Year: 2025
Type: dissertation
Abstract: The Set Covering Problem (SCP) is an NP-hard combinatorial optimization problem with applications in telecommunication, logistics, and transportation. Solving SCP is computationally challenging due to the combinatorial explosion of potential solutions, particularly for large instances. This study proposes a Graph Convolutional Network (GCN) to approximate optimal solutions for SCP. A bipartite graph representation of SCP is employed to predict node priority, serving as a warm start for the Gurobi solver. The GCN is trained on solutions from a classical greedy algorithm. The method integrates GCNs and Gurobi, unifying data-driven prediction and exact solver for better computation efficiency and scalability. Experimental evaluations on benchmark SCP instances show that the Hybrid Model reduces computational time and enhances Gurobi's performance, offering a robust framework for SCP and other large-scale combinatorial optimization problems. Ultimately, this research will help in my future work to predict and identify conservation regions in ecological conservation.
Author: Cobbinah, Hagar
Institution: University of Tennessee at Chattanooga
Language: English
Subfield: Computational Theory and Mathematics
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7028657694
Cites:
Cited by:
Related to: