EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Train Operations Simulation and Evaluation in a Railway Marshalling Yard: A Digital-Twin Based Multi-Agent Framework

Shujin Mao, Hexin Hu, Gongyuan Lu, Hongxiang Zhang, Chengrong Wei

Systems · 23.10.2026 · 14 (11) · 1346

DOI: 10.3390/systems14111346 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Systems, online erschienen am 23.10.2026. Verfasst von Shujin Mao und 4 weiteren Autor:innen. Die Arbeit nennt 21 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministry of Science and Technology of the People's Republic of China, China Railway Shanghai Group Co., Ltd..

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Large railway marshalling yards involve tightly coupled train movements, railcar classification processes, shunting operations, and infrastructure occupation, making it difficult to evaluate operational plans in a static manner alone. To address this issue, this study proposes a digital-twin-based multi-agent framework for operational simulation and performance evaluation in a railway marshalling yard. The framework integrates yard status data from the Computer Integrated Process System (CIPS) and train arrival and departure plans from the Train Dispatching Management System (TDMS) to construct a digital-twin operational data foundation that serves as real-time input for the simulation model. Trains, railcars, and locomotives are modeled as interacting agents, enabling accurate simulation of train arrival, disassembly, accumulation, assembly, and departure operations based on topological yard representations. Simulating train operation processes given by an operational plan in the simulation model therefore offers multi-level indicators to evaluate train and railcar throughput, sub-yard occupancy, railcar accumulation, and shunting locomotive workload. Implementations at a real-world large-scale complex railway marshalling yard in China show that the proposed framework can complete the simulation of a 4-h operational plan within only 7 s, capture the temporal and spatial evolution of over 60 trains and 3000 railcars, and deliver various performance indicators for bottleneck identification and throughput analysis.

Förderung

  • Ministry of Science and Technology of the People's Republic of China
  • China Railway Shanghai Group Co., Ltd.

Zitieren

APA

Mao, S., Hu, H., Lu, G., Zhang, H., & Wei, C. (2026). Train Operations Simulation and Evaluation in a Railway Marshalling Yard: A Digital-Twin Based Multi-Agent Framework. Systems, 14(11), 1346. https://doi.org/10.3390/systems14111346

Vancouver

Mao S, Hu H, Lu G, Zhang H, Wei C. Train Operations Simulation and Evaluation in a Railway Marshalling Yard: A Digital-Twin Based Multi-Agent Framework. Systems. 2026;14(11):1346. doi:10.3390/systems14111346

BibTeX
@article{Mao2026,
  author = {Mao, Shujin and Hu, Hexin and Lu, Gongyuan and Zhang, Hongxiang and Wei, Chengrong},
  title = {Train Operations Simulation and Evaluation in a Railway Marshalling Yard: A Digital-Twin Based Multi-Agent Framework},
  journal = {Systems},
  year = {2026},
  volume = {14},
  number = {11},
  pages = {1346},
  doi = {10.3390/systems14111346}
}
Kostenlos starten →