In the plant-wide hierarchical optimization for mineral and metallurgical processes, the technical indices are numerous, and a certain coupling relationship exists among them. Aiming at this problem, a plant-wide hierarchical optimization method based on the decomposition of coupling relationship between technical indices is proposed. On the basis of the plant-wide hierarchical optimization, the proposed method uses two metrics to decompose the technical indices, and then the procedure-level optimization problem is converted into the corresponding sub-problems. This method can simplify the original procedure-level optimization problem and improve the efficiency of solving the plant-wide hierarchical optimization problem for mineral and metallurgical processes, which has a strong global convergence. Simulation results show that the proposed optimization method has better performance in comparison with the conventional plant-wide optimization method and the pure plant-wide hierarchical optimization method.