Skip to main content
Infrastructure Resilience Conference 2018

Full Program »

Dynamic Resilience Index and Survivability of Nodes In Correlation-Based Networks

Complex, interdependent infrastructure system has been one area, in which resilience studies gained considerable attention. Methods and approaches for the quantitative assessment of resilience have been abounded in literature. Although several methods are well-known and claimed to be representative, a generic approach is still missing to capture measurable elemental dimensions and dynamically characterise system resilient behaviour. We propose a generic performance-based metric for modelling individual dynamic resilience index. Our metric provides a significant illustration of system adaptability with time-series manner, which structured upon ``Drawdown-Drawup" cycles. A performance profile of a stock index is used as proxy model to demonstrate its capability. In addition, a dynamic process of correlation-based stock network evolution is visualised to study the node resilient behaviour and its associated survivability in complex network measures. We found that: (1) The dynamic resilience index could be one appropriate indicator for system performance, especially in capturing the adaptive resilient behaviour in functionality performance. (2) Drawdowns, drawups and dynamic resilience index have fat-tail distribution features, which may infer a slow-decay preference in dynamic system behaviour. (3) Survivability of listed equities has associative relationship with their network measures in long-term dynamic evolution process. For future works, the resilience of complex network should be further studied, by joining both topological measures and component performance (service or purpose performed by network components) into consideration, and it would require a thorough understanding and comprehensive metric to fulfil the purpose.

Junqing Tang
ETH Zurich, Future Resilient Systems, Singapore-ETH Centre.
China

Hans Rudolf Heinimann
ETH Zurich, Future Resilient Systems, Singapore-ETH Centre.
Switzerland

 

Powered by OpenConf®
Copyright ©2002-2016 Zakon Group LLC