We develop the scientific foundations and associated algorithmic tools for compactly representing and analyzing heterogeneous data streams from sensor/information-rich networked dynamical systems. We take a unified dynamics-based and data-driven approach for the design of passive and active monitors for anomaly detection in such systems. Dynamical models naturally capture temporal (i.e., causal) relations within data streams. Moreover, one can use hybrid and networked dynamical models to capture, respectively, logical relations and interactions between different data sources. We study structural properties of networks and dynamics to understand fundamental limitations of anomaly detection from data. By recasting information extraction problem as a networked hybrid system identification problem, we bring to bear tools from computer science, system and control theory and convex optimization to efficiently and rigorously analyze and organize information. The applications include diagnostics, anomaly and change detection in critical infrastructure such as building management systems, transportation and energy networks.
In the next-generation power systems (Smart Grid), a large number of distributed energy devices (e.g., distributed generators, distributed energy storage, loads, smart meters) are connected to each other in an internet-like structure. Incorporating millions of new energy devices will require wide-ranging transformation of the nation’s aging electrical grid infrastructure. The key challenge is to efficiently manage a great amount of devices through distributed intelligence. The distributed grid intelligence (DGI) agent is the brain of distributed energy devices. DGI enables every single energy device to not only have a certain intelligence to achieve optimal management locally, but also coordinate with others to achieve a common goal. The massive volume of real-time data collected by DGI will help the grid operators gain a better understanding of a large-scale and highly dynamic power systems. In conventional power systems, the system operation is performed using purely centralized data storage and processing approaches. However, as the number of DGIs increases to more than hundreds of thousands, it is rather intuitive that the state-of-the-art centralized information processing architecture will no longer be sustainable under such big data explosion. The ongoing research work illustrates how advanced ideas from IT industry and power industry can be combined in a unique way. The proposed high-availability distributed file system and data processing framework can be easily tailored to support other data-intensive applications in a large-scale and complex power grids. For example, the proposed DGI nodes can be embedded into any distributed generators (e.g., roof-top PV panel), distributed energy storage devices (e.g., electric vehicle), and loads (e.g., smart home) in a future residential distribution system. If implemented successfully, we can translate Smart Grid with high-volume, high-velocity, and high-variety data to a completely distributed cyber-physical system architecture. In addition, the proposed work can be easily extended to support other cyber-physical system applications (e.g., intelligent transportation system).
Using GIS, visual analytics, and spatiotemporal modeling, Dr. Rybarczyk examines the utility of Big Data for gaining insight into the causal mechanisms that influence travel patterns and urban dynamics. In particular, his research sets out to provide a fuller understanding of â€œwhatâ€ and â€œwhereâ€ micro-scale conditions affect human sentiment and hence wayfinding ability, movement patterns, and travel mode-choices.
Recent works: Rybarczyk, G. and S. Banerjee. (2015) Visualizing active travel sentiment in an urban context, Journal of Transport and Health, 2(2): 30
My research focuses on developing and applying computational and data-enabled methodology in the broader area of sustainability. Main thrusts are as follows:
1. Human mobility dynamics. I am interested in mining large-scale real-world travel trajectory data to understand human mobility dynamics. This involves the processing and analyzing travel trajectory data, characterizing individual mobility patterns, and evaluating environmental impacts of transportation systems/technologies (e.g., electric vehicles, ride-sharing) based on individual mobility dynamics.
2. Global supply chains. Increasingly intensified international trade has created a connected global supply chain network. I am interested in understanding the structure of the global supply chain network and economic/environmental performance of nations.
3. Networked infrastructure systems. Many infrastructure systems (e.g., power grid, water supply infrastructure) are networked systems. I am interested in understanding the basic structural features of these systems and how they relate to the system-level properties (e.g., stability, resilience, sustainability).
Our research is concerned with evidence-based optimization, the idea of optimizing complex systems holistically, exploiting the unprecedented amount of available data. It is driven by an exciting convergence of ideas in big data, predictive analytics, and large-scale optimization (prescriptive analytics) that provide, for the first time, an opportunity to capture human dynamics, natural phenomena, and complex infrastructures in optimization models. We apply evidence-based optimization to challenging applications in environmental and social resilience, energy systems, marketing, social networks, and transportation. Key research topics include the integration of predictive (machine learning, simulation, stochastic approximation) and prescriptive analytics (optimization under uncertainty), as well as the integration of strategic, tactical, and operational models.
The video above is of a planned evacuation of 70,000 persons for a 1-100 year flood in the Hawkesbury-Nepean Region using both predictive and prescriptive analytics and large data sets for the terrain, the population, and the transportation network.
Sy Banerjee studies the impact of mobile devices on consumer behavior and on the processing of signals emerging from location-based Social Media posts. He teaches a MBA class on digital marketing and Big Data and collaborates with researchers from Business, GIS and Computer Science. Some of his recent works include:
Assessing Prime-Time for Geotargeting With Mobile Big Data, Sy Banerjee, Vijay Viswanathan, Kalyan Raman, Hao Ying, Journal of Marketing Analytics, 2013, Vol. 1(3), pp 174-183.
“Visualizing active travel sentiment in an urban context” with Greg Rybarczyk, International Conference on Transport & Health MINETA Transportation Institute, San Jose, California, July 2016.
“Assigning Geo-Relevance of Sentiments Mined from Location-Based Social Media Posts” with R. Sanborn and M. Farmer, in Advances in Intelligent Data Analysis XIV, LNCS
“Understanding In-Store Consumer Experiences via User Generated Content from Social Media”, working paper with Karthik Sridhar and Ashwin Aravindakshan
“Tweeted Customer Emotions as Currency for Competitive Performance: A Framework of Location-Based Social Media Listening”, working paper with Amit Poddar, Karthik Sridhar, Nanda Kumar
Amy Cohn, PhD, is an Associate Professor and Thurnau Professor in the Department of Industrial and Operations Engineering at the University of Michigan College of Engineering and Director of the Center for Healthcare Engineering and Patient Safety. Her primary research interest is in robust and integrated planning for large-scale systems, predominantly in healthcare and aviation applications. She also collaborates on projects in satellite communications, vehicle routing problems for hybrid fleets, and robust network design for power systems and related applications. Her primary teaching interest is in optimization techniques, at both the graduate and undergraduate level.