Integrating ICA and DEA Model for Efficiency Measurement

In this paper, a two-stage approach by integrating independent component analysis (ICA) and data envelopment analysis (DEA) model is proposed for efficiency measurement. Basically, the former aims to search for latent source signals where no relevant signal mixture mechanisms are available; and the latter is a method for measuring the relative efficiencies of decision making units (DMUs). The proposed method firstly uses ICA to the input variables for generating independent components. After selecting the ICs representing the independent sources of input variables, the selected ICs are used as new input variables to DEA model. The experiments from a hospital data have shown that the proposed method can reduce the number of efficient DMUs and improve the discriminatory capability of the efficiency measurement of DEA.