Scheduling is a problem that often occurs in the world of work that working hours divided into two, namely day and morning. To balance work hours requires more precision and long enough time. Genetic algorithms can create good scheduling, how to solve them through new structural chromosomes. Genetic algorithm is a population-based algorithm, when compared with other algorithms have advantages can be applied to the optimization of the problem large enough. Based on the results of this study, the genetic algorithm is able to schedule well. Workers get balanced results, three days into the afternoon, three days into the morning, and one day off. By the size of iteration parameters 150, individual 100, the probability of crossover 0.9 and the mutation probability 0.04, that is, the result of the fitness value 100%.