【摘要】 铁路隧道环境封闭且地质条件复杂，火灾、塌方等突发事件易引发连锁灾害，对应急资源调度的鲁棒性提出了严峻挑战。针对现有救援方案在应对需求波动及运输时间不确定性方面的不足，研究提出了一种基于两阶段鲁棒优化的应急资源调度模型。该模型将决策过程分为灾前预防性资源配置与灾后应急动态调度2个阶段，引入鲁棒性系数与不确定性集合，系统刻画了多重不确定参数的影响，并采用列与约束生成算法进行高效求解。通过结合典型铁路隧道应急场景进行仿真实验，鲁棒调度方案通过18.9%的平均成本溢价，将路径网络扩展至8～9条，显著提升了系统的抗风险能力，验证了所提方法在保障救援效率的同时，能够兼顾经济性与可靠性。本研究不仅为铁路隧道应急管理提供了科学的决策支持工具，其方法体系还可拓展至多部门协同救援等复杂应用场景，具有较高的理论价值与实践意义。更多还原

【Abstract】 Railway tunnels have a closed environment and complex geological conditions. Emergencies such as fires and collapses are prone to triggering cascading disasters, posing severe challenges to the robustness of emergency resource scheduling. To address the shortcomings of existing rescue plans in addressing demand fluctuation and transportation time uncertainty, this paper proposed an emergency resource scheduling model based on two-stage robust optimization. The model divided the decisionmaking process into two stages: pre-disaster preventive resource allocation and post-disaster emergency dynamic scheduling. By introducing robustness coefficients and uncertainty sets, it systematically characterized the impact of multiple uncertain parameters and used the column and constraint generation algorithm for efficient solution. Simulation experiments were conducted by combining typical railway tunnel emergency scenarios. The robust scheduling scheme expands the path network to 8～9 routes with an average cost premium of 18.9%, significantly improving the risk resistance of the system, which verifies that the proposed method can ensure rescue efficiency while balancing economy and reliability. This paper not only provides a scientific decision support tool for emergency management of railway tunnels but also its method system can be extended to complex application scenarios such as multi-department collaborative rescue, which has high theoretical value and practical significance.更多还原