Temporal modeling and resilience analysis of supply chain networks under cascading failures

Abstract

Enhancing supply chain resilience amid global disruptions remains a pressing challenge. While prior research has largely focused on static assessments that emphasize system degradation, the dynamic recovery and reconfiguration processes of Supply Chain Networks (SCNs) are often overlooked. To bridge this gap, we propose a Temporal Supply Chain Network (TSCN) model that captures key temporal dynamics, including firm entry and exit, relationship turnover, and operational state transitions. To simulate disruption propagation and recovery, we develop a cascading failure model driven by underload mechanisms and incorporate node-level recovery dynamics. Furthermore, we introduce a time-dependent resilience metric based on node reachability over adjustable time windows, enabling a granular assessment of network functionality restoration. Through simulations on both synthetic TSCNs and empirical data from the temporal global wheat trade network, we evaluate resilience under random and targeted disruptions. Our findings reveal that resilience is governed not only by static topology but also by the evolving interplay of node capacities, load distributions, and recovery potentials. The proposed framework offers a dynamic perspective on SCNs resilience, providing actionable insights for designing more adaptive and robust supply chain systems.