Abstract
This thesis addresses two critical areas in transportation infrastructure: the resilience of multimodal freight networks and the deterioration of bridge components. The first part of the research presents a systematic review of resilience in multimodal freight networks, focusing on how these networks respond to disruptions, both natural and human made. The review compiles definitions of resilience from the literature and evaluates various modeling approaches, including topology-based models, finite element simulations, mathematical optimization models, and probabilistic methods. While these models provide insights into network performance during disruptions, the review identifies a gap in the application of advanced approaches, such as real-time data models and image-based simulations. Additionally, the findings emphasize the need for greater validation with real-world data to enhance practical applications and improve network resilience in real-world scenarios.
The second part of this research presents an empirical analysis of bridge deck deterioration patterns across different bridge types in Idaho, based on inspection data from the National Bridge Inventory (NBI) spanning four decades (1983–2023). Using Kaplan-Meier survival analysis, the study constructs survival curves that reveal distinct deterioration trajectories for each bridge type. Prestressed concrete girders, steel girders, and concrete girders demonstrate slower deterioration rates, maintaining higher condition ratings over extended periods, while prestressed concrete slabs and concrete slabs show faster early-stage deterioration. Concrete frames exhibit a more moderate pattern, with steady deterioration after an initial phase.
Additionally, multiple regression analysis examines the influence of various factors on deterioration rates, including bridge age, freeze-thaw cycles, and snowfall exposure. Results indicate that environmental factors such as freeze-thaw cycles and snowfall significantly accelerate deterioration, while bridge age appears as a key determinant of condition decline. In contrast, structural characteristics, such as bridge length and maximum span length, have limited impact on deterioration rates. The reduced regression model accounts for 89.4% of the variance in bridge deck deterioration, establishing a strong predictive basis for managing maintenance needs. This thesis demonstrates the importance of an integrated approach to transportation infrastructure, combining network resilience modeling with empirical analysis of structural deterioration. By improving freight network resilience through advanced simulation techniques and addressing bridge vulnerabilities with data-driven maintenance strategies, this research ensures the long-term sustainability, adaptability, and reliability of transportation systems in the face of both routine demands and unexpected disruptions.