Mitigating Cascading Failure with Adaptive Networking

The increase of a network connectivity may improve network performance, but at the same time, it may also increase the chance of extremely large risk contagion. If external shocks or excess loads at some agents are propagated to the other connected agents due to failure, the domino effects often come with disastrous consequences. How to prevent cascading failures due to external shocks is an important emerging issue. In this paper, we propose mechanisms of mitigating flow-based cascading failure. Our aim is to improve the network's resilience actively and topologically. In the scenario of how to increase cascade resilience actively, we provide a simple micro-foundation based on coordinated incentives to absorb external shocks in order to survive collectively. We propose two types of risk sharing protocols: The topology-based and non-topology-based risk sharing in which network topology plays an important role. These rules employ local sharing algorithms to achieve global shock balancing. The models of shock transfer are designed to investigate some stylized facts on how external or innate shocks tend to be allocated in a network, and how this allocation changes agents' failure probability. In the scenario of how to increase cascade resilience topologically, we provide a rewiring method in which a network is self-organizable to reduce the damage of cascading failure. Simulation results indicate that risk management and adaptive network may dramatically reduce the average size of large cascading failures.