Solving Fuzzy Graph Coloring Problem Using A Hybrid Slime Mould Algorithm

• Published in: 2025 International Conference on Intelligent and Cloud Computing (ICoICC) pp. 1 - 6
• Main Authors: Tripathy, Siba Prasada, Jena, Alok Kumar
• Format: Conference Proceeding
• Language: English
• Published: IEEE 02.05.2025
• Subjects: Benchmark testing; Color; Heuristic algorithms; Metaheuristics; Navigation; NP-hard problem; Registers; Search problems; Tuning; Uncertainty
• DOI: 10.1109/ICoICC64033.2025.11052044

The Graph Coloring Problem (GCP) is a traditional NP-hard problem that is conceptually related to the map coloring problem, which was studied extensively in the 20th century. GCP has been researched extensively over time, and many versions have been created. Although much advance has been made, it is still a subject of research, regularly motivating new strategies for the solution. The primary objective of the Graph Coloring Problem (GCP) is to assign colors to the vertices of a graph in a way that ensures no two connected vertices share the same color. In this study, we explore a variant of GCP in which the chromatic number is predefined, and we propose an innovative solution utilizing the Slime Mould Algorithm (SMA) to address this constraint-driven coloring challenge. The vertex colours are assigned dynamically according to fitness evaluation to strike a balance between exploration and exploitation based on algorithms inspired by the adaptive and oscillatory foraging behaviours of slime moulds. We further embed a fuzzy logic-based mechanism in the coloring process to adapt the coloring problem to the real world with uncertainty. We evaluate our method on the DIMACS benchmark dataset. Synthetics results show that the SMA-based approach yields the best results compared to others. This problem and different solutions have applications in several areas, e.g. scheduling systems, frequency assignment in mobile networks, coloring maps, testing of circuits, and register assignment in compilers.