Authors:
Marouane El Abbassi, Karim Rhofir, Najib Mouhassine

Addresses:
Department of LaSTI Laboratory, National School of Applied Sciences, Sultan Moulay Slimane University, Beni Mellal, Béni Mellal-Khénifra, Morocco.

Abstract:

The Vehicle Routing Problem, also known as the VRP, is an important difficulty in the world of logistics. It entails determining the most effective routes for the delivery of goods or services in the most time-efficient manner. Traditional methods, which are based on heuristic and exact optimisation, have the objective of reducing trip distance and delivery time, but they struggle to deal with the dynamic nature of the conditions that exist in the real world environment. Recent developments in machine learning (ML) have made it possible to apply predictive, adaptive, and data-driven strategies to the question of how to solve the dynamic routing (DR) problem in virtual router protocol (VRP). The purpose of this study is to investigate the application of machine learning techniques, notably reinforcement learning (RL), supervised learning, and deep learning, in order to forecast demand patterns, find the ideal routes in real time, and alter the routing strategy as the conditions change. Solutions that are based on machine learning have the ability to streamline routes, minimise operating costs, and improve service quality by utilising high-dimensional data that is always growing. In order to highlight how machine learning has the potential to be a game-changer and the necessity of adaptive, real-time routing models to deal with the increasing demands and complexity of modern logistics networks, a case study that is supported by research that is currently accessible has been presented.

Keywords: Vehicle Routing Problem (VRP); Operational Expenses; Dynamic Routing (DR); Reinforcement Learning (RL); Optimisation and Operations; Customer Satisfaction.

Received on: 15/01/2025, Revised on: 25/03/2025, Accepted on: 30/05/2025, Published on: 07/12/2025

DOI: 10.69888/FTSIN.2025.000552

FMDB Transactions on Sustainable Intelligent Networks, 2025 Vol. 2 No. 4, Pages: 207-222

• Views : 113
• Downloads : 15

Download PDF