A MAS architecture for dynamic, realtime rescheduling and learning applied to railway transportation

作者：

Highlights：

• Multi-agent model to reschedule disrupted railway traffic on a single track.

• Look-ahead dynamic ranking algorithm for dispatch priority decisions by bidding.

• Objective function is total delay minimization.

• Learning is incorporated as a proof-of concept to enhance rescheduling autonomy.

• Evaluation with real-life data and other resolution procedures.

摘要

•Multi-agent model to reschedule disrupted railway traffic on a single track.•Look-ahead dynamic ranking algorithm for dispatch priority decisions by bidding.•Objective function is total delay minimization.•Learning is incorporated as a proof-of concept to enhance rescheduling autonomy.•Evaluation with real-life data and other resolution procedures.

论文关键词：Conflict,Deadlocks,Disruption,Dynamic priority,Resolution,Multi-agent system (MAS),Single-track bi-directional railway traffic,Dynamic real-time rescheduling,Learning

论文评审过程：Available online 15 November 2014.

论文官网地址：https://doi.org/10.1016/j.eswa.2014.11.013