration time histories compatible with the
site specific elastic response spectrum. Considering a time independent probabilistic distribution of the peak ground
accelerations for the region and a given discount rate, the expected losses can be estimated in present value terms for the
desired time horizon.
Even considering only direct losses, mitigation is shown to be cost effective for all but the very short time-horizons
provided that the cost of life is far higher than the material costs. The methodology presented here is attractive for
application at a regional level incorporating different types of soils and different types of structures and it may be of use
to support urgent decisions on retrofitting strategies. A time dependent probability distribution of accelerations can be
readily implemented in the methodology.
DE: 7212 Earthquake ground motions and engineering
DE: 7223 Seismic hazard assessment and prediction
SC: U
MN: 2003 Fall Meeting