Rethinking Resilience Analytics

Abstract

The concept of "resilience analytics" has recently been proposed as a means to leverage the promise of big data to improve the resilience of interdependent critical infrastructure systems and the communities supported by them. Given recent advances in machine learning and other data-driven analytic techniques, as well as the prevalence of high-profile natural and man-made disasters, the temptation to pursue resilience analytics without question is almost overwhelming. Indeed, we find big data analytics capable to support resilience to rare, situational surprises captured in analytic models. Nonetheless, this article examines the efficacy of resilience analytics by answering a single motivating question: Can big data analytics help cyber-physical-social (CPS) systems adapt to surprise? This article explains the limitations of resilience analytics when critical infrastructure systems are challenged by fundamental surprises never conceived during model development. In these cases, adoption of resilience analytics may prove either useless for decision support or harmful by increasing dangers during unprecedented events. We demonstrate that these dangers are not limited to a single CPS context by highlighting the limits of analytic models during hurricanes, dam failures, blackouts, and stock market crashes. We conclude that resilience analytics alone are not able to adapt to the very events that motivate their use and may, ironically, make CPS systems more vulnerable. We present avenues for future research to address this deficiency, with emphasis on improvisation to adapt CPS systems to fundamental surprise.

In search of network resilience: An optimization‐based view

“…It should be noted that if the number of perturbations is large, then this may not be computationally effective and advanced sampling strategies may be necessary. In general, the model should be carefully constructed in order to provide the insights into the desired characteristics of the system and its limitations should be carefully understood [72].…”

“…In resilience literature, network extensibility refers to models and measures for, “how well systems stretch to handle surprises,” [220]. The term surprise is used to describe stressors where network operations and functional requirements may not be known a priori [72]. Methods for analyzing network robustness and rebound are limited for managing surprises because they only consider perturbations within the predefined set, , or produce results constrained by predefined network structure ( G = ( N , A )), design options, and objectives.…”

“…One is the impact of the 2001 WTC attack on the infrastructure networks of lower Manhattan as there was no prediction or contingency plan for when an entire section of the infrastructure network disappeared [146]. Another example occurred in October 2017 when Hurricane Ophelia hit Ireland, making it the easternmost Atlantic hurricane; it was completely unexpected that a hurricane could travel like Ophelia and impact Western Europe, and analytic tools designed to model hurricane winds had to be reconfigured in real time to adapt [72].…”

“…The development and application of network optimization models and algorithms can provide a critical perspective on the resilience of systems. In practice, the resilience of real‐world networks depends on both the construction and analysis of the optimization model itself and how decision makers interact with it [72]. Within the Networks research community, we identify three practical ways network optimization advances knowledge of resilience: (1) building aspects of resilience into a network optimization model; (2) using network optimization models to study the resilience of a system; and (3) adaptation by users of optimization models for resilience.…”

Risk Analysis: Celebrating the Accomplishments and Embracing Ongoing Challenges

“…This set of papers describes how to visualize overall system performance, relationships among variables, and predicted likelihoods of events. Prescriptive analytics suggest and evaluate proposed risk management solutions (Eisenberg, Seager, & Alderson, 2019;Zheng & Albert, 2019). Automatic collection, storage, and transmission of massive numbers of measurements offer the possibility of building data-informed resilience and enabling more careful human supervision, as well as encouraging machine learning and self-corrective mechanisms supervised by expert staff.…”