Asian Journal of Innovation and Policy

Asian Society for Innovation and Policy (아시아기술혁신학회)
권호 표지
DOI QR코드

DOI QR Code

Integrating Machine Learning with Data Envelopment Analysis for Enhanced R&D Efficiency & Optimizing Resource Allocation in the Specialized Field

  • Seokki Cha (R&D Strategy Center, Korea Institute of Science and Technology Information (KISTI)) ;
  • Kyunghwan Park (REM Policy Team, Korea Institute of Radiological and Medical Sciences (KIRAMS))
  • Received : 2024.02.20
  • Accepted : 2024.05.08
  • Published : 2024.04.30
Download PDF
⟨ Previous Next ⟩

Abstract

Enhancing the efficiency of research and development (R&D) is crucial for organizations to remain competitive and generate innovative solutions. Data Envelopment Analysis (DEA) has emerged as a powerful tool for evaluating R&D efficiency. However, traditional DEA models heavily rely on the selection of input and output variables, which can limit their effectiveness. To overcome this dependency and improve the robustness of DEA, this study proposes a novel methodology that integrates machine learning techniques with DEA for determining the most suitable input and output variables. The proposed approach is particularly relevant for specialized R&D fields, such as Radiation Emergency Medicine (REM). REM is a critical domain that deals with the medical and public health consequences of nuclear emergencies. The selection of REM as the focus of this study is motivated by several factors, including the unique challenges posed by the field, the potential for significant societal impact, and the need for efficient resource allocation in emergency situations. By leveraging machine learning algorithms, such as Support Vector Machines (SVM), the proposed methodology aims to identify the most relevant input and output variables for DEA in the context of REM. The integration of machine learning enables the DEA model to capture complex relationships and non-linearities in the data, leading to more accurate and reliable efficiency assessments. The effectiveness of the proposed methodology is demonstrated through a comprehensive evaluation using real-world REM data. The results highlight the superior performance of the machine learning-integrated DEA approach compared to traditional DEA models. This study contributes to the advancement of R&D efficiency assessment in specialized fields and provides valuable insights for decision-makers in REM and other critical domains.

Keywords

Acknowledgement

This work was supported by the Korea Institute of Science and Technology Information (K-23-L05-C02-S16).

