Scholarlink

<p>Incorporating AI-powered algorithms into business operations is becoming inevitable to gain a comparative advantage in the global business landscape. In this paper, we minimize the total travel distance and time of automated guided vehicles (AGVs) by coordinating routes and waiting time, using a hybrid method that integrates deep reinforcement learning (DRL) and heuristic search in an automated container terminal (ACT). Modern technology and data analytics provide intelligent, automated solutions for managing port resources such as berths, AGVs, quay cranes, and yard cranes. The deep deterministic policy gradient (DDPG) and advanced A-star search algorithms are employed, enhancing stochastic sequential decision-making for integrated scheduling and routing of port equipment. Our test results confirm the effectiveness of the hybrid approach for optimizing port resources and services. Decision-making strategies driven by a hybrid algorithm can optimize container handling and improve operational efficiency for competent maritime logistics. This paradigm shift in business operations contributes to the growing field of intelligent port management by paving the way for adaptive and innovative solutions in the maritime logistics industry, enhancing performance and productivity.</p>

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

Heuristic-assisted deep learning for integrated scheduling of multiple equipment in automated container terminals

  • 期刊论文

  • 发表日期：

    2025年8月14日

查看全文
  • Ho Van Roi,
  • Sam-Sang You,
  • Hwan-Seong Kim,
  • Le Ngoc Bao Long,
  • Truong Ngoc Cuong,
  • Duy Anh Nguyen

摘要

Incorporating AI-powered algorithms into business operations is becoming inevitable to gain a comparative advantage in the global business landscape. In this paper, we minimize the total travel distance and time of automated guided vehicles (AGVs) by coordinating routes and waiting time, using a hybrid method that integrates deep reinforcement learning (DRL) and heuristic search in an automated container terminal (ACT). Modern technology and data analytics provide intelligent, automated solutions for managing port resources such as berths, AGVs, quay cranes, and yard cranes. The deep deterministic policy gradient (DDPG) and advanced A-star search algorithms are employed, enhancing stochastic sequential decision-making for integrated scheduling and routing of port equipment. Our test results confirm the effectiveness of the hybrid approach for optimizing port resources and services. Decision-making strategies driven by a hybrid algorithm can optimize container handling and improve operational efficiency for competent maritime logistics. This paradigm shift in business operations contributes to the growing field of intelligent port management by paving the way for adaptive and innovative solutions in the maritime logistics industry, enhancing performance and productivity.