In 2m factorial designs, to save on the number of runs, a fraction of the runs of the factorial design are used to test the significance of the effects. In the screening stage, when the number of factors is large, fractional main effects designs are used. For further savings and considering the principle of sparsity of active factors, supersaturated designs were proposed. In this paper, supersaturated designs and the E(S2) criterion are introduced as an optimality criterion for two-level supersaturated designs. Then, in order to determine the degree of desirability of these designs, a lower bound for E(S2) is presented.
