Date of Graduation


Document Type


Degree Name

Doctor of Philosophy in Engineering (PhD)

Degree Level



Industrial Engineering


Haitao Liao

Committee Member

Pohl, Edward

Second Committee Member

Zhang, Shengfan

Third Committee Member

Yadav, Om Prakash

Fourth Committee Member

Sullivan, Kelly


Deep Reinforcement Learning, Interconnected Critical Infrastructures (ICI), Network Reliability, Quantum Computing, Simulation, Systems Reliability


This dissertation presents a framework for developing data-driven tools to model and improve the performance of Interconnected Critical Infrastructures (ICIs) in multiple contexts. The importance of ICIs for daily human activities and the large volumes of data in continuous generation in modern industries grant relevance to research efforts in this direction. Chapter 2 focuses on the impact of disruptions in Multimodal Transportation Networks, which I explored from an application perspective. The outlined research directions propose exploring the combination of simulation for decision-making with data-driven optimization paradigms to create tools that may provide stakeholders with optimal policies for a wide array of scenarios and conditions. The flexibility of the developed simulation models, in combination with cutting-edge technologies, such as Deep Reinforcement Learning (DRL), sets the foundation for promising research efforts on the performance, analysis, and optimization of Inland Waterway Transportation Systems. Chapter 3 explores data-driven models for condition monitoring and prognostics, with a focus on using Deep Learning (DL) to predict the Remaining Useful Life of turbofan engines based on sequential sensor measurements. A myriad of approaches exist for this type of problems, and the main contribution for future efforts might be centered around combining this type of data-driven methods with simulation tools and computational methods in the context of network resilience optimization. Chapter 4 revolves around developing data-driven methods for estimating all-terminal reliability of networks with arbitrary structures and outlines research directions for data-driven surrogate models. Furthermore, the use of DRL for network design optimization and maximizing all-terminal network reliability is presented. This poses a promising research venue that has been extended to network reliability problems involving dynamic decision-making on allocating new resources, maintaining and/or improving the edges already in the network, or repairing failed edges due to aging. The outlined research presents various data-driven tools developed to collaborate in the context of modeling and improvement for Critical Infrastructures. Multiple research venues have been intertwined by combining various paradigms and methods to achieve this goal. The final product is a line of research focused on reliability estimation, design optimization, and prognostics and health management for ICIs, by combining computational methods and theory.