References

  1. Abdullah, A., Saraswat, S., and Talib, F. (2023). Impact of Smart, Green, Resilient, and Lean Manufacturing System on SMEs' Performance: A Data Envelopment Analysis (DEA) Approach. Sustainability, 15, 1379.
  2. Altmann, A., et al. (2010). Permutation importance: a corrected feature importance measure. Bioinformatics, 26(10), 1340-1347.
  3. Banker, R.D., Charnes, A., and Cooper, W.W. (1984). Some models for estimating technical and scale inefficiencies in data envelopment analysis. Management science, 30(9), 1078-1092.
  4. Boussofiane, A., Dyson, R.G., and Thanassoulis, E. (1991). Applied data envelopment analysis. European journal of operational research, 52(1), 1-15.
  5. Cha, S., et al. (2020). Examining the effectiveness of initial response training program for nuclear emergency preparedness. International Journal of Radiation Research, 18(4), 715-721.
  6. Charnes, A., Cooper, W.W., and Rhodes, E. (1978). Measuring the efficiency of decision making units. European journal of operational research, 2(6), 429-444.
  7. Chen, L., and Wang, Y.M. (2021). Limitation and optimization of inputs and outputs in the inverse data envelopment analysis under variable returns to scale. Expert Systems with Applications, 183, 115344.
  8. Coleman, C.N., and Lurie, N. (2012). Emergency medical preparedness for radiological/nuclear incidents in the United States. Journal of Radiological Protection, 32(1), N27.
  9. Cooper, W.W., Seiford, L.M., and Tone, K. (2007). Data envelopment analysis: a comprehensive text with models, applications, references and DEA-solver software. New York: Springer.
  10. Degenhardt, F., Seifert, S., and Szymczak, S. (2019). Evaluation of variable selection methods for random forests and omics data sets. Briefings in bioinformatics, 20(2), 492-503.
  11. Disha, R.A., and Waheed, S. (2022). Performance analysis of machine learning models for intrusion detection system using Gini Impurity-based Weighted Random Forest (GIWRF) feature selection technique. Cybersecurity, 5(1), 1.
  12. Emrouznejad, A., and Yang, G.L. (2018). A survey and analysis of the first 40 years of scholarly literature in DEA: 1978-2016. Socio-economic planning sciences, 61, 4-8.
  13. Fang, L., and Zhang, C.Q. (2008). Resource allocation based on the DEA model. Journal of the operational Research Society, 59(8), 1136-1141.
  14. Golany, B., and Roll, Y. (1989). An application procedure for DEA. Omega, 17(3), 237-250.
  15. Goldstein, A., et al. (2015). Peeking inside the black box: Visualizing statistical learning with plots of individual conditional expectation. journal of Computational and Graphical Statistics, 24(1), 44-65.
  16. Guo, Y., Zhang, Z., and Tang, F. (2021). Feature selection with kernelized multi-class support vector machine. Pattern Recognition, 117, 107988.
  17. Inekwe, J.N. (2015). The contribution of R&D expenditure to economic growth in developing economies. Social indicators research, 124(3), 727-745.
  18. Kabir, S., and Farrokhvar, L. (2022). Nonlinear gradient-based feature selection for precise prediction of diseases. International Journal of Data Mining, Modelling and Management, 14(3), 248-268.
  19. Khan, N.M., et al. (2020). Analysis on improving the performance of machine learning models using feature selection technique. In Intelligent Systems Design and Applications: 18th International Conference on Intelligent Systems Design and Applications (ISDA 2018) held in Vellore, India, December 6-8, 2018, Volume 2 (pp. 69-77). Springer International Publishing.
  20. Lampe, H.W., and Hilgers, D. (2015). Trajectories of efficiency measurement: A bibliometric analysis of DEA and SFA. European journal of operational research, 240(1), 1-21.
  21. Lim, S.M. (2009). A Method for Select of Input-Output Factors in DEA. Industrial Engineering (IE interfaces), 22(1), 44-55.
  22. Lovell, C.A.K., and Pastor, J.T. (1999). Radial DEA models without inputs or without outputs. European Journal of operational research, 118(1), 46-51.
  23. Malik, M. (2012). Technopolitics: how technology shapes relations among nations. The Interface of Science, Technology & Security, 12, 21-29.
  24. Mienye, I.D., Sun, Y., and Wang, Z. (2019). Prediction performance of improved decision tree-based algorithms: a review. Procedia Manufacturing, 35, 698-703.
  25. Min, G., et al. (2023). Development of an Explainable Traffic Flow Prediction Model Using Shapley Additive Explanation and Elliptic Bivariate Relationship. Journal of Korean Society of Transportation, 41(2), 239-253.
  26. Min, J.H., and Kim, J.H. (1998). A selection process of input and output factors using partial efficiency in DEA. Journal of the Korean Operations Research and Management Science Society, 23(2), 201-217.
  27. Min, J.H., and Jeong, C.W. (2006). Nonparametric Approach to Bankruptcy Prediction: Developing Cross Peeling Technique integrating DEA and Negative DEA. Korean Management Review, 35(4), 1157-1180.
  28. Mirmozaffari, M., Yazdani, M., Boskabadi, A., Ahady Dolatsara, H., Kabirifar, K., and Amiri Golilarz, N. (2023). A Novel Machine Learning Approach Combined with Optimization Models for Eco-efficiency Evaluation. Appl. Sci., 10, 5210.
  29. Muthukrishnan, R., and Rohini, R. (2016). LASSO: A feature selection technique in predictive modeling for machine learning. In 2016 IEEE international conference on advances in computer applications (ICACA) (pp. 18-20). IEEE.
  30. Sengupta, J.K. (1999). A dynamic efficiency model using data envelopment analysis. International Journal of Production Economics, 62(3), 209-218.
  31. Shokrollahpour, E., Lotfi, F.H., and Zandieh, M. (2016). An integrated data envelopment analysis-artificial neural network approach for benchmarking of bank branches. Journal of Industrial Engineering International, 12(2), 137-143.
  32. Sinuany-Stern, Z., Mehrez, A., and Barboy, A. (1996). Academic departments efficiency via DEA. Computers & Operations Research, 23(5), 513-513.
  33. Smyth, G.K. (2005). Limma: linear models for microarray data. Bioinformatics and computational biology solutions using R and Bioconductor, 397-420.
  34. Tatsuzaki, H., et al. (2022). Reorganization of advanced radiation emergency medicine systems in Japan after the Fukushima Nuclear Power Plant accident. Environmental Advances, 8, 100197.
  35. Tofallis, C. (1996). Improving discernment in DEA using profiling. Omega, 24(3), 361-364.
  36. Wang, D.H.M., Yu, T.H.K., and Liu, H.Q. (2013). Heterogeneous effect of high-tech industrial R&D spending on economic growth. Journal of Business Research, 66(10), 1990-1993.