94-0003 A Markov Model for Local and Global Damage Indices in Seismic Analysis S.Rahman, M.Grigoriu 1994 Cornell University 288 PB94-206000 20 This study considers several important issues regarding probabilistic seismic performance of structural systems. Three major directions of research are pursued: 1) evaluation of effects of simplifications in reliability-based design codes; 2) development of a new methodology based on Markov model for seismic reliability of degraded structures; and 3) development of analytical relations between local and global damage indices for seismic analysis of shear type buildings. The Markov model involves simple but realistic characterization of seismic hazard, nonlinear dynamic analysis for estimating structural response, uncertainty in the initial state of structural systems, and failure conditions incorporating damage accumulation during consecutive seismic events. From the proposed model, both event and lifetime reliabilities can be calculated, thus providing a designer more control in seismic performance evaluation. It can be used to determine the damage probability evolution during several earthquakes allowing investigation on seismic vulnerability of new and existing structures. The model can be used to compute mean first passage time determining average number of seismic events before the structure will suffer potential damage. It can also evaluate sensitivity of seismic reliability due to the variability in the initial state of structural systems. Results from this study indicate that the seismic reliability based on lifetime largest load effects can differ significantly from that obtained from seismic hazard based on damage accumulation between seismic events and the uncertainty in initial condition can yield significant variation in the seismic reliability estimate.<BR> Local Damage Indices, Global Damage Indices, Second Order Reliability Methods (SORM), First Order Reliability Methods (FORM), Reliability Analysis, Stochastic Processes, Damage Assessment, Nondegrading Systems, Degrading Systems, Markov Processes, and Earthquake Engineering.