Multivariate Process Capability, Process Validation and Risk Analytics Based on Product Characteristic Sets: Case Study Piston Rod

  • Stefan Bracke
  • Bianca BackesEmail author
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 637)


Manufacturing processes of technically complex products require highly standardised methods to fulfil technical and customer specifications. To accomplish the demanded specifications, various methods, which can be applied at different phases of the product life cycle, have been developed. One of these methods, within the manufacturing phase, is the process capability index (PCI). The determination of the PCI allows the visualisation of risk with one indicator and failure probability with regard to a manufacturing process. State-of-the-art is the univariate calculation of the PCI based on the analysis of one product characteristic. This paper outlines different approaches for the determination of multidimensional process capability indices (MPCI) based on a product characteristic set including symmetric and asymmetric product characteristic distribution models. The goal of the explained methods is the analysis of risks, the determination of risk indicator MPCI and failure probabilities with regard to complex manufacturing processes.


Multivariate manufacturing analysis Failure probabilities Process capability index Statistical process control Crankshaft Piston rod Cp/Cpk 


  1. 1.
    DIN ISO 3534-2:2013-12: Statistics - Vocabulary and Symbols - Part 2: Applied Statistics (ISO 3534-2:2006)Google Scholar
  2. 2.
    Goethals, P.L., Cho Byung, R.: The development of a target-focused process capability index with multiple characteristics. Qual. Reliab. Eng. Int. 27(3), 297–311 (2011). doi: 10.1002/qre.1122 CrossRefGoogle Scholar
  3. 3.
    Tano, I., Vännman, K.: Comparing confidence intervals for multivariate process capability indices. Qual. Reliab. Eng. Int. 28, 481–495 (2011). doi: 10.1002/qre.1250 CrossRefGoogle Scholar
  4. 4.
    Bracke, S.: Prozessfähigkeit bei der Herstellung komplexer technischer Produkte. Statistische Mess- und Prüfanalyse. Springer, Berlin (2016). (in German language)CrossRefGoogle Scholar
  5. 5.
    Sachs, L.: Applied Statistics. Springer, Berlin (2002)zbMATHGoogle Scholar
  6. 6.
    Dietrich, E., Schulze, A.: Statistical Procedures for Machine and Process Qualification. Munich Hanser Publications, Cincinnati (2010)Google Scholar
  7. 7.
    Pfeifer, T.: Quality Management. Hanser, München (2002)CrossRefGoogle Scholar
  8. 8.
    Härdle, W., Simar, L.: Applied Multivariate Statistical Analysis. Springer, Heidelberg (2015)CrossRefzbMATHGoogle Scholar
  9. 9.
    Kotz, S., Balakrishnan, N., Johnson, N.: Continuous Multivariate Distributions. Wiley, Hoboken (2005)zbMATHGoogle Scholar
  10. 10.
    Rinne, H.: The Weibull Distribution. CRC Press, Boca Raton (2009)zbMATHGoogle Scholar
  11. 11.
    Genz, A., Bretz, F.: Computation of Multivariate Normal and Probabilities. Springer, Berlin (2009)CrossRefzbMATHGoogle Scholar
  12. 12.
    Rizzo, M.L.: Statistical Computing with R. Computer Science and Data Analysis Series. Chapman and Hall/CRC, Boca Raton (2008)Google Scholar
  13. 13.
    Hamerle, A., Fahrmeir, L.: Multivariate statistische Verfahren. Walter de Gruyter, Berlin (1984)zbMATHGoogle Scholar
  14. 14.
    Bracke, S., Backes, B.: Multidimensional failure probabilities based on symmetric and asymmetric product characteristic distribution models within high precision manufacturing processes. In: Proceedings CIE45, 28th–30th October 2015, Metz, France (2015)Google Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  1. 1.Chair of Reliability Engineering and Risk AnalyticsUniversity of WuppertalWuppertalGermany

Personalised recommendations