In this paper we study a single machine scheduling problem with a position-based learning effect as well as taking into account the imprecise or fuzzy nature of due dates. The position-based learning effect of a job is assumed to be a function of its position. The fuzzy due date of a job is represented by the degree of satisfaction with respect to the job completion time. The objective is to find a job sequence maximizing the minimum satisfaction degree. We show the problem is polynomially solvable under certain agreeable conditions.