This text makes a formal description for Course Timetabling Problem, and proposed a hybrid heuristic algorithm for course timetabling problem by combining simulated annealing with iterative local search algorithm. First, we generate an initial feasible solution based on graph coloring algorithm, and then apply the simulated annealing algorithm to find the optimal solution. In the process of annealing algorithm, we use two neighborhoods iteratively in order to escape from local optimum to search the global optimum. Computational results show that it signicantly improves the quality of solution.