Survival Traffic Ratio Analysis for Cascading Failure in Interdependent Networks

This paper proposes a new measure for the impacts of cascading failures in interdependent networks. This measure adds information about traffic paths, expressing the routes conveying traffic on a telecommunication network, to the existing model. The survival traffic ratio of this improved model is defined as the ratio of the surviving capacity of traffic which telecommunication network can ensure after the cascading failure has occurred. While the existing measure is unreasonable because it focuses on only the topology of networks, our measure is advantageous in terms of considering the traffic capacity degradation caused by cascading failure, which affects customer satisfaction. We implemented a software program to compute both the traditional measure and the survival traffic ratio, and the numerical results demonstrate usefulness of our measure as it surely clarified what nodes should be watched and given high priority in guarding against cascading failures.