Improved efficiency measures through directional distance formulation of data envelopment analysis

This paper develops a model for an improved efficiency measure through directional distance formulation of data envelopment analysis. It deals with cases where positive and negative values co-exist as production factors. The developed model has properties such as scalar quantity for measuring efficiency and it identifies all sources of inefficiency. The measure is weakly monotonic; units and translation do not vary with respect to inputs and outputs. The proposed model, under some restrictions, reduces to basic Data Envelopment Analysis (DEA) models such as Charnes-Cooper-Rhodes (CCR), Banker-Charnes-Cooper (BCC), and a slack based model (SBM). In addition to the above, the proposed model includes the closest targets for a given inefficient unit to achieve efficiency with less effort. The proposed model is validated using a case study done on Information Technology firms operating in India.