A Unified Parallel DEA Model and Efficiency Modeling of Multi-Activity and/or Non-Homogeneous Activity

Abstract

Data envelopment analysis (DEA), as originally proposed, is a methodology for evaluating the relative efficiencies of peer decision making units (DMUs) under some general assumptions. DEA models with non-homogeneous DMUs and multi-activity structures are two different subjects referring to relaxing various assumptions. In this paper, we show that these two formulations are both derived by embedding the corresponding process into a general parallel DEA model. Furthermore, following the parallel DEA framework, general DEA models for multi-activity and non-homogeneity are proposed, which are able to handle many situations where different aspects of non-homogeneity or multi-activities exist. This study provides important insights into the existing DEA models for non-homogeneity and multi-activity.