Learning Large Neighborhood Search for Maritime Inventory Routing Optimization

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Chen, Rui, Liu, Defeng, Jiang, Nan, Gupta, Rishabh, Kilinc, Mustafa, Lodi, Andrea
Format:	Preprint
Published:	2025
Subjects:	Optimization and Control
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

_version_	1866908496527622144
author	Chen, Rui Liu, Defeng Jiang, Nan Gupta, Rishabh Kilinc, Mustafa Lodi, Andrea
author_facet	Chen, Rui Liu, Defeng Jiang, Nan Gupta, Rishabh Kilinc, Mustafa Lodi, Andrea
contents	Maritime inventory routing optimization is an important yet challenging combinatorial optimization problem. We propose a machine learning-based local search approach for finding feasible solutions of large-scale maritime inventory routing optimization problems. Given the combinatorial complexity of the problems, we integrate a graph neural network-based neighborhood selection method to enhance local search efficiency. Our approach enables a structured exploration of different neighborhoods by imitating an optimization-based expert neighborhood selection policy, improving solution quality while maintaining computational efficiency. Through extensive computational experiments on realistic instances, we demonstrate that our method outperforms direct mixed-integer programming as well as benchmark local search approaches in solution time and solution quality.
format	Preprint
id	arxiv_https___arxiv_org_abs_2502_15244
institution	arXiv
publishDate	2025
record_format	arxiv
spellingShingle	Learning Large Neighborhood Search for Maritime Inventory Routing Optimization Chen, Rui Liu, Defeng Jiang, Nan Gupta, Rishabh Kilinc, Mustafa Lodi, Andrea Optimization and Control Maritime inventory routing optimization is an important yet challenging combinatorial optimization problem. We propose a machine learning-based local search approach for finding feasible solutions of large-scale maritime inventory routing optimization problems. Given the combinatorial complexity of the problems, we integrate a graph neural network-based neighborhood selection method to enhance local search efficiency. Our approach enables a structured exploration of different neighborhoods by imitating an optimization-based expert neighborhood selection policy, improving solution quality while maintaining computational efficiency. Through extensive computational experiments on realistic instances, we demonstrate that our method outperforms direct mixed-integer programming as well as benchmark local search approaches in solution time and solution quality.
title	Learning Large Neighborhood Search for Maritime Inventory Routing Optimization
topic	Optimization and Control
url	https://arxiv.org/abs/2502.15244