Scholarlink

We adopt reinforcement learning approach to automatically optimize waiting time, according to constraints, in a queueing system with limited flexibility by changing priorities and/or connections between servers and queues. We use constraint solver to generate priorities of queues that satisfy constraints. Simulation of queueing system with limited flexibility is used to obtain reward.

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

Customizing Characteristics of Multi-queue Multi-server Systems

  • 章节

  • 发表日期：

    2026年

查看全文
  • Olga Grinchtein,
  • Johan Karlsson

摘要

We adopt reinforcement learning approach to automatically optimize waiting time, according to constraints, in a queueing system with limited flexibility by changing priorities and/or connections between servers and queues. We use constraint solver to generate priorities of queues that satisfy constraints. Simulation of queueing system with limited flexibility is used to obtain reward.