Journal of Korean Society of Industrial and Systems Engineering (산업경영시스템학회지)

Society of Korea Industrial and System Engineering (한국산업경영시스템학회)
권호 표지
DOI QR코드

DOI QR Code

A Study on the Remaining Useful Life Prediction Performance Variation based on Identification and Selection by using SHAP

SHAP를 활용한 중요변수 파악 및 선택에 따른 잔여유효수명 예측 성능 변동에 대한 연구

  • Yoon, Yeon Ah (Department of Industrial and Management Engineering Kyonggi University Graduate School) ;
  • Lee, Seung Hoon (Department of Industrial and Management Engineering Kyonggi University Graduate School) ;
  • Kim, Yong Soo (Department of Industrial and Management Engineering Kyonggi University)
  • 윤연아 (경기대학교 일반대학원 산업경영공학과) ;
  • 이승훈 (경기대학교 일반대학원 산업경영공학과) ;
  • 김용수 (경기대학교 산업경영공학과)
  • Received : 2021.09.23
  • Accepted : 2021.11.29
  • Published : 2021.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

Recently, the importance of preventive maintenance has been emerging since failures in a complex system are automatically detected due to the development of artificial intelligence techniques and sensor technology. Therefore, prognostic and health management (PHM) is being actively studied, and prediction of the remaining useful life (RUL) of the system is being one of the most important tasks. A lot of researches has been conducted to predict the RUL. Deep learning models have been developed to improve prediction performance, but studies on identifying the importance of features are not carried out. It is very meaningful to extract and interpret features that affect failures while improving the predictive accuracy of RUL is important. In this paper, a total of six popular deep learning models were employed to predict the RUL, and identified important variables for each model through SHAP (Shapley Additive explanations) that one of the explainable artificial intelligence (XAI). Moreover, the fluctuations and trends of prediction performance according to the number of variables were identified. This paper can suggest the possibility of explainability of various deep learning models, and the application of XAI can be demonstrated. Also, through this proposed method, it is expected that the possibility of utilizing SHAP as a feature selection method.

Keywords

Acknowledgement

This work was supported by Kyonggi University's Graduate Research Assistantship 2021.

