Abstract: The flexible job-shop scheduling problem (FJSP) is addressed and a hybrid genetic algorithm (GA) is proposed to solve this problem. A heuristic is used to create a set of relatively good schedules for the problem, from which the initial population is created by selecting the relatively good ones. Furthermore, priority rules are integrated into crossover, mutation, exchange, and selection to avoid that it converges to a local optimum. In this way, a better solution can be obtained. Benchmark FJSPs are used to test the proposed method and comparison with some existing approaches is done. Results show that it is effective and efficient for solving FJSPs.