The concept of a FDH and the associated method of FDH analysis were introduced by Deprins, Simar, and Tulkens (1984). Subsequently, in a number of papers, Tulkens and his associates (especially Tulkens [1993]) have further refined the methodology of FDH analysis within the broad framework of dominance analysis. FDH analysis provides a method of efficiency measurement without the assumption of convexity and relies on dominance relations between observed input–output bundles to measure efficiency. It is shown to be a special case of the BCC or CCR problems with additional {0, 1} constraints on the λj ’s.
Input-oriented FDH model
The input-oriented FDH model under the variable returns-to-scale assumption is expressed as:

Output-oriented FDH model
The output-oriented FDH model under the variable returns-to-scale assumption is expressed as: