Setup planning is an intermediate phase of process planning, and automating setup planning constitutes a critical component of computer-aided process planning. The common absence of a clearly defined optimality of setup plans has been a major obstacle to converting automatic setup planning research progress into real advancement in practice. The optimality of a setup plan is a multiple attributes problem associated with uncertainties, and human interaction is vital in clarifying the optimality structure in a dynamic setup planning environment. In this paper, a quantitative setup plan evaluation system driven by multiple-attributes utility analysis coupled with manufacturing error simulation is proposed to serve three purposes: (1) to clarify what is optimality of setup plans, (2) to provide a systematic method of evaluating setup plan alternatives quantitatively, and (3) to incorporate in existing automatic setup planning systems a human interface to fulfill their potential values.