Energy efficiency evaluation and energy saving based on DEA integrated affinity propagation clustering: Case study of complex petrochemical industries

Data envelopment analysis (DEA) has been widely used in the energy efficiency analysis of industrial production processes. However, the traditional DEA model is not high in the division of the efficiency value of decision making units (DMUs), and produces a large number of DMUs with an efficiency value equal to 1, making it difficult to identify their merits and demerits. Therefore, a novel DEA model based on the affinity propagation (AP) clustering algorithm (AP-DEA) is proposed. Through the AP clustering algorithm, high influence input data of the energy efficiency can be obtained. The merits and demerits of DMUs can then be identified with a high degree of discrimination to obtain better efficiency groups. Finally, the proposed model is applied to evaluate the energy efficiency and optimize the energy configuration of the ethylene and pure terephthalic acid (PTA) production processes in complex petrochemical industries. The experimental results show that this proposed model can improve the efficiency value discrimination of efficiency values by effective DMUs better than the traditional DEA. Moreover, the energy saving potentials of ethylene and PTA production systems are approximately 0.49% and 24.74%, respectively, and the carbon emission reduction of the ethylene production system is approximately 10.04%. (C) 2019 Elsevier Ltd. All rights reserved.