Learning to reschedule platforms: A graph neural network based deep reinforcement learning method for the train platforming and rescheduling problem

•A novel deep reinforcement learning framework for the Train Platforming and Rescheduling Problem.•Integrating the microscopic simulation into the deep reinforcement learning framework.•High solution quality competitive with modern heuristics.•High and stable solving efficiency, regardless of the se...

Celý popis

Podrobná bibliografie

Vydáno v:	Transportation research. Part C, Emerging technologies Ročník 183; s. 105453
Hlavní autoři:	Zhang, Hongxiang, D’Ariano, Andrea, Zhu, Yongqiu, Wu, Yaoxin, Hu, Liuyang, Lu, Gongyuan
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	Elsevier Ltd 01.02.2026
Témata:	Deep reinforcement learning Graph neural network Neural combinatorial optimization Train platforming Train rescheduling Neural combinatorial optimization Train rescheduling Deep reinforcement learning Train platforming Graph neural network
ISSN:	0968-090X
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!