Executive Summary

This project develops a framework for qualitative and quantitative analysis of urban region resilience based on identification and exercise of system metrics pertinent across infrastructure sectors and over geography and time. Complex system response following stress application (human- or nature-induced) will be assessed through study of performance response functions with a focus on the New York/New Jersey region.

Research Problem and Background

Over the past century, our nation has experienced dramatic changes in demographics, and existing sociotechnical systems have become more complex and increasingly networked. To complicate matters, our cyberphysical infrastructure has not been maintained, causing unexpected vulnerabilities and cascading failures (ASCE, 2009; AWWA, 2001). As extreme events frequency and magnitude of resulting disasters have increased, emergent behavior, unexpected performance response, and lack of resilience have been noted (Sanford Bernhardt and McNeil, 2008). While there is success in modeling complex response and predicting behaviors of our urban sociotechnical networks under stress, the models have grown so complex that data is not available to validate the model predictions (NRC 2009). It is clear that we need to understand our sociotechnical system dynamics and resilience at a fundamental level. We define resilience as the ability (sufficient capacity and/or flexibility) of a system to experience unexpected shocks or perturbations, and to respond and recover functionality at some acceptable level of performance or action. We have an urgent need for improved understanding of the genesis and evolution of resilience, in particular in urban transportation systems. This will allow us to build and enhance social and ecological capital and community resilience, as well as to increase system adaptive capacity (including self-organization) and improve the cost-effectiveness of investments in infrastructure systems. An interdisciplinary approach is needed that captures attributes of the complex systems in a region. This requires assembling varied and deep information reflecting current and future conditions, response and usage so that we can expand our knowledge and validate our discoveries and predictions for system performance response. We need to assemble and create information and modeling resources, develop a framework of variables and relationships that will support a cross-disciplinary and cross-sector exploration of resilience, and build knowledge as we develop and test theory and models. In the long term, this will allow us to answer important questions including: 

• What observations (evidence) can we make (identify) to indicate qualitatively whether a specific system or network will demonstrate resiliency?
• What metrics can be used to evaluate the capacity of a system or network for resilient response?
• How does resilience response develop, and what factors control or influence the development? Is it a process with thresholds, tipping points, state changes, or is it a continuous function?
• What can we understand about when investment or adaptive management is warranted to improve resiliency of a system or networks of interdependent systems?

The research proposed here will focus on identifying the basic metrics and models that can be used to develop representations of performance response that can be used to define resilience in urban environments, and to bring together data resources that can be investigated to understand and validate the interactive behavior of our complex transportation infrastructure systems..

Research Objectives

In order to generate the deep interdisciplinary knowledge and understanding of the urban world that will advance knowledge of resilience, we need to identify metrics for system performance representation. These metrics need to be applicable across modes of transportation systems. These metrics also need to be applicable to other sectors of infrastructure since our transportation systems are highly interdependent. With metrics identified, appropriate modeling and analytical tools need to be identified that can used to model single mode, multi-mode and multisector infrastructure systems. Finally, data resources need to be identified and assembled to support both fundamental inquiry, and also to validate interpretations and outcomes from modeling efforts. Therefore, this research is directed toward completing a thorough investigation of potential metrics, models and data resources that can be used to further the understanding of complex system behaviors and resilience. With these resources developed, future research can be conducted that will lead to fundamental investigations into performance response of real systems, and for systems of systems. Then we can begin to approach an understanding of community resilience that includes socio-economic, natural and builtenvironment systems interacting to produce behavioral and interacting responses that can be validated and applied to resilience prediction. The knowledge produced may well support the development of a theory of resilience by aggregating knowledge into important guiding/fundamental questions, identifying hypotheses and building scenarios and testing hypotheses against system and simulation data, and against experiments and testbeds.