000018384 001__ 18384
000018384 005__ 20170118182219.0
000018384 04107 $$aeng
000018384 046__ $$k2017-01-09
000018384 100__ $$aPillai, Sujith Mangalathu Sivasubramanian
000018384 24500 $$aSensitivity of Fragility Curves To Parameter Uncertainty Using Lasso Regression

000018384 24630 $$n16.$$pProceedings of the 16th World Conference on Earthquake Engineering
000018384 260__ $$b
000018384 506__ $$arestricted
000018384 520__ $$2eng$$aFragility curves are one of the essential tools in seismic risk assessment. Various source of uncertainties such as material, geometric and ground motion have to be considered in the generation of fragility curves for regional risk assessment of bridges. The identification of parameters that have a significant effect on the seismic demand or fragility curves often requires computationally expensive and time-consuming screening methods. This paper proposes an approach to identify the significant variables using an advanced regression technique called Lasso regression. The proposed approach helps to identify and remove the less significant parameters during the generation of fragility curves without any reiteration. The proposed approach is demonstrated in this paper through the case study of a two span box girder bridge with rigid diaphragms. Although the approach is illustrated with one specific bridge type, the methodology and approach is relevant to other structural systems.

000018384 540__ $$aText je chráněný podle autorského zákona č. 121/2000 Sb.
000018384 653__ $$afragility curves, sensitivity, Lasso regression, box-girder bridges.

000018384 7112_ $$a16th World Conference on Earthquake Engineering$$cSantiago (CL)$$d2017-01-09 / 2017-01-13$$gWCEE16
000018384 720__ $$aPillai, Sujith Mangalathu Sivasubramanian$$iSoleimani, Farahnaz$$iJiang, Jiqing
000018384 8560_ $$ffischerc@itam.cas.cz
000018384 8564_ $$s441849$$uhttps://invenio.itam.cas.cz/record/18384/files/135.pdf$$yOriginal version of the author's contribution as presented on USB, paper 135.
000018384 962__ $$r16048
000018384 980__ $$aPAPER