As the key enabling technology for grid perspective, grid workflow receives more and more attention of researchers and has been applied to many projects. Dynamic workflow generation is considered as one of the most important issues in grid workflow researches, because it has essential impact on system usability, flexibility and robustness. Currently, exploiting AI planning technology for workflow construction is emerging in some researches. Although it works well in specific problem domain, the lack of explicit knowledge jeopardizes its scalability and significance in grid workflow. In this work, we put forward process pattern as a vehicle for knowledge representation to capture process expertise in business level, and based on it, a planning approach is proposed for automated workflow generation; Compared with existing planning method, our pattern-oriented approach minimizes user-visible complexity and makes system more scalable and flexible by utilizing explicit knowledge support.