Prediction of DMUs' Performance Using Improved SVMs Method Based on DEA

Document Type : Research Paper

Authors
1 Department of Mathematics, Yadegar-e- Emam Khomeini (RAH) Branch, Islamic Azad University, Tehran, Iran
2 Department of Computer, Abarkouh Branch, Islamic Azad University, Abarkouh, Iran
Abstract
Data Envelopment Analysis (DEA) is an effective method for measuring the efficiency of Decision Making Units (DMUs). DMUs have been generated based on production function, cost function and measuring efficiency for evaluation and selection. Prediction of DMUs' Performance (PDMUP) is a problem to select the best among DMUs based on input and output data of the DMUs. PDMUP includes evaluating of ranking and efficiency for DMUs. Improving the accuracy and computation time in measuring the efficiency of PDMUP have been two main challenges. Hence, in this paper, a method is proposed using DEA and Support Vector Machine (SVM) for predicting of DMUs. This method have been compared using datasets in earlier research for large DMUs. The experimental results of comparisons between PDMUP method and early methods demonstrate that the method can significantly improve the accuracy and reduce the computation time in predicting the efficiency of large DMUs.
Keywords

[1] S. Abe, Support vector machines for pattern classification, Springer, London, 2010.
[2] A. I. Ali, Streamlined computation for data envelopment analysis, European Journal of Operational Research 64 (1993), no. 1, 61–67.
[3] A. I. Ali, Computational aspects of dea, Data Envelopment Analysis: Theory, Methodology, and Applications, Springer Netherlands, 1994, pp. 63–88.
[4] A. I. Ali and L. M. Seiford, Computational accuracy and infinitesimals in data envelopment analysis, INFOR 31 (1993), no. 4, 290–297.
[5] R. D. Banker, A. Charnes, and W. W. Cooper, Some models for estimating technical and scale inefficiencies in data envelopment analysis, Management Science 30 (1984), no. 9, 1078–1092.
[6] R. S. Barr, K. A. Killgo, T. F. Siems, and S. Zimmel, Evaluating the productive efficiency and performance of US commercial banks, Managerial Finance 28 (2002), no. 8, 3–25.
[7] R. Carbonneau, K. Laframboise, and R. Vahidov, Application of machine learning techniques for supply chain demand forecasting, European Journal of Operational Research 184 (2008), no. 3, 1140–1154.
[8] D. Celebi and D. Bayraktar, An integrated neural network and data envelopment analysis for supplier evaluation under incomplete information, Expert Systems with Applications 35 (2008), no. 4, 1698–1710.
[9] C. -C. Chang and C. -J. Lin, LIBSVM: a library for support vector machines, ACM Transactions on Intelligent Systems and Technology 2 (2011), no. 3, 27: 1–27: 27.
[10] A. Charnes, W. W. Cooper, and E. Rhodes, Measuring the efficiency of decision making units, European Journal of Operational Research 2 (1978), no. 6, 429–444.
[11] C. -M. Chen and J. van Dalen, Measuring dynamic efficiency: theories and an integrated methodology, European Journal of Operational Research 203 (2010), no. 3, 749–760.
[12] W. W. Cooper, L. M. Seiford, and K. Tone, Data envelopment analysis: A comprehensive text with models, applications, references and DEA-Solver software, 2 ed. , Springer, New York, 2007.
[13] N. Cristianini and J. Shawe-Taylor, An introduction to support vector machines and other kernel-based learning methods, Cambridge University Press, Cambridge, 2000.
[14] J. H. Dulá, A computational study of DEA with massive data sets, Computers & Operations Research 35 (2008), no. 4, 1191–1203.
[15] A. Emrouznejad and E. Shale, A combined neural network and DEA for measuring efficiency of large scale datasets, Computers & Industrial Engineering 56 (2009), no. 1, 249–254.
[16] M. Farahmand, M. I. Desa, and M. Nilashi, A combined data envelopment analysis and support vector regression for efficiency evaluation of large decision making units, International Journal of Engineering and Technology (2014), 2310–2321.
[17] M. Farahmand, M. I. Desa, and M. Nilashi, Hybrid data envelopment analysis and neural networks for suppliers efficiency prediction and ranking, Proceedings of the International Conference on Recent Trends in Information and Communication Technologies, 2014, pp. 392–400.
[18] R. Färe and C. A. K. Lovell, Measuring the technical efficiency of production, Journal of Economic Theory 19 (1978), no. 1, 150–162.
[19] M. J. Farrell, The measurement of productive efficiency, Journal of the Royal Statistical Society: Series A (General) 120 (1957), no. 3, 253–281.
[20] H. Guosheng and Z. Guohong, Comparison on neural networks and support vector machines in suppliers' selection, Journal of Systems Engineering and Electronics 19 (2008), no. 2, 316–320.
[21] B. Jiang, W. Chen, H. Zhang, and W. Pan, Supplier's efficiency and performance evaluation using DEA-SVM approach, Journal of Software 8 (2013), no. 1, 25–30.
[22] C. Kao, Malmquist productivity index based on common-weights DEA: the case of taiwan forests after reorganization, Omega 38 (2010), no. 6, 484–491.
[23] Z. -J. Kong and J. -B. Xue, A comparative study of supplier selection based on support vector machine and RBF neural networks, International Asia Conference on Industrial Engineering and Management Innovation (IEMI2012) Proceedings, Springer, Berlin, 2013, pp. 917–926.
[24] T. C. Koopmans, Analysis of production as an efficient combination of activities, Activity Analysis of Production and Allocation (T. C. Koopmans, ed. ), Wiley, New York, 1951, pp. 33–97.
[25] R. K. Mishra and G. Patel, Supplier development strategies: a data envelopment analysis approach, Business Intelligence Journal 3 (2010), no. 1, 99–110.
[26] D. Ozdemir and G. T. Temur, DEA ANN approach in supplier evaluation system, World Academy of Science, Engineering and Technology 54 (2009), no. 538, 343–348.
[27] J. T. Pastor, J. L. Ruiz, and I. Sirvent, An enhanced DEA Russell graph efficiency measure, European Journal of Operational Research 115 (1999), no. 3, 596–607.
[28] S. Samoilenko and K. -M. Osei-Bryson, Determining sources of relative inefficiency in heterogeneous samples: methodology using cluster analysis, DEA and neural networks, European Journal of Operational Research 206 (2010), no. 2, 479–487.
[29] B. Schölkopf, A. J. Smola, R. C. Williamson, and P. L. Bartlett, New support vector algorithms, Neural Computation 12 (2000), no. 5, 1207–1245.
[30] B. Schölkopf and A. J. Smola, Learning with kernels: Support vector machines, regularization, optimization, and beyond, MIT Press, Cambridge, MA, 2002.
[31] V. Vapnik, The nature of statistical learning theory, Springer, New York, 1995.
[32] L. Wen and J. Li, Research of credit grade assessment for suppliers based on multi-layer SVM classifier, Proceedings of the Sixth International Conference on Intelligent Systems Design and Applications (ISDA'06) (Jinan, China), 2006.
[33] D. Wu, Supplier selection: a hybrid model using DEA, decision tree and neural network, Expert Systems with Applications 36 (2009), no. 5, 9105–9112.
[34] C. -C. Yeh, D. -J. Chi, and M. -F. Hsu, A hybrid approach of DEA, rough set and support vector machines for business failure prediction, Expert Systems with Applications 37 (2010), no. 2, 1535–1541.

Articles in Press, Accepted Manuscript
Available Online from 16 November 2026

  • Receive Date 21 May 2022
  • Accept Date 10 June 2022