Abstract: Automated guided vehicles (AGVs) are widely used for material transportation in flexible manufacturing workshops due to their advantages of high automation and strong handling capacity. To improve the energy efficiency and handling efficiency of multi-load AGVs, an energy-saving scheduling model considering the objectives of energy consumption and handling distance is established, and a two-stage optimal scheduling method is proposed. In the method, a combination of proximity allocation and path coincidence approach is designed to optimize task allocation among multi-load AGVs, and an improved genetic algorithm is proposed to optimize the sequence of loading and unloading tasks. Thus, the comprehensive performance of the handling distance and energy consumption of multi-load AGVs is improved. The case study shows that the proposed two-stage optimal scheduling method is effective.