Deep learning-enhanced quantum optimization for integrated job scheduling in container terminals

This study explored the challenge of multi-equipment collaborative scheduling and management within container terminals, focusing on essential equipment such as reach stackers, automated guided vehicles (AGVs), quay cranes, and yard cranes. The job shop scheduling problem (JSSP) in container handlin...

Full description

Saved in:
Bibliographic Details
Published in:Engineering applications of artificial intelligence Vol. 148; p. 110431
Main Authors: Cuong, Truong Ngoc, Kim, Hwan-Seong, Bao Long, Le Ngoc, You, Sam-Sang, Tan, Nguyen Duy
Format: Journal Article
Language:English
Published: Elsevier Ltd 15.05.2025
Subjects:
Container terminal
Deep learning
Job scheduling
Multiple equipment
Quantum optimization algorithm
Deep learning
Multiple equipment
Container terminal
Job scheduling
Quantum optimization algorithm
ISSN:0952-1976
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This study explored the challenge of multi-equipment collaborative scheduling and management within container terminals, focusing on essential equipment such as reach stackers, automated guided vehicles (AGVs), quay cranes, and yard cranes. The job shop scheduling problem (JSSP) in container handling was modeled using a quadratic unconstrained binary optimization (QUBO) formulation. A novel scheduling approach is developed based on the synergistic combination of an artificial intelligence (AI) algorithm and quantum computing (QC) (or AI-QC scheduler). This strategy integrates deep reinforcement learning (DRL) and a quantum approximate optimization algorithm (QAOA) to minimize the makespan (total completion time) with accurate coordination of port equipment. Based on operating scenarios in port environments, the proposed AI-QC scheduler offers significant advantages over traditional methods such as first-in-first-out (FIFO), shortest processing time (SPT), and genetic algorithms (GA), particularly in handling stochastic disturbances in processing times. The hybrid technique based on deep learning, quantum computing, and (DRL-QAOA) has self-optimization capabilities, adjusting to changes in market demand in port operations. The AI-QC scheduler can outperform benchmark algorithms in allocating and scheduling port resources, achieving a lower makespan and enhanced operational efficiency across various scenarios. Robust performance indicates optimizing equipment utilization and enhancing terminal throughput, particularly in unpredictable scenarios. Hence, deep-learning AI-driven quantum optimization strategies can improve the competitiveness and efficiency of container terminal operations in a rapidly evolving global market. •AI-QC scheduler effectively models the JSSP in container terminals using QUBO.•Hybrid algorithm with DRL-QAOA is presented to optimize port equipment scheduling.•DRL-QAOA approach outperforms FIFO, SPT, and GA under stochastic processing times.•Detailed analysis demonstrates the efficacy of job scheduler in handling JSSP challenges.•Quantum scheduler enhances equipment utilization and productivity, providing visibility.
ISSN:0952-1976
DOI:10.1016/j.engappai.2025.110431