A DEA cross-efficiency approach based on bargaining theory

Abstract

Cross-efficiency evaluation is a Data Envelopment Analysis (DEA) approach that enables the efficiency and ranking of Decision-Making Units (DMUs) to assess the possibility of idiosyncratic weighting schemes. However, the existence of multiple optimal weights in multiplier DEA models implies the non-uniqueness of the cross-efficiency evaluation. This is normally dealt with using secondary goals that enable the optimal weighting vectors to be chosen. In this paper, a new method to discriminate between optimal weighting profiles is proposed based on bargaining problems and the Kalai-Smorodinsky solution. In this approach, the input and output multipliers are agreed upon by the peer DMUs.