The genetic algorithm is a powerful method to analyze many complex issue, especially in the optimization problems. The main challenges of genetic algorithm are premature convergence on local minimum and long convergence time. In this paper, a new genetic algorithm, named partial mutation in GA (PMGA) is proposed for tackling of these problems. PMGA is using elitism selection and improved mutation operator to increase diversity and efficiency. In this method, mutation probability is dynamic and executed on population when the chromosomes became stable. Mutation probability is determined by simulated annealing algorithm. In fact, the novel proposed method is considered as a combination of genetic algorithm and simulated annealing. The resulting performances show the successful and promising capabilities of the proposed algorithm.