Statistical, empirical and mechanistic fragility analysis of concrete bridges

Performance of a highway transportation network in a seismically active region depends on the damageability of its components, particularly bridges. Therefore, bridge damageability information is an important ingredient in the seismic performance and risk evaluation of highway network systems. Expression of this damageability in the form of fragility curves is a commonly practiced technique and can easily be incorporated into the scheme of risk analysis of network systems. The primary objective of this dissertation is statistical, empirical and mechanistic fragility analysis of reinforced concrete bridges. Nonlinear time history analysis of bridges is performed under earthquake ground motions having different hazard levels with an emphasis on the gradual time-progression of structural damage towards failure. Bridge damageability thus simulated is expressed in the form of fragility curve which is assumed to be a two-parameter lognormal distribution function. A mechanistic model is developed in calibration with the empirical fragility curves constructed on the basis of damage data obtained from the 1994 Northridge earthquake. This calibration quantifies threshold values of bridge damage states in accordance with the physical damage described in HASUZ (1999). This mechanistic model is transportable and applicable for any type and size of bridges. In addition, the calibrated definitions are compared with experimental result obtained from a large-scale shaking table test conducted at the University of Nevada, Reno. A practice-oriented nonlinear static analysis method for concrete bridges is proposed in line with the concept of performance-based structural design. Bridge seismic performance is evaluated through the interaction of capacity spectrum of the bridge and inelastic demand spectrum of earthquake ground motion. In this method, bridge nonlinear behavior and influences of its higher modes are considered. Study indicates that this method can serve as a good alternative of nonlinear time history analysis as fragility curves developed from these two procedures are in excellent agreement. Furthermore, this research deals with the directionality effect of earthquake ground motions in seismic vulnerability assessment of bridges. Even focusing only on two horizontal translational components of ground motions, it is not possible to identify the principal directions of these components and orientation of structures relative to them. Modeling of these components into 2D vector random process is too complex and not useful because of lack of experimental data for the ensuring damageability analysis. Hence, a practice-oriented concept is discussed here considering that bridges must resist seismic ground motions approaching from all possible directions to structural axes. It is believed that the theoretical discussions and analytical tools presented in this research will add to the existing knowledge of fragility analysis of concrete bridges, and upgrade the accuracy of seismic risk assessment of highway transportation systems.