For solving crane scheduling problem in steelmaking-continuous casting process, a hybrid method of integrating genetic algorithm and simulation was formulated to minimize the operation time of tasks by considering full ladles, empty ladles and auxiliary tasks as transportation tasks. The crane selected sequence was designed as chromosome coding. The initial population was generated by employing the crane selection rules in the simulation model, and then evolved optimized by using the genetic operations such as selection, crossover and mutation. The chromosome is evaluated through generating a feasible crane schedule by employing the attribute updating rules and collision eliminating rules in the simulation model. The new crane scheduling could be generated through simulation model based on the new species. Therefore, more excellent individuals will be produced along with the evolutionary process. Thus a better crane scheduling scheme could be obtained finally. In the process of initialization, the high quality initial population was generated in the simulation model. In the iterative process, simulation model also could generate the feasible crane scheduling schemes based on the given cranes selection sequence. To validate this model, experiments were conducted by using the production data in the casting span of a steel plant. The results demonstrate the feasibility and efficiency of this method, which provides a useful tool for crane scheduling in actual production.