In traditional printing enterprises, production scheduling is highly complex. This is due to the wide variety of products, scattered processes, uneven automation levels, frequent changes in plans, and low standardization. These factors turn scheduling into a multi-objective optimization problem with multiple constraints. Based on flexible demand for production scheduling in an intelligent printing workshop, this study integrates the learning effect of production personnel with the static scheduling problem found in printing workshops. Under the constraints of resource, operation, and process layers, this study establishes a multi-objective static scheduling optimization model for printing workshops, with the goals of minimizing the penalties for makespan of orders, total load time of equipment, maximum load time of bottleneck equipment, total production cost of orders, total quality failure rate of orders and total order delay/early completion time; and proposes an improved NSGA-II algorithm that is suitable for solving complex large-scale scheduling problems existing in printing workshops. The effectiveness and rationality of design model and algorithm are verified by combining with Kacem dataset, Brandimarte standard examples and actual enterprise cases.