Evaluation of port disruption impacts in the global liner shipping network

The global container shipping network is vital to international trade. Current techniques for its vulnerability assessment are constrained due to the lack of historical disruption data and computational limitations due to typical network sizes. We address these modelling challenges by developing a new framework, composed by a game-theoretic attacker-defender model and a cost-based container assignment model that can identify systemic vulnerabilities in the network. Given its focus on logic and structure, the proposed framework has minimal input data requirements and does not rely on the presence of extensive historical disruption data. Numerical implementations are carried in a global-scale liner network where disruptions occur in Europe's main container ports. Model outputs are used to establish performance baselines for the network and illustrate the differences in regional vulnerability levels and port criticality rankings with different disruption magnitudes and flow diversion strategies. Sensitivity analysis of these outputs identifies network components that are more susceptible to lower levels of disruption which are more common in practice and evaluates the effectiveness of component-level interventions seeking to increase the resilience of the system.