This study examines the crew pairing problem, which is one of the most comprehensive problems encountered in airline planning, to generate a set of crew pairings that has minimal cost, covers all flight legs and fulfils legal criteria. In addition, this study examines current research related to crew pairing optimization. The contribution of this study is developing heuristics based on an improved dynamic-based genetic algorithm, a deadhead-minimizing pairing search and a partial solution approach (less-costly alternative pairing search). This study proposes genetic algorithm variants and a memetic algorithm approach. In addition, computational results based on real-world data from a local airline company in Turkey are presented. The results demonstrate that the proposed approach can successfully handle medium sets of crew pairings and generate higher-quality solutions than previous methods.