Hard X-ray Modulation Telescope (HXMT) satellite is China's first space X-ray astronomy satellite. The planning and scheduling of HXMT observations is a complex multi-objective optimization problem, and the optimization object is to maximize simultaneously the scientific return and the observation efficiency. By analyzing the characteristics of HXMT observations and all the important factors in the efficient scheduling of a celestial source, such as celestial constraints and the additional scientific constraints, the HXMT long-term planning model is established. The greedy algorithm and genetic algorithm are combined to solve the model. Simulation results show that the approach is effective in solving the HXMT long-term planning problem, based on the analysis of the first year observations of Suzaku Mission.