Journal of the Korea Academia-Industrial cooperation Society (한국산학기술학회논문지)

The Korea Academia-Industrial cooperation Society (한국산학기술학회)
권호 표지
DOI QR코드

DOI QR Code

A Development of Defeat Prediction Model Using Machine Learning in Polyurethane Foaming Process for Automotive Seat

머신러닝을 활용한 자동차 시트용 폴리우레탄 발포공정의 불량 예측 모델 개발

  • 최낙훈 (공주대학교 대학원 미래융합공학과) ;
  • 오종석 (공주대학교 기계자동차공학부) ;
  • 안종록 (이스턴에프티씨 부설연구소) ;
  • 김기선 (이스턴에프티씨 부설연구소)
  • Received : 2021.02.19
  • Accepted : 2021.06.04
  • Published : 2021.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

With recent developments in the Fourth Industrial Revolution, the manufacturing industry has changed rapidly. Through key aspects of Fourth Industrial Revolution super-connections and super-intelligence, machine learning will be able to make fault predictions during the foam-making process. Polyol and isocyanate are components in polyurethane foam. There has been a lot of research that could affect the characteristics of the products, depending on the specific mixture ratio and temperature. Based on these characteristics, this study collects data from each factor during the foam-making process and applies them to machine learning in order to predict faults. The algorithms used in machine learning are the decision tree, kNN, and an ensemble algorithm, and these algorithms learn from 5,147 cases. Based on 1,000 pieces of data for validation, the learning results show up to 98.5% accuracy using the ensemble algorithm. Therefore, the results confirm the faults of currently produced parts by collecting real-time data from each factor during the foam-making process. Furthermore, control of each of the factors may improve the fault rate.

최근 4차 산업혁명으로 인해 제조업계에서는 제조업의 인공지능을 접목시켜 효율성을 극대화하는 스마트 팩토리 붐이 일어나고 있다. 특히 자동차 부품 제조 및 생산에 널리 적용되어 불량을 낮추는 연구들이 활발히 진행되고 있다. 이에 본 연구에서는 머신러닝을 통한 불량예측을 시트 폼 발포공정에 접목시켜 발포공정의 효율성을 극대화하는 연구를 진행하였다. 자동차 시트폼 에서 주로 사용되는 폴리우레탄 폼(polyurethane foam)은 폴리올(polyol, 이하 POL)과 이소시아네이트(isocyanate, 이하 ISO)를 혼합 및 발포하는 공정으로 제조되며, 각 원료의 혼합비율과 온도의 변화에 따라 제품의 특성이 변화한다. 이에 본 연구에서는 발포공정에서 수집되는 인자별 데이터값을 머신러닝에 적용하여 불량을 예측하고자 한다. 머신러닝에 사용되는 알고리즘으로는 의사결정트리, kNN, 앙상블 알고리즘을 사용하였으며 학습은 5,147개의 데이터를 사용하였으며, 학습된 결과를 1,000개의 검증용 데이터에 적용한 결과, 세 알고리즘 중 앙상블 알고리즘에서 최대 98.5 %의 정확도를 확인할 수 있었다. 이러한 결과를 통해 발포공정에서 실시간으로 수집되는 데이터를 통해 현재 생산되는 부품의 불량 여부를 확인할 수 있으며, 나아가 각 인자를 조절하여 불량률을 개선할 수 있음을 짐작할 수 있다고 사료된다.

Keywords

Acknowledgement

본 논문은 2020년도 중소벤처기업부의 상반기 중소기업 연구인력(고경력) 지원사업으로 수행되었음.

References

  1. S. H. Woo, Y. B. Cho, "Major Technologies and Introduction of Smart Factory." 2018.
  2. C. S. Seo, S. J. Jeong, S. C. Kim, "Establishing a Smart Factory to Improve Enterprise Productivity", The Journal of The Korean Institute of Communication Sciences, Vol35. No.6 43-49, 2018.
  3. S. Y. Kim, H. J. Kim, H. S. Ji, K. S. Kim, O. W. Kim, Y. D Jun, "Characteristics of Static Comfort with Changing Isocyanate/Polyol Mixing Ratio of Polyurethane Foam", The Korean Society Of Automotive Engineers, 954-954, 2019.
  4. H. S. Kim, J. W. Youn, "A study on foaming characteristics of polyurethane depending on environmental temperature and blowing agent content." Transactions of Materials Processing Vol18. No.3 256-261, 2009. DOI:https://doi.org/10.5228/KSPP.2009.18.3.256
  5. Bikard, J., Bruchon, J., Coupez, T., & Silva, L. Numerical simulation of 3D polyurethane expansion during manufacturing process. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 309(1-3), 49-63, 2007. DOI:https://doi.org/10.1016/j.colsurfa.2007.04.025
  6. H. R. Yoon, "A Empirical Study on the Financial Stability Prediction Model of South Korea's Public Enterprises with Machine Learning Techniques", Ph. D, Hansung University, 19-22.
  7. M. M. C. Han, Y. S. Kim, C. K. Lee, "A Study on Defect Prediction Method Using Sensor Data and Machine Learning in Manufacturing Process", Entrue Journal of Information Technology, Vol17. No.1, 89-98, 2019.
  8. J. E. Ahn, J. Y. Jung, "Predicting and Interpreting Quality of CMP Process for Semiconductor Wafers Using Machine Learning." The Journal of Bigdata Vol4. No.2, 61-71, 2019. https://doi.org/10.36498/kbigdt.2019.4.2.61
  9. J. H. Choi, D. S. Seo. "Decision Trees and Its Applications", Journal of statistical analysis., Vol.4 No.1 61-83, 1999.
  10. .Navada, A., Ansari, A. N., Patil, S., & Sonkamble, B. A. "Overview of use of decision tree algorithms in machine learning." 2011 IEEE control and system graduate research colloquium. IEEE, 37-42, 2011. DOI:https://doi.org/10.1109/ICSGRC.2011.5991826
  11. Che, D., Liu, Q., Rasheed, K., & Tao, X, "Decision tree and ensemble learning algorithms with their applications in bioinformatics". Software tools and algorithms for biological systems. Springer, New York, NY, 191-199. 2011 DOI:https://doi.org/10.1007/978-1-4419-7046-6_19
  12. Y. S. Jang, B. J. Park, C. Y. Park, "Comparison study of K-nearest neighborhood classification algorithms", Journal of the Korean Data And Information Science Society, Vol30. No.5, 977-985, 2019. DOI:https://doi.org/10.7465/jkdi.2019.30.5.977
  13. Bzdok, Danilo, Martin Krzywinski, Naomi Altman, "Machine learning: supervised methods", Nat Methods 15, 5-6, 2018. DOI:https://doi.org/10.1038/nmeth.4551
  14. S. H. Min, "Improving an Ensemble Model by Optimizing Bootstrap Sampling", Journal of Internet Computing and Services, Vol17. No.2, 49-57, 2016. DOI:https://doi.org/10.7472/jksii.2016.17.2.49
  15. Vega-Pons, Sandro, Jose Ruiz-Shulcloper., Faramarz Gordaninejad, Xiaojie Wang, "A survey of clustering ensemble algorithms", International Journal of Pattern Recognition and Artificial Intelligence, Vol25. No.3 337-372, 2011. DOI:https://doi.org/10.1142/S0218001411008683
  16. Polikar, Robi. "Ensemble learning." Ensemble machine learning. Springer, Boston, MA, 1-34, 2012. DOI:https://doi.org/10.1007/978-1-4419-9326-7_1