Information sharing in DEA: A cooperative game theory approach

In this paper, a cooperative DEA game is proposed based on the idea that different organizations can gain if they share data on the input consumption and output production of their processing units. A larger data sample allows a more complete description of the Production Possibility Set and, consequently, of the efficient frontier. However, since not all organizations will benefit in the same way, it is convenient to use cooperative game theory in order to share the benefits of the cooperation. The proposed cooperative DEA game is shown to be subadditive and balanced. Models for computing the τ-value of the game and checking its stability, taking into account the exponential number of coalitions that exist when the number of organizations is large, are also presented. The approach is illustrated with a dataset from the literature.