DOI QR코드

DOI QR Code

Machine Learning based on Approach for Classification of Abnormal Data in Shop-floor

제조 현장의 비정상 데이터 분류를 위한 기계학습 기반 접근 방안 연구

  • Shin, Hyun-Juni (Department of Electrical, Electronics & Communication Engineering, Korea University of Technology and Education(KOREATECH)) ;
  • Oh, Chang-Heon (Department of Electrical, Electronics & Communication Engineering, Korea University of Technology and Education(KOREATECH))
  • Received : 2017.10.25
  • Accepted : 2017.11.03
  • Published : 2017.11.30

Abstract

The manufacturing facility is generally operated by a pre-set program under the existing factory automation system. On the other hand, the manufacturing facility must decide how to operate autonomously in Industry 4.0. Determining the operation mode of the production facility itself means, for example, that it detects the abnormality such as the deterioration of the facility at the shop-floor, prediction of the occurrence of the problem, detection of the defect of the product, In this paper, we propose a manufacturing process modeling using a queue for detection of manufacturing process abnormalities at the shop-floor, and detect abnormalities in the modeling using SVM, one of the machine learning techniques. The queue was used for M / D / 1 and the conveyor belt manufacturing system was modeled based on ${\mu}$, ${\lambda}$, and ${\rho}$. SVM was used to detect anomalous signs through changes in ${\rho}$.

스마트 공장은 미리 입력된 프로그램에 의해 생산시설이 수동적으로 움직이는 공장 자동화 작업 방식과는 달리, 생산 설비 스스로 작업 방식을 결정하여야 한다. 생산 설비 스스로 작업 방식을 결정이라 함은, 이를테면 제조 현장에서 설비의 노후, 문제 발생 예측, 제품의 불량 검출 등과 같은 이상 징후가 발생할 시 이를 조기에 발견한 후 스스로 문제를 해결하는 것을 의미한다. 본 논문에서는 제조 현장의 제조 공정 이상 징후 감지를 위해 대기행렬을 이용한 제조 공정 모델링을 제시하고 해당 모델링에서 이상 징후를 기계학습 기술 중 하나인 SVM을 이용하여 이를 감지하도록 한다. 해당 대기행렬을 M/D/1을 사용하였으며, ${\mu}$, ${\lambda}$, ${\rho}$를 기반으로 컨베이어 벨트 제조 시스템을 모델링하였다. SVM을 이용하여 ${\rho}$의 변화량을 통해 이상 징후를 감지했다.

Keywords

References

  1. M. Taisch, B. Stahl and G. Tavola, "ICT in Manufacturing: Trends and Challenges for 2020 - an European View," Industrial Informatics (INDIN), In Proceedings of the IEEE International Conference on, Houston, USA, pp. 941-946, Sep. 2012.
  2. J. M. Park, "Technology and Issue on Embodiment of Smart Factory in Small-Medium Manufacturing Business," The Journal of The Korean Institute of Communication Science, vol.40, no. 12, pp. 2491-2502, Dec. 2015.
  3. J. B. Kim, "An Empirical Study on the Success Cases of Digital Manufacturing System Implementation in Manufacturing Industry-Focused on Smart Factory," Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology, vol.5, no.4, pp. 1-8, Aug. 2015. https://doi.org/10.14257/AJMAHS.2015.06.05
  4. K. W. Cho, M. H. Jeon and C. H. Oh, "Development of Equipment Control System based on DB Access Method for Industrial IoT," Journal of the Korea Institute of Information and Communication Engineering, vol. 20, no. 6, pp. 1142-1147, Jun. 2016. https://doi.org/10.6109/jkiice.2016.20.6.1142
  5. C. Cortes and V. Vapnik, "Support-vector Networks," Machine Learning, vol. 20, no. 3, pp. 273-297, Sep. 1995. https://doi.org/10.1007/BF00994018
  6. M. D. Odom and R. Sharda, "A Neural Network Model for Bankruptcy Prediction," In Proceedings of the International Joint Conference on Neural networks, San Diego, USA, pp. 163-168, Jun. 1990.
  7. Y. G. Oh, H. S. Park, A. R. Yoo, N. H. Kim, Y. H Kim, D. C. Kim, J. U. Choi, S. H. Yoon and H. J. Yang, "A Product Quality Prediction Model Using Real-Time Process Monitoring in Manufacturing Supply Chain," Journal of the Korean Institute of Industrial Engineers, vol. 39, no. 4, pp. 271-277, Aug. 2013. https://doi.org/10.7232/JKIIE.2013.39.4.271
  8. B. U. Narayan, An Introduction to Queueing Theory Modeling and Analysis in Application, 2nd ed. Birkhauser, United States, 2008.
  9. A. A. Alsebae, M. S. Leeson and R. J. Green, "SimEventsbased Modeling and Simulation Study of Stop-and-Wait Protocol," In Proceedings of the International Conference on Modelling, Identification and Control (ICMIC), Cairo, Egypt, pp. 239-244, Aug-Sep 2013.