Open Collections

Decision tree based seismic retrofit selection for non-code conforming reinforced concrete buildings

University of British Columbia

Pacific Earthquake Engineering Research (PEER) Center has developed a comprehensive framework for quantitative assessment of performance level of structures. The framework relies on integrated work of four consecutive stages to provide probabilistic description of system level performance in terms of repair cost, downtime, casualties, deaths or any other parameter of interest to engineers and stakeholders. This is for the purpose of communicating behaviour of facility under earthquake in term of identified damage states and expected economic losses, thus treats possible disconnection between engineers and stakeholders on the desired performance target for the facility. Key objective of this dissertation is to present simplified version of the PEER framework to conduct earthquake-related financial loss studies for structures in a computationally efficient manner. The presented framework is utilized in this investigation to examine and compare efficiency of alternative seismic strengthening technique to control earthquake-induced monetary losses of a non-ductile hotel building, representative of 1960s construction. The framework integrates knowledge obtained by analyzing seismic environment at building site, investigation of structural demand, and quantifying levels of structural damage and consequential financial losses. Damage measures are computed, by generating fragility models, to link structural response directly to monetary losses. Seismic-induced economic losses are predicted by converting fragility information (i.e. damage probabilities) into financial losses utilizing inventory and monetary losses data of HAZUS-MH. The economic losses computed in this investigation included direct costs, such as construction cost of retrofit, and repair and replacement cost of the facility. In addition, indirect costs, such as losses due damage of building content and business interruption, as well as consequential losses, such as job and housing losses were also considered. Finally, decision tree model was implemented, as a final component of the framework, to establish a decision-assisting platform that enables transparent comparison and selection of the best retrofit option to reduce owner’s susceptibility for financial losses.

Text

2012-11-01

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/43564

Civil Engineering

University of British Columbia

UBCO

http://creativecommons.org/licenses/by-nc-nd/4.0/