With the transformation of manufacturing towards intelligence and flexibility, the workshop scheduling problem, as a typical combinatorial optimization problem, is facing increasingly complex constraints and requirements. In the actual production process, order constraints and assembly constraints often significantly affect production efficiency and scheduling effectiveness. The traditional centralized scheduling method is no longer able to meet the requirements for flexibility, real-time performance, and efficiency in modern manufacturing environments. To solve this problem, a distributed workshop scheduling model based on order constraints and assembly constraints is proposed, aiming to improve the scheduling efficiency of the workshop under complex constraint conditions. For this model, an improved greedy algorithm was combined to solve it, and particle swarm optimization algorithm was introduced to enhance the local and global search capabilities of the distributed scheduling system, effectively addressing the computational complexity in large-scale production environments. The experimental results show that the research model performs well in optimizing scheduling, with a maximum completion time of 40.49 hours, an average processing waiting time of 1.46 hours, and a resource utilization rate of over 95%. These results indicate that the proposed solution algorithm can significantly shorten the production cycle, improve resource utilization, and have high real-time scheduling capabilities, making it particularly suitable for solving workshop scheduling problems with complex orders and assembly constraints.