English

Towards designing robust coupled networks

Statistical Mechanics 2013-06-13 v3 Disordered Systems and Neural Networks Computational Physics Data Analysis, Statistics and Probability

Abstract

Natural and technological interdependent systems have been shown to be highly vulnerable due to cascading failures and an abrupt collapse of global connectivity under initial failure. Mitigating the risk by partial disconnection endangers their functionality. Here we propose a systematic strategy of selecting a minimum number of autonomous nodes that guarantee a smooth transition in robustness. Our method which is based on betweenness is tested on various examples including the famous 2003 electrical blackout of Italy. We show that, with this strategy, the necessary number of autonomous nodes can be reduced by a factor of five compared to a random choice. We also find that the transition to abrupt collapse follows tricritical scaling characterized by a set of exponents which is independent on the protection strategy.

Keywords

Cite

@article{arxiv.1106.3234,
  title  = {Towards designing robust coupled networks},
  author = {Christian M. Schneider and Nuri Yazdani and Nuno A. M. Araujo and Shlomo Havlin and Hans J. Herrmann},
  journal= {arXiv preprint arXiv:1106.3234},
  year   = {2013}
}
R2 v1 2026-06-21T18:23:22.405Z