Modern manufacturing enterprises are increasingly faced with fierce international competition and fluctuating market conditions. To gain flexibility from a functional manufacturing system and efficiency from mass production, group technology has emerged as an effective compromise between the two. Identification and grouping of parts, that share similar processes into cells, is the basic problem in the design of cellular manufacturing systems. The design of cellular manufacturing systems is a complex, multi- criteria and multi -step process which can have significant implication for the entire organization. The formation of machine cell and part families have a number of constraints such as number of cells, cell size, technological requirements, and any other constraints like cost and time. This research proposes an expert system called CAEGT, which controls a number of constraints during the formation of machine cells and part families by using three methods. The system was tested in a manufacturing company and it gives an efficient results.