Deep Learning Assisted Heuristic Tree Search for the Container Pre-marshalling Problem

TL;DR

This paper introduces DLTS, a deep learning-based heuristic tree search method that automates the creation of problem-specific heuristics for the container pre-marshalling problem, achieving near-optimal solutions efficiently.

Contribution

The paper presents a novel deep learning-assisted heuristic tree search approach that automates heuristic design for CPMP, outperforming existing methods in solution quality.

Findings

Achieves gaps to optimality below 2% on real-world instances.

Produces the highest quality heuristic solutions to date for CPMP.

Automates heuristic development using deep neural networks trained on optimal solutions.

Abstract

The container pre-marshalling problem (CPMP) is concerned with the re-ordering of containers in container terminals during off-peak times so that containers can be quickly retrieved when the port is busy. The problem has received significant attention in the literature and is addressed by a large number of exact and heuristic methods. Existing methods for the CPMP heavily rely on problem-specific components (e.g., proven lower bounds) that need to be developed by domain experts with knowledge of optimization techniques and a deep understanding of the problem at hand. With the goal to automate the costly and time-intensive design of heuristics for the CPMP, we propose a new method called Deep Learning Heuristic Tree Search (DLTS). It uses deep neural networks to learn solution strategies and lower bounds customized to the CPMP solely through analyzing existing (near-) optimal solutions to CPMP instances. The networks are then integrated into a tree search procedure to decide which branch to choose next and to prune the search tree. DLTS produces the highest quality heuristic solutions to the CPMP to date with gaps to optimality below 2% on real-world sized instances.