Volume 18, Issue 3 (9-2021)                   jor 2021, 18(3): 73-92 | Back to browse issues page

XML Persian Abstract Print

Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:

Homayounfar M, Salahi F, Daneshvar A, Khatami Firouzabadi S M A. Applying a Hybrid DEA-ANN Approach in Evaluation of Balanced Efficiency of the Tehran Stock Exchange Pharmaceutical Companies. jor. 2021; 18 (3) :73-92
URL: http://jamlu.liau.ac.ir/article-1-1908-en.html
Faculty of Management and Accounting, Allameh Tabataba’i University, Tehran, Iran
Abstract:   (507 Views)
Performance evaluation is one of the most important ways to deal with the performance of organizations compared to their previous situation or other competitors, which makes it possible to take actions for performance improvement. In this study, using a hybrid data envelopment analysis and artificial neural network approach, the performance of the Iran Daroo pharmaceutical company has been evaluated. For this purpose, first by reviewing the literature, the company’s evaluation criteria were studied based on the balanced scorecard perspectives and then more important criteria were identified according to the experts' judgments. Accordingly, due to the necessity of studying and evaluating the performance of the Iran Daroo pharmaceutical company, its performance was evaluated over the 4-year period (2014-2017) using DEA, under the assumptions of constant return to scale and output-oriented approach. Next, using an artificial neural network algorithm, the performance of the company was predicted. Finally, the results of different artificial neural networks under different layers were examined and the results of the network with the most appropriate number of layers were compared with the other machine learning algorithms based on the accuracy, correctness, invocation and error indexes. The results illustrate better performance of the proposed model compared to the decision tree, naive Bayes, support vector machine and K- nearest neighbor algorithms.
Full-Text [PDF 798 kb]   (158 Downloads)    
Type of Study: Applicable | Subject: Special
Received: 2020/07/2 | Accepted: 2021/03/13

Add your comments about this article : Your username or Email:

Send email to the article author

Rights and permissions
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.