### Abstract

A significant practical disadvantage of multivariate statistical process control is that it is difficult to determine which of the monitored variables is responsible for the out-of-control signal. While there have been several proposed techniques and diagnostics, most of these have disadvantages. Rather than an ad hoc selection, we use three approaches to develop a diagnostic for a chi-square chart and generate the same statistic from each approach. Consequently, intuitive interpretations of the diagnostic are provided and previous proposals are simplified by a focus on a single metric.

Original language | English (US) |
---|---|

Pages (from-to) | 2203-2213 |

Number of pages | 11 |

Journal | Communications in Statistics - Theory and Methods |

Volume | 25 |

Issue number | 10 |

State | Published - 1996 |

### Fingerprint

### Keywords

- Assignable cause
- Chi-square chart
- T chart
- X chart

### ASJC Scopus subject areas

- Safety, Risk, Reliability and Quality
- Statistics and Probability

### Cite this

**Contributors to a multivariate statistical process control chart signal.** / Montgomery, Douglas.

Research output: Contribution to journal › Article

*Communications in Statistics - Theory and Methods*, vol. 25, no. 10, pp. 2203-2213.

}

TY - JOUR

T1 - Contributors to a multivariate statistical process control chart signal

AU - Montgomery, Douglas

PY - 1996

Y1 - 1996

N2 - A significant practical disadvantage of multivariate statistical process control is that it is difficult to determine which of the monitored variables is responsible for the out-of-control signal. While there have been several proposed techniques and diagnostics, most of these have disadvantages. Rather than an ad hoc selection, we use three approaches to develop a diagnostic for a chi-square chart and generate the same statistic from each approach. Consequently, intuitive interpretations of the diagnostic are provided and previous proposals are simplified by a focus on a single metric.

AB - A significant practical disadvantage of multivariate statistical process control is that it is difficult to determine which of the monitored variables is responsible for the out-of-control signal. While there have been several proposed techniques and diagnostics, most of these have disadvantages. Rather than an ad hoc selection, we use three approaches to develop a diagnostic for a chi-square chart and generate the same statistic from each approach. Consequently, intuitive interpretations of the diagnostic are provided and previous proposals are simplified by a focus on a single metric.

KW - Assignable cause

KW - Chi-square chart

KW - T chart

KW - X chart

UR - http://www.scopus.com/inward/record.url?scp=25844527344&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=25844527344&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:25844527344

VL - 25

SP - 2203

EP - 2213

JO - Communications in Statistics - Theory and Methods

JF - Communications in Statistics - Theory and Methods

SN - 0361-0926

IS - 10

ER -