Evolutionary algorithm framework for optimizing truck scheduling in multi-dock truck cross-docking centers

• Published in: Evolutionary intelligence Vol. 18; no. 1
• Main Authors: Nogueira, Thiago Henrique, Coutinho, Felipe Provezano, Peixoto, Maria Gabriela Mendonça, Carrano, Eduardo Gontijo, Ravetti, Martín Gómez
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01-02-2025
• Subjects: Applications of Mathematics; Artificial Intelligence; Bioinformatics; Control; Engineering; Mathematical and Computational Engineering; Mechatronics; Research Paper; Robotics; Statistical Physics and Dynamical Systems; Truck; Scheduling; Cross-docking; Evolutionary algorithm; Logistics
• ISSN: 1864-5909; 1864-5917
• DOI: 10.1007/s12065-024-00992-x

Cross-docking optimizes logistics by reducing storage and handling times in warehouses, where cargos are unloaded from inbound trucks and loaded directly onto outbound vehicles. This paper addresses the critical challenge of minimizing makespan in a multi-dock, parallel machine setting within cross-docking centers. We propose a novel framework that integrates an evolutionary algorithm (EA) with a machine learning (ML) based hyperparameter tuning mechanism to optimize truck sequencing. This study fills the research gap by offering a quantifiable improvement over traditional heuristic methods, delivering up to a 37% improvement in the GAP metric compared to state-of-the-art techniques. It also achieves a 30% enhancement over the PCH constructive heuristic used for generating initial solutions. Additionally, our ML-based tuning strategy provides up to a 21% performance increase over static tuning methods. Notably, these improvements are attained while maintaining a competitive computational time as reported in the literature.