85
views
0
recommends
+1 Recommend
1 collections
    0
    shares
       
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      Performance Prediction Of Production Lines Using Machine Learning Algorithm

      Preprint
      In review
      research-article
      Bookmark

            Abstract

            Manufacturing today considers data-drive business operations at different levels leading to the growth of various paradigms in manufacturing, of which emerged smart manufacturing.

            However data can be used to predict equipment failure rates, streamline and optimize inventory management and prioritize processes.

            The use of parameter tuning and optimization, grid-search, cross-validation, to predict the best performing machine learning algorithm. This research work evaluates the time potential failure-rates, against the lines which peaks and drops depending on its components RUL(Remaining Useful Life). The accuracy of the machine learning algorithms that are employed in this studies, are hence subjected to some metrics for evaluation, these are : MCC and AUC-ROC.

            This study has analyzed and evaluated some annoymized dataset from a manufacturing company, using some metrics and machine learning algorithms for performance prediction of their production lines using unsupervised learning. This study would served as a good reference for anyone wanting to use the best performance model, for further research work.

            Content

            Author and article information

            Journal
            ScienceOpen Preprints
            ScienceOpen
            3 November 2021
            Affiliations
            [1 ] Dundalk Institute of Technology
            Author notes
            Article
            10.14293/S2199-1006.1.SOR-.PPA7BE8.v1
            a66ca8d1-5399-4fda-a995-fdd0cf5f0188

            This work has been published open access under Creative Commons Attribution License CC BY 4.0 , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Conditions, terms of use and publishing policy can be found at www.scienceopen.com .


            The datasets generated during and/or analysed during the current study are available in the repository: https://www.kaggle.com/ c/bosch-production-line-performance.
            Computer science,Statistics,Engineering
            Machine learning,Algorithms,MCC(Mattews correlation coefficient),AUC-ROC(Area under curve the receiver operating characteristics)

            Comments

            Comment on this article