DOI QR코드

DOI QR Code

A Machine Learning-based Total Production Time Prediction Method for Customized-Manufacturing Companies

주문생산 기업을 위한 기계학습 기반 총생산시간 예측 기법

  • 박도명 (동아대학교 경영정보학과) ;
  • 최형림 (동아대학교 경영정보학과) ;
  • 박병권 (동아대학교 경영정보학과)
  • Received : 2021.01.26
  • Accepted : 2021.03.18
  • Published : 2021.03.31

Abstract

Due to the development of the fourth industrial revolution technology, efforts are being made to improve areas that humans cannot handle by utilizing artificial intelligence techniques such as machine learning. Although on-demand production companies also want to reduce corporate risks such as delays in delivery by predicting total production time for orders, they are having difficulty predicting this because the total production time is all different for each order. The Theory of Constraints (TOC) theory was developed to find the least efficient areas to increase order throughput and reduce order total cost, but failed to provide a forecast of total production time. Order production varies from order to order due to various customer needs, so the total production time of individual orders can be measured postmortem, but it is difficult to predict in advance. The total measured production time of existing orders is also different, which has limitations that cannot be used as standard time. As a result, experienced managers rely on persimmons rather than on the use of the system, while inexperienced managers use simple management indicators (e.g., 60 days total production time for raw materials, 90 days total production time for steel plates, etc.). Too fast work instructions based on imperfections or indicators cause congestion, which leads to productivity degradation, and too late leads to increased production costs or failure to meet delivery dates due to emergency processing. Failure to meet the deadline will result in compensation for delayed compensation or adversely affect business and collection sectors. In this study, to address these problems, an entity that operates an order production system seeks to find a machine learning model that estimates the total production time of new orders. It uses orders, production, and process performance for materials used for machine learning. We compared and analyzed OLS, GLM Gamma, Extra Trees, and Random Forest algorithms as the best algorithms for estimating total production time and present the results.

4차 산업혁명 기술의 발전으로 사람이 처리하지 못하는 부분을 기계학습 등 인공지능 기법을 활용하여 개선해 보려는 노력이 확대되고 있다. 주문형 생산 기업에서도 주문에 대한 총생산시간을 예측하여 납기 지연 등의 기업 리스크를 줄이고자 하나 주문마다 총생산시간이 모두 달라 이를 예측하는데, 어려움을 겪고 있다. 주문 처리량 증대, 주문 총비용 절감을 위해 효율성이 가장 낮은 영역을 찾아 그 영역을 강화하는 TOC(Theory of constraints) 이론이 개발되었으나 총생산시간 예측은 제시하지 못하였다. 주문생산은 고객의 다양한 요구로 인해 주문마다 그 특성이 모두 다르므로 개별적인 주문의 총생산시간을 사후에 측정할 수는 있으나 사전 예측을 하기는 어렵다. 기존 주문의 이미 측정된 총생산시간도 모두 달라 표준 시간으로 활용할 수 없는 한계성이 있다. 이에 따라 경험이 많은 관리자는 시스템의 이용보다는 감에 의존하고 있고, 경험이 부족한 관리자는 간단한 관리지표(예, 원재료가 파이프이면 총생산시간 60일, 철판이면 총생산시간 90일 등)를 사용하고 있다. 불완전한 감이나 지표를 기초로 하여 작업 지시를 너무 빨리하면 정체가 발생하여 생산성이 저하되고, 너무 늦게 하면 긴급 처리로 인해 생산비용이 증가하거나 납기를 지키지 못하는 경우가 발생한다. 납기를 지키지 못하면 지체상금을 배상해야 하거나 영업, 수금 등의 부문에 악영향을 미친다. 본 연구에서는 이러한 문제를 해결하기 위하여 주문생산시스템을 운영하는 기업의 신규 주문 총생산시간을 추정하는 기계학습 모델을 찾고자 한다. 기계학습에 활용된 자료는 수주, 생산, 공정 실적을 사용한다. 그리고 총생산시간의 추정에 가장 적합한 알고리즘으로 OLS, GLM Gamma, Extra Trees, Random Forest 알고리즘 등을 비교 분석하고 그 결과를 제시하고자 한다.

