This paper puts forward a multi-objective hybrid difference optimization algorithm to solve multi-objective flow-shop scheduling problem (FSP). The hybrid algorithm inherits the merits of differential evolution vector operation, and makes dynamic adjustments to the search direction based on historical data. However, the basic differential evolution algorithm is prone to the local optimum trap, due to the low population diversity in the later stage of evolution. To solve the problem, a hybrid sampling strategy was introduced obtain the distribution information of solution sets and to design the mutation operator of differential evolution, thus improving the convergence of the hybrid algorithm. Finally, our algorithm was applied to solve FSPs through simulation. The simulation results show... that our algorithm greatly outperformed the basic multi-objective evolutionary algorithm in convergence and distribution performance.