Lu Huang

Urban rail transit (URT) train-door failures are safety-critical and can cause cascading service disruptions, yet existing emergency operation schemes (EOSs) are often static, difficult to adapt to evolving fault patterns, and hard to verify against updated regulations. This study proposes a retrieval-augmented large language model (LLM) framework for executable and evidence-traceable EOS generation. Multi-source heterogeneous incident evidence (structured work orders, operational impact records, and unstructured maintenance/dispatch narratives) is normalized into a structured incident representation, and a hybrid retriever (dense + BM25) with cross-encoder reranking selects compact regulatory clauses and historical cases under a fixed context budget. The generator is fine-tuned with structured objectives to enforce schema compliance, role assignment, and citation grounding. Experiments on 776 passenger-door incidents from Shanghai URT (2019–2024) show that Hybrid + rerank achieves the best retrieval quality (Recall@5 = 0.78; Coverage@B = 0.71; FirstHit/B = 0.46). For generation, the full setting improves operational usability, reaching SchemaPass = 0.88, RoleAcc = 0.91, CiteCov = 0.73, and UsableAns = 0.83, compared with 0.15 UsableAns for a pure LLM baseline and 0.26 for prompting with RAG only. These results indicate that combining high-utility retrieval with structure- and citation-aware fine-tuning substantially improves the executability and verifiability of safety-critical operation schemes.

Climate-induced natural hazards are increasingly disrupting metro operations in megacities, necessitating robust and generalizable frameworks for system-wide resilience. While current studies often treat infrastructure degradation, operational adjustment, and passenger flow redistribution as separate problems, this study proposes a resilience-oriented decision framework that couples these universal processes together to address diverse disruptive events. Taking extreme rainfall as a critical representative scenario, a multi-objective recovery optimization model is developed to jointly optimize repair resource cost and average section saturation. Resilience is quantified through the demand satisfaction ratio over the disruption–recovery process, ensuring the framework’s applicability across different hazard types. A case study of the Shanghai metro system under a real extreme rainfall event demonstrates the model’s efficacy in capturing complex system dynamics. Results show a clear Pareto trade-off between repair resource cost and average section saturation, while increasing service capacity on adjacent lines improves the Pareto frontier. Prioritizing repairs on lines with the fewest damaged sections effectively reduces network saturation by restoring corridor throughput. The resilience curve proves that higher repair resources not only shorten recovery time but also raise the minimum demand satisfaction ratio. These findings provide a scalable methodology for designing resilient metro recovery strategies under various climate-related disruptions globally.

Intercity railway is an important system for the development of urban agglomeration, and the site selection of the Line Management Department of Intercity Railway (LMDIR) is a significant task for the railway department when constructing intercity railways. Owing to the complexity and uncertainty during the selection process, we constructed a multiple expert multi-criteria decision making (MEMCDM) method to provide a rational decision support model for a railway management department in the Fermatean cubic fuzzy set context. In this regard, an innovative extension called Fermatean cubic fuzzy sets (FCFSs) that integrates Fermatean fuzzy sets (FFSs) and cubic sets; several basic theories of FCFSs, including the score and accuracy functions; and distance measures are also given. Then, a series of Fermatean cubic fuzzy Hamacher operators are put forward to flexibly fuse Fermatean cubic fuzzy information, and the corresponding valuable characteristics of these operators are also investigated. Thirdly, the Fermatean cubic fuzzy logarithmic percentage-change-driven objective weighting (LOPCOW) approach is recommended based on the score function to recognize the importance of criteria, and the similarity-based method is deployed to identify the expert weight information. Fourthly, a hybrid MEMCDM methodology integrating the proposed Fermatean cubic fuzzy Hamacher operators, the LOPCOW method, whose evaluation is based on distance from average solution (EDAS) method based on regret theory, is designed to ascertain the prioritization of the schemes. Consequently, an empirical test concerning the site selection of LMDIR is shown to validate the feasibility and usefulness of the designed MEMCDM approach. The analysis involving the sensibility test and comparison study with prior methods is displayed to emphasize the effectuality and advantages of the propounded method. The outcomes demonstrate that the hybrid method recommended in this research possesses superior robustness and feasibility to cope with complicated decision issues. The findings of this research show that the presented method can recommend more credible site selection of LMDIR when encountering uncertainties and abundant impact factors.