References

  1. Adadi, A. and Berrada, M., Peeking inside the black-box: a survey on explainable artificial intelligence(XAI), IEEE Access, 2018, Vol. 6, pp. 52138-52160. https://doi.org/10.1109/access.2018.2870052
  2. Ahn, G., Yoo, J. H., Lee, S. H. and Kim, S. B., Explainable convolutional neural networks for multi-sensor data, Journal of the Korean Institute of Industrial Engineers, 2019, Vol. 45, No. 2.
  3. Babu, G.S., Zhao, P., and Li, X.L., Deep convoultional neural network based regression approach for estimation of remaining useful life, International Conference on Database Systems for Advanced Applications, 2016, pp. 214-228.
  4. Choi, Y. H. and Kim, H. S., Prognostics and health management for battery useful life prediction based on electrochemisty model: A tutorial, The Journal of Korean Institute of Communications and Information Sciences, 2017, pp. 939-949.
  5. Hancer, E., Xue, B., and Zhang, M., Differential evolution for filter feature selection based on information theory and feature ranking, Knowledge-based systems, 2018, Vol. 140, pp. 103-119. https://doi.org/10.1016/j.knosys.2017.10.028
  6. Hong, C.W., Lee, C., Lee, K., and Ko, M.S., Kim, D. E. and Hur, K., Remaining useful life prognosis for turbofan engine using explainable deep neural networks with dimensionality reduction, Sensors, 2020, Vol. 20, No. 22, pp. 6626. https://doi.org/10.3390/s20226626
  7. Hong, C.W. and Hur, K., Prognosis of the remaining useful life of a turbofan engine using deep neural network and pearson correlation coefficient, Journal of the KNST, 2021, Vol. 4, No. 1. pp. 078-083. https://doi.org/10.31818/JKNST.2021.03.4.1.78
  8. Jung, C.Y. and Lee, H.J., A comparative study of explainable AI techniques for process analysis, Journal of the Institute of Electronics and Information Engineers, 2020, Vol. 57, No. 8, pp. 51-59. https://doi.org/10.5573/ieie.2020.57.8.51
  9. Jung, S.J. and Hur, J.W., Deep learning approaches to RUL prediction of lithium-ion batteries, Journal of the Korean Society of Manufacturing Process Engineers, 2020, vol. 19, no. 12, pp. 21-27. https://doi.org/10.14775/ksmpe.2020.19.12.021
  10. Ko, G.H., Lim, G.M., and Cho, H.M., Survey on feature attribution methods in explainable AI, Journal of KIISE, 2020, pp. 1181-1191. https://doi.org/10.5626/jok.2020.47.12.1181
  11. Kwon, H.G., Lee, D.K., and Shin, M.S., Dynamic forecasts of bankruptcy with recurrent neural network model, Journal of Intelligence and Information Systems, 2017, Vol. 23, No. 3, pp. 139-153. https://doi.org/10.13088/JIIS.2017.23.3.139
  12. Laguna, J.O., Olaya, A.G., and Borrajo, D., A dynamic sliding window approach for activity recognition, International conference on user modeling, adaptation, and personlization, 2011, pp. 219-230.
  13. Li, J., Li, X., and He, D., A directed acyclic graph network combined with CNN and LSTM for remaining useful life prediction, IEEE Access, 2019, Vol. 7, pp. 75464-75475. https://doi.org/10.1109/access.2019.2919566
  14. Wang, J. Wen, G., Yang, S., and Liu, Y., Remaining useful life estimation in prognostics using deep bidrectional lstm neural network, IEEE, In 2018 prognostics and system health management conference(PHM-chongqing), 2018, pp. 1037-1042.
  15. Marcilio, W.E. and Eler, D.M., From explanations to feature selection: accessing SHAP values as feature selection mechanism, IEEE, In 2020 33rd SIBGRAPI conference on Graphics, Patterns and Images(SIBGRAPI), 2020, pp. 340-347.
  16. Oh, J.Y., Lee, Y.G., and Kim, G.B., Improvement of solar power forecasting using interpretation of artificial intelligence, The transactions of the Korean Institute of Electrical Engineers, 2020, Vol. 69, No. 7, pp. 1111-1116. https://doi.org/10.5370/kiee.2020.69.7.1111
  17. Seijo-Pardo, B., Porto-Diaz, I., Bolon-Canedo, V., and Alonso-Betanzos, A., Ensemble feature selection: homogeneous and heterogeneous approaches, Knowledge-Based Systems, 2017, Vol. 118, pp. 124-139. https://doi.org/10.1016/j.knosys.2016.11.017
  18. We, J.W., Choi, D.J., Lee, H.B., Lim, J.T., Lim, H.J., Bok, K.S., and Yoo, J.S., K-NN query optimization scheme based on machine learning using a dnn model, Journal of the Korea contents association, 2020, Vol. 20, No. 10, pp. 715-725. https://doi.org/10.5392/JKCA.2020.20.10.715
  19. Yu, S., Wu, Z., Zhu, X., and Pecht, M., A domain adaptive convolutional LSTM model for prognostic remaining useful life estimation under variant conditions, IEEE, In 2019 Prognostics and System Health Management Conference(PHM-Paris), pp. 130-137.
  20. Yun, Y.R., Kim, S.G., Cho, S.H., and Choi, J.H., Neural network based aircraft engine health management using C-mapss data, International Journal of Aerospace System Engineering, 2019, Vol. 13, No. 6, pp. 17-25.
  21. Zhang, A., Wang, H., Li, S., Cui, Y., Liu, Z., Yang, G., and Hu, J., Transfer learning with deep recurrent neural networks for remaining useful life estimation, Applied Sciences, 2018, Vol. 8, No. 12, pp. 2416.
  22. Zhang, X., Wu, G., Dong, Z., and Crawford, C., Embedded feature-selection support vector machine for driving pattern recognition, Journal of the Franklin Institute, 2015, Vol. 352, No. 2, pp. 669-685. https://doi.org/10.1016/j.jfranklin.2014.04.021