Predictive Quality: Towards a New Understanding of Quality Assurance Using Machine Learning Tools
Product failures are dreaded by manufacturers for the associated costs and resulting damage to their public image. While most defects can be traced back to decisions early in the design process they are often not discovered until much later during quality checks or, at worst, by the customer. We propose a machine learning-based system that automatically feeds back insights about failure rates from the quality assurance and return processes into the design process, without the need for any manual data analysis. As we show in a case study, this system helps to assure product quality in a preventive way.
KeywordsData analytics Machine learning Neural networks Quality assurance Preventive quality Predictive quality
This work is based on Preventive Quality Assurance, a project partly funded by the German ministry of education and research (BMBF), reference number 01S17012D. The authors are responsible for the publication’s content.
- 1.FMEA - Fehlermöglichkeits- und Einflussanalyse. Whitepaper, Deutsche Gesellschaft für Qualität (2012)Google Scholar
- 2.Morlock, S.: Qualitätsgesicherte Produktentwicklung in der Bekleidungsindustrie (2013). https://www.hohenstein.de/de/inline/pressrelease_45248.xhtml?excludeId=45248
- 4.Veronesi, L., Kifonti, R., Cattaneo, G.M.: Data-driven innovation in the manufacturing industry. Eur. Data Market Study (2014)Google Scholar
- 5.Stamatis, D.H.: Failure Mode and Effect Analysis: FMEA from Theory to Execution. ASQ Quality Press, Milwaukee (2003)Google Scholar
- 6.Ericson, C.A.: Fault tree analysis. In: Hazard Analysis Techniques for System Safety, pp. 183–221 (2005)Google Scholar
- 7.Oakland, J.S.: Statistical Process Control. Routledge, New York (2007)Google Scholar
- 8.Schulte, R., Pape, D.: Advanced digital mock-up based on the usage of assistances. Des. Principles Pract. 5, 195–206 (2011)Google Scholar
- 11.Quality management systems - fundamentals and vocabulary. Standard, International Organization for Standardization (2015)Google Scholar
- 14.scikit-learn. http://scikit-learn.org
- 15.Python data analysis library (pandas). https://pandas.pydata.org
- 16.Dal Pozzolo, A., Caelen, O., Johnson, R.A., Bontempi, G.: Calibrating probability with undersampling for unbalanced classification. In: 2015 IEEE Symposium Series on Computational Intelligence, pp. 159–166 (2015)Google Scholar
- 17.Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)