Scholarlink

Reinforcement learning (RL) has recently been applied to solve railway timetabling problems. In this paper, we provide a detailed overview of the state-of-the-art research for RL in railway timetabling. Specifically, we categorize RL into basic RL and deep reinforcement learning (DRL), and further divide the research of railway timetabling into scheduling and rescheduling for exhaustively review and discussion. The present research on RL in railway timetabling is still in the primary stage and the scale of problems that can be solved is still limited. However, the applications of RL shows great promise and excitement, with significant potential for addressing various challenges in railway planning and management in the future.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Literature Review of Reinforcement Learning in Railway Timetabling

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Yan Wang,
  • Jiaming Fan,
  • Ruihao Han,
  • Angyang Chen,
  • Junyuan He,
  • Bo Li,
  • Peiyu Zhou,
  • Junren Wei

摘要

Reinforcement learning (RL) has recently been applied to solve railway timetabling problems. In this paper, we provide a detailed overview of the state-of-the-art research for RL in railway timetabling. Specifically, we categorize RL into basic RL and deep reinforcement learning (DRL), and further divide the research of railway timetabling into scheduling and rescheduling for exhaustively review and discussion. The present research on RL in railway timetabling is still in the primary stage and the scale of problems that can be solved is still limited. However, the applications of RL shows great promise and excitement, with significant potential for addressing various challenges in railway planning and management in the future.