Abstract: In order to realize the multiple-objective joint optimization of flexible workshop, such as completion time, machine load, delivery delay time and workshop energy consumption, a multi-objective scheduling method for flexible workshop based on adaptive penalty MOEA/D is proposed. The flexible workshop scheduling problem with multiple production machines, multiple processing tasks and multiple processes is described, and an optimization model is established. A flexible job shop scheduling method based on MOEA/D algorithm is proposed. Aiming at the problem that constant penalty factors cannot meet the different adjustment requirements of different neighborhoods for convergence and chromosome diversity, a penalty factor that can adjust adaptively with the density of adjacent chromosomes is proposed, and a flexible workshop scheduling process based on adaptive penalty MOEA/D algorithm is formulated. In the production scheduling experiment with 8 machine tools and 8 workpieces with 28 processes, the Pareto frontier solution searched by the adaptive MOEA/D algorithm can dominate that of the standard MOEA/D and the improved NSGA-II algorithm; In the production experiment of the equal weight optimal solution, the completion time, machine load, delivery delay time and workshop energy consumption of the adaptive MOEA/D algorithm scheduling scheme are less than those of the standard MOEA/D algorithm and the improved NSGA-II algorithm. The experimental results show that the adaptive penalty MOEA/D algorithm is effective and superior in flexible workshop scheduling.