<ul><li>Column Generation (CG) is a popular method for enhancing computational efficiency in large scale Combinatorial Optimization problems.</li><li>A new approach combines CG with Graph Neural Network (GNN) and unsupervised learning to reduce the size of the Elementary Shortest Path Problem with Resource Constraints (ESPPRC).</li><li>The reduced problem is then solved using local search techniques, resulting in significant improvements in convergence compared to previous reduction techniques.</li><li>The method has shown over 9% improvement in objective values for larger instances of Capacitated Vehicle Routing Problems with Time Windows.</li></ul>

Graph Reduction with Unsupervised Learning in Column Generation: A Routing Application

Discover more