Project DescriptionNSF EFRI-1441231, "Collaborative Research: RIPS Type 2: Vulnerability Assessment and Resilient Design of Interdependent Infrastructures"
Modern infrastructure systems, such as power grids, communication networks, and transportation networks are interdependent in such a way that a failure of an element in one system may cause multiple failures of elements in other systems. This process can propagate back and forth between interdependent systems in a cascading fashion, resulting in a catastrophic widespread failure. In addition, the diverse human behaviors to disruptions, such as drivers' reaction to gridlock, can further complicate the cascading behaviors. Radically new models and analytical techniques are needed to assess and design resilient interdependent systems.
In this project, a team of five investigators from the domains of computer science, optimization, transportation systems, power engineering, and social science will work together to gain a better understanding of cascading failure phenomena, develop tractable mathematical models for designing resilient interdependent systems, and investigate innovative strategies to enhance the resilience of interdependent systems by preventing the occurrence of cascading failures and quickly restoring system operations. This research will lay a foundation in understanding the fundamental properties that contribute to the robustness of interdependent systems under disruptions, and thus, advancing the state-of-the-art in modern complex network theory and optimization algorithms. The transformative contributions of the project are as follows. The investigators will offer the first models that can characterize the scale and depth of cascading failures in interdependent systems, introduce the new concept of "human vulnerability", and provide the first model on identifying critical network elements based on serviceability. The findings of the research will provide timely support for public and private agencies to better understand the impacts of cascading failures and the implications of protecting critical elements, and develop policies to enhance the resilience of the interdependent infrastructure systems. In particular, the findings can potentially diversify the choices of these policies for managing transportation networks and power grids. The research results will also enrich the literature in the areas of network science, graph theory, optimization, communications, transportation systems, power engineering, and social science. The project will involve students at all levels, with emphasis on attracting students from underrepresented groups. The real-world applications will offer an ideal platform to engage undergraduate and K-12 students and to reach out to practitioners and policy makers.
Via a combination of theoretical (mathematical modeling and optimization) and applied (domain expertise) approaches, this project will comprehensively investigate vulnerability and resilience issues in interdependent systems. As specific steps towards this goal, the investigators will pursue five interdisciplinary research tasks: 1) analyzing the mechanisms of cascading failures in interdependent systems by mathematically quantifying the "depth" and "breadth" of cascades; 2) identifying critical elements (nodes and/or links) whose removal yields the most significant loss of resilience of interdependent systems; 3) enhancing the resilience of interdependent systems via optimal addition of inter-network links and finding adaptive control strategies to rapidly react to the cascading behaviors before the systems decay into full-blown failure; 4) investigating "human vulnerability" associated with critical elements, and deriving metrics of human vulnerability, which will be further integrated into the mathematical models of interdependent systems to refine the detection of critical elements; 5) applying the proposed rigorous mathematical models and algorithms to the real-world interdependent networks in Florida, which consist of power grids, communication networks, and transportation networks, with an impact of human behavior.