Performance evaluation of hybrid genetic algorithm for assembly line scheduling

In this paper, we present a new approach to tackle scheduling problems in manufacturers' assembly line. Former solutions also provide results, yet they turn out to be ineffective or time-consuming. Our approach involves several new schemes in crossover and mutation, which reduce its processing time. In order to avoid premature convergences of the chromosomes, we choose a self-adaptive mutation rate and a clone-replacement approach. We then try an alternative called derivative tree crossover. Finally, the paper examines this algorithm's efficiency, which outperforms the previous methods