Performance evaluation of thermal power plants considering CO2 emission: A multistage PCA, clustering, game theory and data envelopment analysis

Abstract Data envelopment analysis is a relative performance assessment method to evaluate performance of a group of decision making units. Empirically, when the number of decision making units is insufficient, the classical data envelopment analysis models cannot discriminate the efficient units perfectly. To overcome this issue, in this paper, several mathematical approaches, including “multivariate data analysis techniques”, “game theory”, “Shannon entropy” and “the technique for order of preference by similarity to ideal solution”, are combined with data envelopment analysis. The proposed framework is applied to evaluate performance of Iranian thermal power plants. Inefficient performance of thermal power plants may end up in serious economic and environmental problems for example CO2 emission. Therefore, evaluating performance of thermal power plants and identifying their weaknesses in order to improve their performance is a necessity. The obtained results are analyzed, and some practical suggestions are provided to achieve sustainable performance and a cleaner production system.