We associate to a Variational Inequality an equivalent optimization problem by means of a suitable Gap function using the separation theory and a Lagrangean function whose saddle point is related to the solution of the Variational Inequality.