A combined machine learning algorithms and DEA method for measuring and predicting the efficiency of Chinese manufacturing listed companies

Data Envelopment Analysis (DEA) is a linear programming methodology for measuring the efficiency of Decision Making Units (DMUs) to improve organizational performance in private and public sectors. However, if new DMU needs be known its score, DEA analysis would have re-conducted, especially nowadays, datasets from many fields been growing rapidly real world, which will need huge amount computation. Following previous studies, this paper aims establish linkage between method machine learning (ML) algorithms, proposes an alternative way that combines with ML (ML-DEA) algorithms measure predict DMUs. Four ML-DEA are discussed, namely DEA-CCR model combined back-propagation neural network (BPNN-DEA), genetic algorithm (GA) integrated (GANN-DEA), support vector machines (SVM-DEA), improved (ISVM-DEA), respectively. To illustrate applicability above models, Chinese manufacturing listed companies 2016 measured, predicted compared scores obtained by model. The empirical results show average accuracy DMUs about 94%, comprehensive order four ranked good poor GANN-DEA, BPNN-DEA, ISVM-DEA, SVM-DEA.