Nowadays, Problem-Based Learning (PBL for short) is a popular teaching strategy for teachers. In the PBL environment students can learn how to solve the real life problems by discussing with their teammates and developing their brainstorm map. How a teacher decides what kinds of teaching materials are appropriate to students according to their learning performance is an important issue. Therefore, this paper proposes a sequencing mechanism to provide the teacher recommendatory sequence of remedial teaching for each group. In this paper the Rough Set theory is used to discover the hidden data pattern in the tree-like concept map which is transformed from the brainstorm map constructed by the students.