A new crossover for solving constraint satisfaction problems

摘要: In this paper we investigate the applicability of Genetic Algorithms (GAs) for solving Constraint Satisfaction Problems (CSPs). Despite some success GAs when tackling CSPs, they generally suffer from poor crossover operators. order to overcome limitation in practice, propose a novel specifically designed CSPs. Together with variable ordering heuristic and an integration into parallel architecture, proposed enables large hard problem instances as demonstrated by experimental tests conducted on randomly generated CSPs based model RB. We will indeed demonstrate, through these tests, that our method is superior known GA techniques addition, show are able compete efficient MAC-based Abscon 109 solver random instances.