Effects of mutual traffic redistribution on robustness of interdependent networks to cascading failures under fluctuant load

Abstract

Failure in real interdependent networks can be induced by temporal and fluctuant overload on nodes. Moreover, failure of nodes will lead to loss of flow to some extent due to the decline of system delivery ability. In this paper, we propose a cascading failure model of interdependent networks based on mutual traffic redistribution under fluctuant load. In the model, the flow loss that is correlated with network existing resource is considered by defining a traffic loss parameter. The interdependence between coupled layers is realized by mutual traffic redistribution. Different types of artificial interdependent networks and a Bus-Metro network example are investigated. The results show larger node tolerance and larger traffic loss parameter can lead to higher network robustness. Moreover, cascading time synchronization of different layers induced by mutual traffic redistribution is observed. Although the interdependence usually makes the entire systems fragile, the optimal mutual flow redistribution rules that are beneficial to the whole network robustness can be obtained by our model. The model and results in this paper can provide some references on robust traffic flow assignment and network topology design for real interdependent systems.