Efficient Graph Coloring by Evolutionary Algorithms

In this paper we present an Evolutionary Algorithm (EA) for the Graph Coloring Problem (GCP). GCP has many important applications in VLSI CAD. The problem in general is known to be NP-hard. Our EA approach incorporates problem specific knowledge, i.e. we make use of different heuristics to obtain efficient solutions. We perform a large set of experiments and compare our EA to state-of-the-art heuristics. It turns out that the combination of EA methods with “classical” heuristics clearly outperforms previous approaches.