Keywords

References

  1. Aleksei, G., "Histogram-Based Algorithm for Building Gradient Boosting Ensembles of Piecewise Linear Decision Trees", International Conference on Analysis of Images, Social Networks and Texts, (2019), 38-50
  2. Alexey, N., K. Alois, "Gradient boosting machines, a tutorial", US National Library of Medicine National Institutes of Health, (2013)
  3. Cho, H. J., J. I. Park, "A Method for Construction Productivity index for Increasing Productivity", Korean Institute Of Industrial Engineers, (2010), 832
  4. Choi, J. H., H. J. Kwon and J. Y. Woo and P. Park, "A case study of work improvement adapting standard time and line balancing on a manufacturing process", Journal of the Korea Institute of Plat Engineering, Vol3(1998), 172
  5. Davis, F. D., "Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology", MIS Quarterly, Vol.13(1989), 319-342 https://doi.org/10.2307/249008
  6. Dwight, D. E., "A speech to the National Defense Executive Reserve Conference in Washington, DC", (1957)
  7. Goldratt, E., The Goal second revised edition., Dongyangbooks, (2002)
  8. Epstein, M. J., M. J. Roy, "Improving sustainability performance: Specifying, implementing and measuring key principles", Journal of General Management, Vol.29(2003), 15-31 https://doi.org/10.1177/030630700302900101
  9. Friedman, M., The social responsibility of business is to increase its profits, The New York Times Magazine, September 13 (1970)
  10. Harris, D., Improving Regressors using Boosting Techniques, (1997)
  11. Kim, D. W., J. Y. Lee and K. H. Cho, "Algorithms for establishing production schedule plans for custom production plants and systems for establishing production schedule plans using them", Korea Intellectual Property Office, (2012)
  12. Kim, G. H., "A Study on the Evaluation of MBO Operation and Effectiveness", The Korea Association for Policy Studies, Vol13(2004), 52-54
  13. Kim, S. M., J. K. Ahn, "Verification of ERP Standard Time Using TOC Technique and Improvement of MES Routing Point", Journal of Society of Korea Industrial and System Engineering, Vo141(2018), 30-31
  14. Kim, S. Y., H. Y. Soon and S. W. Choi, "Intelligent Digital Factory for Productivity Innovation: Remodeling of A Tractor Factory", Korea Productivity Association, Vol23(2009), 6-7
  15. Kim, Y. Y., "Manufacturing Innovation and HPC(High Performance Computing) Utilization", Korea Technology Innovation Society, Vol19 (2016), 250
  16. Lee, D. H., S. Y. Cho and H. W. Kim, "A Study on the Increase of Satisfaction and Productivity of Smart Work", Korea Society of Management Information Systems, (2013), 298-299
  17. Lee, J., H. S. Kang, Environmental management theory, Hyungseol publishing house, 2003
  18. Lee, K. J., "Establishment of Standard Time for Stranding Process of a Cable Company in Small Quantity Batch Production System", Journal of the Society of Korea Industrial and System Engineering, Vol35(2012), 100-101
  19. Lee, S. K., Y. H. Lee, "Production Scheduling employing ERP in the make-to-order manufacturing system", IE Interfaces, Vol12(1999), 427-429
  20. Maltz, E., A. K. Kohla, "Market Intelligence Dissemination Across Functional Boundaries", Journal of Marketing Research, Vol33(1996), 47-61 https://doi.org/10.2307/3152012
  21. Mishra, G. J., D, P, Sehgal, "Quantitative Structure Activity Relationship study of the AntiHepatitis Peptides employing Random Forests and Extra-trees regressors", Bioinforamtion, Vol13(2017),
  22. Navver, j., S. Slater, "The Effect of a Market Orientation on Business Profitability," Journal of Marketing, Vol54(1990), 20-35 https://doi.org/10.2307/1251757
  23. Park, H. S., Hands-On Machine Learning with Scikit-Learn & TensorFlow, Hanbit Media Inc, (2018)
  24. Park, J. M., "A Study on the Perception of Corporate Members on the Factors of Productivity Improvement", Korea Productivity Association, Vol10(1995), 178-181
  25. Park, S. H., S. W. Choi, "Establishment of Standard Times Using STD Technique in Multi-product and One-unit Production System", Korean Association of Industrial Business Administration, Vol10(1987), 28-30
  26. Perter, M., N. John, Generalized Linear Models, Second Edition, Chapman and Hall/CRC, (1989)
  27. Ryu, H. K., "OLAP and Decision Tree Analysis of Productivity Affected by Construction Duration Impact Factors", Journal of the Korea Institute of Building Construction, Vol11 (2011), 106
  28. Scikit-learn developers, User Guide, 2007-2020, https://scikit-learn.org/stable/modules/linear_model.html
  29. Scikit-learn developers, User Guide, 2007-2020, https://scikit-learn.org/stable/modules/ensemble.html
  30. Taylor., F. W, The Principles of Scientific Management, ModiBooks, (2016)
  31. Wikipedia, (2020), https://en.wikipedia.org/wiki/Time_and_motion_study
  32. Woo. M. S., J. W. Yoo. "Impact of relative investment in SFA technology on new product success through market information acquisition process and perceived intelligence quality: Moderation Effect of Market Uncertainty and Technical Turbulence", Korea Productivity Association, Vol32(2018), 174-176