Recovery Modeling and Robustness Study after Cascading Failures in Logistics-based Networks

Abstract

In order to ensure the normal operation of the logistics network and improve the robustness of the network in case of cascading failure faults, this study introduces the concept of node recovery threshold based on the existing failure model to optimize the recovery time lag. Further, a criticality-first recovery model is proposed, which defines the capacity of a recovery node as a function related to its original capacity and opens the recovery node selectively to critical neighboring nodes to reduce the risk of secondary failure. Finally, a postal logistics network in Northwest China is used as a case study to investigate the recovery robustness of this network when it encounters cascading failures. The effects of various parameter variations on the network robustness are examined through experimental simulations. The experimental results show that timely recovery measures can significantly reduce the number of failed nodes when cascade failure occurs in the logistics network; setting a higher recovery threshold can reduce the impact of cascade failure on the network, effectively reduce the scale of network failure, and thus significantly improve the robustness of the logistics network; at the same time, increasing the capacity parameter can effectively delay the time of cascade failure in the network, and can slightly improve the robustness of the network.