Document Type



Doctor of Philosophy


Civil Engineering

First Adviser

Frangopol, Dan M.

Other advisers/committee members

Wilson, John L.; Yen, Ben T.; Naito, Clay J.; Cheng, Liang


Infrastructure systems are under continuous deteriorating effects due to various environmental and mechanical stressors. These effects can be generated by sudden threats such as earthquakes, tornadoes, blast, and fire, or gradual deterioration due to fatigue and corrosion. Moreover, as indicated in the 2013 American Society of Civil Engineers (ASCE) Report Card of America's Infrastructure, the United States' infrastructure systems are highly deteriorating with a required estimated investment of 3.6 trillion USD to improve their condition within the next seven years. Given the limited financial resources, rational methodologies are required to support the optimum budget allocation while maintaining maximum possible safety levels. Uncertainties associated with the performance prediction, damage initiation and propagation, damage detection capabilities, and the effect of maintenance and retrofit on the structural performance add more challenges to this allocation process. In this context, life-cycle engineering provides rational means to optimize budget allocation and manage an infrastructure system starting from the initial design and construction to dismantling and replacing the system at the end of its service life.This study provides novel management methodologies which support the decision-making process for civil and marine large-scale structural systems under fatigue and corrosion deterioration. Multi-objective optimization models that seek the optimal trade-offs between conflicting life-cycle management (LCM) aspects such as the life-cycle cost and the projected service life are proposed. These models provide the optimum intervention schedules (e.g., inspections and maintenance actions) which fulfil the LCM goals. For the first time in the field of life-cycle management, an approach capable of establishing the optimum inspection, monitoring, and repair actions simultaneously is proposed. Maximizing the expected service life, minimizing the total life-cycle cost, minimizing the maintenance delay, and maximizing the probability of damage detection are examples of the considered optimization goals. It is shown that the implementation of optimum solutions resulting from the proposed management plans can significantly reduce the life-cycle cost. A methodology for planning inspection actions for bridges with multiple critical fatigue details is proposed. This is considered a step forward from the traditional approaches which are only capable of considering one critical fatigue detail. Additionally, this study provides methodologies for the reliability-based performance evaluation of structures under fatigue deterioration. Furthermore, rational approaches which make use of structural health monitoring (SHM) and non-destructive inspection information for the near real-time decision making for deteriorating structures are proposed. Specifically, an approach to obtain the fatigue reliability of aluminium high-speed naval vessels based on SHM information is proposed. By using the proposed approach, the effect of individual operational conditions encountered by the ship on the overall fatigue damage accumulation can be quantified. This quantification is not possible by using the traditional fatigue life estimation methods. Probabilistic reliability methods and Monte Carlo simulation are implemented to account for uncertainties associated with different aspects of the LCM process. Existing large-scale structural systems are analysed to demonstrate the feasibility and effectiveness of the proposed methodologies.