Multi-objective resilience-oriented optimisation for the global container shipping network against cascading failures
提出多目标逐步优化框架，通过减少运输时间、缓解港口拥堵和保持网络结构完整性，帮助决策者维护全球集装箱航运网络应对级联故障的韧性。
• A new multi-objective stepwise optimisation framework for proactive resilience maintenance against cascading failures in the GCSN. • Solution to complex cargo redistribution in the GCSN concerning the economic efficiency, functionality, and structural completeness. • New load redistribution strategies for the GCSN to mitigate the cascading effects when being disrupted. • Extensive simulations and case studies for model validation and practical insights across 21 major global ports. Disruptive events at ports (e.g., epidemics, natural hazards and regional conflicts) continuously challenge the stability of cargo flows, leading to cascading failures that significantly undermine the resilience of global shipping networks. To address these challenges, this study proposes a new Multi-objective Stepwise Optimisation (MSO) framework that can aid decision-makers in maintaining resilience against cascading failures. Specifically, this study first formulates multiple objectives aimed at minimising adverse impacts on maritime stakeholders by reducing transit time, alleviating port overload, and preserving the network’s structural completeness. Then, to explore ideal load redistribution strategies mitigating the cascading effects, a Stepwise Cascading Mitigation (SCM) model is newly developed. In this model, all feasible target ports are identified, followed by an iterative algorithm applied to determine the equilibrium volumes of load redistributed to each target. An evolutionary procedure is then designed to ensure renewal of diverse solutions and reduce computational complexity. By simulating the entire cascading process, multi-dimensional reductions in shipping network resilience are eventually assessed. Taking the Global Container Shipping Network (GCSN) as a case study, comprehensive experiments, alongside targeted analyses of major international ports, are conducted to validate the effectiveness and superiority of the MSO framework over four benchmarking methods. Sensitivity analysis results further reveal that maintaining appropriate redundancy across the network, combined with the proposed optimal redistribution strategies, can effectively mitigate the adverse impacts of cascading failures on system resilience. Therefore, this study provides stakeholders with adaptable emergency response protocols to alleviate excessive congestion at critical ports, ensuring the timely and reliable movement of goods, thereby proactively protecting the overall robustness and resilience of global supply chains.