AsimplifiedfragilityanalysisoffantypecablestayedbridgesusingProbabilisticRiskAnalysis(PRA)procedureispresentedfordeterminingtheirfailureprobabilityunderrandomgroundmotion.Seismicinputtothebridgesupportisconsideredtobeariskconsistentresponsespectrumwhichisobtainedfromaseparateanalysis.Fortheresponseanalysis,thebridgedeckismodeledasabeamsupportedonspringsatdifferentpoints.Thestiffnessesofthespringsaredeterminedbyaseparate2Dstaticanalysisofcable-tower-decksystem.Theanalysisprovidesacoupledstiffnessmatrixforthespringsystem.Acontinuummethodofanalysisusingdynamicstiffnessisusedtodeterminethedynamicpropertiesofthebridges.Theresponseofthebridgedeckisobtainedbytheresponsespectrummethodofanalysisasappliedtomultidegreeoffreedomsystemwhichdulytakesintoaccountthequasi-staticcomponentofbridgedeckvibration.Thefragilityanalysisincludesuncertaintiesarisingduetothevariationingroundmotion,materialproperty,modeling,methodofanalysis,ductilityfactoranddamageconcentrationeffect.ProbabilityoffailureofthebridgedeckisdeterminedbytheFirstOrderSecondMoment(FOSM)methodofreliability.Athreespandoubleplanesymmetricalfantypecablestayedbridgeoftotalspan689m,isusedasanillustrativeexample.Thefragilitycurvesforthebridgedeckfailureareobtainedunderanumberofparametricvariations.Someoftheimportantconclusionsofthestudyindicatethat(i)notonlyverticalcomponentbutalsothehorizontalcomponentofgroundmotionhasconsiderableeffectontheprobabilityoffailure;(ii)groundmotionwithnotimelagbetweensupportexcitationsprovidesasmallerprobabilityoffailureascomparedtogroundmotionwithverylargetimelagbetweensupportexcitation;and(iii)probabilityoffailuremayconsiderablyincreaseforsoftsoilcondition.
