Abstract: Simple genetic algorithms have poor stability, for they are prone to premature convergence. In order to overcome this disadvantage, a novel algorithm is proposed using a combination of uniform design and genetic operation. A mapping between the solution space of problems and the search space of the algorithm is established by coding, and then crossover operation, mutation operation and uniform design are performed to produce the next generation of solution candidates for iteration until convergence. The algorithm is tested with a typical testing function, and proved feasible. Compared with the simple genetic algorithms, the algorithm proposed in this paper has a higher precision and a faster convergence rate.