Date

2013

Document Type

Dissertation

Degree

Doctor of Philosophy

Department

Structural Engineering

First Adviser

Frangopol, Dan M.

Other advisers/committee members

Wilson, John L.; Sause, Richard; Naito, Clay J.; Casas, Joan R.

Abstract

Structural systems are usually subjected to progressive and/or sudden damage throughout their lifetime. Damage can cause a reduced level of safety and increase the life-cycle cost. In order to keep the safety and proper functionality of structures above prescribed thresholds, maintenance interventions should be well planned. Informed decision making for maintaining the required safety and serviceability levels of structural systems under uncertainty during their lifetime can only be achieved through integrated life-cycle management planning. Structural performance assessment and prediction, optimization of inspection and monitoring activities, updating the performance with information from inspection and monitoring, optimization of maintenance and repair activities and decision making are the main tasks of an integrated life-cycle management framework. Accurate prediction and quantification of life-cycle performance is the most critical task in a life-cycle management framework. Uncertainty is inevitable in all aspects of this framework. Aggressive environmental conditions may cause the strength of a structure to deteriorate progressively in time. The deterioration process is complex and contains high uncertainty. Therefore, probabilistic methods are required for accurate assessment of structural performance. Reliability-based performance measures are the primary tool for structural management optimization frameworks. Extreme events such as floods, earthquakes, and blasts may cause sudden damage to structures. A structure must be able to withstand local damage without experiencing disproportionate consequences. Performance measures such as reliability, redundancy, robustness, vulnerability and damage tolerance should be considered in the life-cycle management of structures. Risk-based approaches provide the means of combining the probability of structural failure with the consequences of this event. There is the need to effectively incorporate the risk-based performance measures into the life-cycle management frameworks by accounting for the probabilities of occurrence of failure events and the associated consequences using scenario-based approaches in a computationally efficient manner. The main objective of this study is to develop means for integrating the reliability-based and risk-based performance indicators in a life-cycle management framework for structures undergoing progressive and sudden damage. An approach for quantifying time-variant reliability, redundancy, vulnerability, and robustness of structural systems in a life-cycle context is developed. A methodology for quantifying lifetime risk associated with the component failure and risk-based robustness of bridge superstructures is proposed. Furthermore, a risk-based maintenance optimization methodology for deteriorating bridges to establish optimum maintenance plans is proposed. A methodology to assess the time-dependent expected losses and risk-based robustness of highway bridge networks consisting of deteriorating bridges is established. In addition, a probabilistic approach for performance assessment of ship hulls under sudden damage accounting for different operational conditions is developed. Finally, the applicable range of simple expressions based on first-order second-moment method for bridge system reliability assessment is provided by investigating the amount of error associated with these simple expressions.

Share

COinS