The worth of machine-part groups in cellular manufacturing systems is evaluated by different grouping measures available in the literature. Three of the well-known grouping efficiency measures will be used, analyzed and tested against two case studies taken from the literature to compare the effectiveness of cell formation. In this paper it is found that in both cases grouping efficiency measures may not lead to choose the optimal or best distribution. For that in the first case the designer will choose the optimal distribution based on optimality rather than efficiency. While in the second case his decision will be based on the constraints on the shop floor. Moreover, it is found that these measures have some deficiencies comparing to optimal manufacturing systems having the same sum of voids and exceptions