Stability of the classifications of returns to scale in data envelopment analysis: A case study of the set of public postal operators

A significant theme in data envelopment analysis (DEA) is the stability of returns to scale (RTS) classification of specific decision making unit (DMU) which is under observed production possibility set. In this study the observed DMUs are public postal operators (PPOs) in European Union member states and Serbia as a candidate country. We demonstrated a sensitivity analysis of the inefficient PPOs by DEA-based approach. The development of this analytical process is performed based on real world data set. The estimations and implications are derived from the empirical study by using the CCR RTS method and the most productive scale size concept (MPSS). First, we estimated the RTS classification of all observed PPOs. After that, we determined stability intervals for preserving the RTS classification for each CCR inefficient PPO under evaluation. Finally, scale efficient inputs and output targets for these PPOs are designated.