The main aim of this paper is to generalize fuzzy decision system for the Assessment of Students in an Engineering colleges or Universities. Roughly speaking, the student's category can be classified into three major classes: so- called Good, Average and Poor. The Teachers who are involved in the course & regularly delivering the lecture Classes, interacting with the students are the best persons to judge or assess a student. The various factors that are taken into consideration for the Student's Assessment are Appearance, Communication skills, IQ, Class Attendance, Punctuality, Alertness, Class Performance and Extra-curricular Activities. Periodically, the University Director is interested in studying the performance Assessment of the Students. Generally, meetings are held where the Teachers involved in the course give their opinion linguistically to the respective Students depending upon the various factors that are taken into consideration for Student's Assessment Process. Then a decision is taken to find out the best Student in the University. A requirement fuzzy subset has been formulated. A relationship is formed between the respective students and the teacher's opinion using the fuzzy subset representation. It is proposed to construct fuzzy decision set which includes the relative merits of all Students in the University. The Index of fuzziness of various Teacher involved in the course are measured and compared. A decision system has been developed using fuzzy distance approach to select the best Student.