CIRC: Planning-C: Synergistic Graph Flow Analytics: An Integrated Infrastructure for Bridging Complexity, Fragmentation, and Interdisciplinary Gaps

Network dynamics manifest themselves across a wide spectrum of domains, from the propagation of rumors over social networks and the transmission of diseases through human interactions, to the transportation of goods via traffic routes, as well as the circulation of blood through the brain's intricate vasculature. However, despite shared similarities in their underlying physical laws, practitioners in each field have traditionally worked in isolation in the past.

This siloed approach has led to a duplication of research efforts, and a lack of understanding of the complex interplay among different network dynamics, such as the dissemination of rumors could amplify the spread of infectious diseases. This project's novelties lie in creating a unified framework to analyze dynamic behaviors across various fields, and developing a comprehensive supporting infrastructure.

Through a comparative study spanning multiple domains, this project offers a schema for "analogy learning" and "interplay learning" in interdisciplinary research, serving as an exemplar for exploring the synergies and interconnections among different networks. It opens new avenues for investigating the intricate effects among multiple disciplines. The project's impacts are not limited to individual scenarios such as epidemiology, circuits, transportation, and banking systems; they will also enhance our understanding of coupled dynamics, including the co-evolution of epidemics and rumors, traffic and power systems, and banks and supply chains, among others.

Consequently, it will enhance our comprehension of the robustness and vulnerability of interconnected societal infrastructure as a whole. Moreover, the project will foster new avenues for increased dialogue and knowledge exchange across various disciplines, promoting novel breakthroughs.

The project outlines a plan to advance the field of network dynamics through the development of a unified computational framework and an experimental platform. It addresses the fragmentation in network dynamics research by integrating various network types and bridging multiple knowledge domains. The central methodologies of this endeavor involve treating each distinct domain of network dynamics as a specific instance within a unified framework, standardizing methodologies by identifying common patterns across diverse network dynamics, and devising a systematic framework that encompasses the structural, routing, and diffusion dimensions of networks.

This approach fosters interdisciplinary cooperation, drawing insights from various fields. The project will analyze both similarities and differences among these dynamics, revealing generalizations through shared principles observed in both virtual dynamics (such as information dynamics) and physical dynamics (such as traffic dynamics), as well as homogeneous and heterogeneous dynamics.

To engage the community in this emerging topic, the approach begins with theoretical framework prototypes, drawing on concepts from computational fluid dynamics and electrical network theory. Furthermore, the project aims to build upon prior work on open datasets and benchmarks, including the XFlow library and the COPE-ID platform, thereby establishing new platforms for network dynamics.

This research advances the technical aspects of network dynamics by constructing a convenient working platform while promoting interdisciplinary collaboration and dialogue, consequently encouraging further interdisciplinary research. Additionally, its open-source nature aims to broadly benefit educational outreach and influence various societal sectors, making significant contributions to network management and research.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.