Optimizing resilience in sea-rail container block train networks under cascading failure uncertainty

Abstract Optimizing resilience in sea-rail intermodal networks faces two key challenges: single-model frameworks are often unable to coordinate resilience responses to both hub and section failures, and deterministic failure assessments cannot capture the nonlinear growth of cascading failure probabilities. To address these limitations, this study develops a two-stage collaborative resilience optimization model for sea-rail networks. Structural reliability theory is further introduced to establish a cascading-risk quantification framework and is integrated with the resilience optimization model. In Stage 1, a stochastic programming model is used to dynamically redistribute loads under arbitrary hub-capacity degradation, with the objective of minimizing expected losses caused by hub failures while suppressing cascading failure propagation. In Stage 2, a transport-organization optimization model is applied to recovery planning under arbitrary section-capacity degradation, minimizing incremental losses and constructing a metric for evaluating resilience enhancement. Empirical and sensitivity analyses based on the New International Land-Sea Trade Corridor sea-rail intermodal network demonstrate that under hub-failure scenarios, the proposed model reduces total losses by an average 95.52% compared with conventional recovery strategies, while maintaining cascading failure probabilities of all alternative nodes within the preset threshold. For critical-section failures, the model achieves an average resilience enhancement of 22.57% across 864 simulated failure cases, with the maximum improvements reaching 57.60%. By establishing a resilience optimization paradigm that integrates recovery-strategy coordination with failure-risk control through interdisciplinary theoretical fusion, this study provides a novel theoretical foundation for enhancing multimodal transport network resilience under uncertain failure scenarios.