An evolutionary algorithm for weighted graph coloring problem

Abstract

One of the optimization problems that is widely studied in the literature is the graph coloring problem. In this paper, we present an evolutionary algorithm for the weighted graph coloring problem that combines genetic algorithms with a local search technique. The proposed algorithm uses a novel pool-based crossover that gathers and combines domain specific information from parents and generates the next offspring. The performance of our algorithm is compared with two evolutionary algorithms in the literature, and the results of the synthetic benchmarks show that our algorithm significantly outperforms these algorithms with respect to total spill cost, total number of spilled nodes and execution time. © 2015 ACM.