1. Proceedings
  2. I3M
  3. 2022
  4. EMSS
  5. 10.46354/i3m.2022.emss.007

Application of Deep Learning based on a Simulation Model to classify Production Orders

  • a Lukas Rissmann,
  • b Konstantin Muehlbauer,
  • c Sebastian Meissner 
  • a,b,c   Technology Centre for Production and Logistics Systems, Landshut University of Applied Sciences, Am Lurzenhof 1,Landshut, Germany
Cite as
Rissmann L., Muehlbauer K., and Meissner S. (2022).,Application of Deep Learning based on a Simulation Model to classify Production Orders. Proceedings of the 34th European Modeling & Simulation Symposium (EMSS 2022). , 007 . DOI: https://doi.org/10.46354/i3m.2022.emss.007

Abstract

The application of artificial intelligence can support employees indecision-making on highly complex issues. The improving performance of computers in combination with the progress in deep learning makes it possible to answer problems with a very high complexity. The approach presented in this paper demonstrates that production orders of an assembly line can be classified with regard to a chosen key performance indicator using deep learning as a surrogate model for a logistics simulation. Evaluating production orders in advance enables a higher performance of a production system without cost intensive process improvements. The aim of the approach, demonstrated on an exemplary use case, is to utilize deep learning to determine which sequence of individual production orders leads to a high throughput in units. Results gained reveal a significant increase of average throughput and therefore showing the feasibility of the approach. The application of artificial intelligence models enables that such complex questions can be solved in a short time. Consequently, the model is able to classify production orders with an accuracy of86%.

References

  1. Bárkányi Á., Chován T., Németh S., and Abonyi J. (2021). Modellingfor Digital Twins—Potential Role of Surrogate Models.Processes9 3, 476.
  2. Blöchl S. J. and Schneider M. (2016).Simulation Game forIntelligent Production Logistics–The PuLL® Learning Factory.Procedia CIRP54, 130–135.
  3. Boysen N., Fliedner M., and Scholl A. (2009). Production planningof mixed-model assembly lines: overview and extensions.Production Planning & Control20 5, 455–471.
  4. Cho S., May G., Tourkogiorgis I., Perez R., Lazaro O., La Maza B. de,and Kiritsis D. (2018). A Hybrid Machine Learning Approach forPredictive Maintenance in Smart Factories of the Future: Advances in Production Management Systems. Smart Manufacturing for Industry 4.0. Cham, Springer InternationalPublishing, 536, 311–317.
  5. Domingos P. (2012). A few useful things to know about machinelearning.Commun. ACM55 10, 78–87.
  6. Dylewski R., Jardzioch A., and Krebs I. (2016). The OptimalSequence of Production Orders, Taking into Account the Costof Delays.Management and Production Engineering Review72, 21–28.
  7. Elbasheer M., Longo F., Nicoletti L., Padovano A., Solina V., andVetrano M. (2022). Applications of ML/AI for Decision-Intensive Tasks in Production Planning and Control.ProcediaComputer Science200, 1903–1912.
  8. Goodfellow, I., Bengio, Y., and Courville, A. (2016). Deep learning.Cambridge, Massachusetts, London, England: The MIT Press(Adaptive computation and machine learning).
  9. Knoll D., Neumeier D., Prüglmeier M., and Reinhart G. (2019).Anautomated packaging planning approach using machinelearning.Procedia CIRP81, 576–581.
  10. Knoll D., Prüglmeier M., and Reinhart G. (2016). PredictingFutureInbound Logistics Processes Using Machine Learning.ProcediaCIRP52, 145–150.
  11. LeCun Y., Bengio Y., and Hinton G. (2015). Deep learning.Nature7553, 436–444.
  12. Li B., Hou B., Yu W., Lu X., and Yang C. (2017). Applications ofartificial intelligencein intelligent manufacturing: a review.Frontiers Inf Technol Electronic Eng18 1, 86–96.
  13. Meissner S. (2010). Controlling just-in-sequence flow-production.Logistics Research2 1, 45–53.
  14. Rabe, M., Spieckermann, S., and Wenzel, S. (2008). Verifikation undValidierung für die Simulation in Produktion und Logistik.Vorgehensmodelle und Techniken. 1. Aufl. Berlin, Heidelberg:Springer. Online verfügbar unterhttps://ebookcentral.proquest.com/lib/kxp/detail.action?docID=364359.
  15. Sun J. and Xue D. (2001). A dynamic reactive scheduling mechanismfor responding to changes of production orders andmanufacturing resources.Computers in Industry46 2, 189–207.
  16. Timm I. J. and Lattner A. D. (2010).Künstliche Intelligenz in derLogistik.Künstl Intell24 2, 99–103.
  17. Ungermann F., Kuhnle A., Stricker N., and Lanza G. (2019).DataAnalytics for Manufacturing Systems–A Data-DrivenApproach for Process Optimization.Procedia CIRP81, 369–374.
  18. Usuga Cadavid J. P., Lamouri S., Grabot B., Pellerin R., and Fortin A.(2020). Machine learning applied in production planning andcontrol: a state-of-the-art in the era of industry 4.0.J IntellManuf31 6, 1531–1558.
  19. Uttendorf S., Eilert B., and Overmeyer L. (2016). A fuzzy logic expertsystem for the automated generation of roadmaps forautomated guided vehicle systems. In:2016 IEEE International Conference on Industrial Engineering and EngineeringManagement (IEEM).Bali, Indonesia, 04.12.2016-07.12.2016.IEEE, 977–981.
  20. Vojdani N. and Erichsen B. (2019). Anwendung von Support VectorRegression zur vorausschauenden Identifikation vonStörungswirkungen in der Produktionslogistik.LogisticsJournal: Proceedings 2019.
  21. Wang K. and Wang Y. (2018). How AI Affects the Future PredictiveMaintenance: A Primer of Deep Learning:AdvancedManufacturing and Automation VII. Singapore, SpringerSingapore, 451, 1–9.
  22. Weichert D., Link P., Stoll A., Rüping S., Ihlenfeldt S., and Wrobel S.(2019). A review of machine learning for the optimization ofproduction processes.Int J Adv Manuf Technol104 5-8,1889–1902.
  23. Woschank M., Rauch E., and Zsifkovits H. (2020). A Review ofFurther Directions for Artificial Intelligence, Machine Learning,and Deep Learning in Smart Logistics.Sustainability12 9, 1–23.
  24. Wuest T., Weimer D., Irgens C., and Thoben K.-D. (2016). Machinelearning in manufacturing: advantages, challenges, andapplications.Production & Manufacturing Research4 1, 23–45.
  25. Zafarzadeh M., Wiktorsson M., and Baalsrud Hauge J. (2021). ASystematic Review on Technologies for Data-DrivenProductionLogistics: Their Role from a Holistic and ValueCreation Perspective.Logistics5 2, 1–32.
  26. Zhang W. J., Yang G., Lin Y., Ji C., and Gupta M. M. (2018). OnDefinition of Deep Learning. In:2018 World AutomationCongress (WAC).Stevenson, WA, 03.06.2018-06.06.2018,IEEE, 1–5.