Finding common weights in DEA using a compromise solution approach

The weights generated by the common weights approach provide a common criterion for ranking the decision-making units (DMUs) in data envelopment analysis (DEA). Existing common weights models in DEA are either very complicated or unable to produce a full ranking for DMUs. This paper proposes a new compromise solution model to seek a common set of weights for full ranking for DMUs. The maximum inefficiency scores calculated from the standard DEA model are regarded as the anti-ideal solution for the DMUs to avoid. A common set of weights that produces the vector of inefficiency scores for the DMUs furthest to the anti-ideal solution is sought. The discrimination power of the new model is tested using two numerical examples and its potential application for fully ranking DMUs is illustrated.