Open access
Oct 2026
A Supervised Learning Framework for Accelerating Solvers and Metaheuristics in Routing Problems
This study introduces a supervised machine learning (ML) framework that assigns a confidence score to each edge, indicating the likelihood of appearing in high‐quality solutions, and develops systematic strategies to reduce the search space by pruning low‐relevance edges.
Johannes Gückel, S. Voigt, P. Fontaine
· Networks · 0 citations