Analysis of axioms and assumptions of data envelopment analysis: application for efficiency measurement in project management contexts

Abstract

Data envelopment analysis (DEA) is increasingly used to measure projects' efficiency, as recent research contributions indicate. However, most studies in project management take the axioms and assumptions underlying DEA for granted or do not pay attention to the type of data they are working with. Lack of attention to these important factors leads to selection of inappropriate DEA models and, consequently, produces biased efficiency scores. In this paper, after arguing that DEA is an appropriate model for project efficiency measurement, the economic meaning of its axioms and assumptions is explained. We also explain how different data types require some modifications in the CCR model. As a result, a guideline is presented to help future project management scholars select an appropriate DEA method tailored for their specific situation. Further, to highlight the importance of paying attention to these issues, we empirically demonstrate the high sensitivity of DEA results to the applicability of underlying DEA axioms and assumptions.