This paper reports on the conceptual development of an architecture for building intelligent fuzzy decision support systems (IFDSS). The changing nature of decision making today’s fast paced environment calls for new techniques for dealing with both quantitative and qualitative components of decision making. Soft computing tools like fuzzy reasoning offer much promise in this area. After reviewing the components of traditional decision support systems architecture, the paper describes the components of a new, more robust type of decision support system based on expert systems and fuzzy reasoning methodologies. The paper concludes with the application of this IFDSS methodology to the domain of educational assessment.