During the past three decades, PCIs – process capability indices – have inspired hundreds of pages of scientific research. The trade-off between simplicity and precision in reproducing an overall process quality prediction is both the reason behind the criticism for Cp and Cpk and the motive for their widespread use. Indeed, their strength in simplifying the assessment of a process control status compensates for some of the statistical shortcomings largely recognized in the literature, amongst which is the normality assumption. Hence, this article aims at overcom‐ ing the main statistical problems of Cp and Cpk indices by proposing a new indicator, still compliant with the tradi‐ tional PCIs approach, but applicable for cases of nonnormal processes while being simple to implement and easy to interpret. The proposed indicator is composed of three sub-indices, each related to a specific process charac‐ teristic: how the process is repeatable, how much the data distribution is skewed about the mean value and how much the process data comfortably lies between the specifications limits. On top of this, a specific parameter allows designers or quality engineers to modify the index value range in order to fine-tune the effect of the cost of Taguchi’s loss function. The article presents the theoretical structure of the new indicator and an extensive numerical test on several different processes with different distributions upon multiple specification limit combinations, along with a comparison to the Cpk index, in order to demonstrate how the new index provides a clearer indication of the process criticalities

Baciarello, L., Schiraldi, M.m. (2015). A Proposal for a Management-oriented Process Capability Index. INTERNATIONAL JOURNAL OF ENGINEERING BUSINESS MANAGEMENT, 7(26), 1-13 [10.5772/62185].

A Proposal for a Management-oriented Process Capability Index

SCHIRALDI, MASSIMILIANO MARIA
2015

Abstract

During the past three decades, PCIs – process capability indices – have inspired hundreds of pages of scientific research. The trade-off between simplicity and precision in reproducing an overall process quality prediction is both the reason behind the criticism for Cp and Cpk and the motive for their widespread use. Indeed, their strength in simplifying the assessment of a process control status compensates for some of the statistical shortcomings largely recognized in the literature, amongst which is the normality assumption. Hence, this article aims at overcom‐ ing the main statistical problems of Cp and Cpk indices by proposing a new indicator, still compliant with the tradi‐ tional PCIs approach, but applicable for cases of nonnormal processes while being simple to implement and easy to interpret. The proposed indicator is composed of three sub-indices, each related to a specific process charac‐ teristic: how the process is repeatable, how much the data distribution is skewed about the mean value and how much the process data comfortably lies between the specifications limits. On top of this, a specific parameter allows designers or quality engineers to modify the index value range in order to fine-tune the effect of the cost of Taguchi’s loss function. The article presents the theoretical structure of the new indicator and an extensive numerical test on several different processes with different distributions upon multiple specification limit combinations, along with a comparison to the Cpk index, in order to demonstrate how the new index provides a clearer indication of the process criticalities
Pubblicato
Rilevanza internazionale
Articolo
Esperti anonimi
Settore ING-IND/17 - Impianti Industriali Meccanici
English
Con Impact Factor ISI
Process Capability; Process Performance; PCI; Statistical Process Control; Cpk
http://cdn.intechopen.com/pdfs-wm/49788.pdf
Baciarello, L., Schiraldi, M.m. (2015). A Proposal for a Management-oriented Process Capability Index. INTERNATIONAL JOURNAL OF ENGINEERING BUSINESS MANAGEMENT, 7(26), 1-13 [10.5772/62185].
Baciarello, L; Schiraldi, Mm
Articolo su rivista
File in questo prodotto:
File Dimensione Formato  
baciarello_cpk.pdf

accesso aperto

Descrizione: full paper
Licenza: Creative commons
Dimensione 3.69 MB
Formato Adobe PDF
3.69 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2108/156547
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 1
social impact