EURO 2025 Leeds
Abstract Submission

2942. An Enhanced Metaheuristic-Machine Learning Algorithm for Multi-Trip Freight Routing in Portsmouth’s Consolidation Center

Invited abstract in session TD-58: Machine Learning and Artificial Intelligence, stream Vehicle Routing and Logistics.

Tuesday, 14:30-16:00
Room: Liberty 1.13

Authors (first author is the speaker)

1. Mohanad AL-Behadili
School of Mathematics and Physics, university of portsmouth
2. Djamila Ouelhadj
Maths, University of Portsmouth
3. Andrew bullock
School of Mathematics and Physics, University of Portsmouth
4. Graham Wall
Department of Mathematics, University of Portsmouth
5. Christopher Bayliss
School of Mathematics and Physics, University of Portsmouth
6. Graham Fletcher
University of Portsmouth

Abstract

This study provides the analysis of the potential sustainability impacts (economic, environmental, social) of implementing a freight consolidation centre on the outskirts of the city of Portsmouth. This research is part of the Solent Future Transport Zone project (FTZ) funded by the UK’s Department for Transport and led by Solent Transport. To analyse the impact of adopting a freight consolidation centre, the freight distribution is modeled as a Multi-Trip Capacitated Vehicle Routing Problem with Time Constraints (MCVRPT). To solve this problem, we propose an Enhanced Jaya optimisation method, integrating the Greedy Randomised Adaptive Search Procedure (GRASP) with Machine Learning (Jaya-GRASP-ML) to generate high-quality solutions. Machine learning guides the local search in the metaheuristic by adaptively selecting improvement operators based on their past performance to enhance the quality of the solutions. This approach considers practical constraints, such as vehicle capacity and time limitations, to enable multiple trips per vehicle. Furthermore, this study aims to validate the effectiveness of the model and Jaya-GRASP-ML through extensive computational experiments, assessing both the solution quality and the sustainability impacts.

Keywords

Status: accepted

Back to the list of papers