Scholarlink

On-time delivery is becoming more and more important for industrial companies, as their customers often require orders to meet specific deadlines. In this paper, we investigate the parallel machine scheduling problem with sequence-dependent setup times and due date windows. This problem has attracted attention due to its significance and relevance in real-world applications. To minimize total weighted earliness tardiness, we propose an efficient adaptive iterated greedy method enhanced with Q-learning (QIG). The performance of our approach is compared against six well selected metaheuristics from the literature. Computational experiments on a benchmark set of 600 instances demonstrate the efficiency of the proposed iterated greedy method.

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

Reinforcement Learning Based Iterated Greedy for Parallel Machine Scheduling with Weighted Earliness Tardiness

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Ahmed Missaoui,
  • Barry O’Sullivan

摘要

On-time delivery is becoming more and more important for industrial companies, as their customers often require orders to meet specific deadlines. In this paper, we investigate the parallel machine scheduling problem with sequence-dependent setup times and due date windows. This problem has attracted attention due to its significance and relevance in real-world applications. To minimize total weighted earliness tardiness, we propose an efficient adaptive iterated greedy method enhanced with Q-learning (QIG). The performance of our approach is compared against six well selected metaheuristics from the literature. Computational experiments on a benchmark set of 600 instances demonstrate the efficiency of the proposed iterated greedy method.