Statistical process control using run sums

Thomas R. Willemain, George Runger

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

We develop a novel and effective way to monitor quality using the large volumes of positively autocorrelated data produced by high-frequency sampling of a process. We regard the process as a sequence of runs above and below the mean. The sums of the observations in these runs behave as independent random variables suitable for charting. Using simulated data, we show that the average run length performance of charts based on run sums compares favorably to that of alternative charts based on ARMA residuals, while avoiding the need for ARMA modeling. Furthermore, we obtained the same relative performance results for iid data. Thus, run sum charts provide a powerful and comprehensive method for SPC in data-rich environments.

Original languageEnglish (US)
Pages (from-to)361-378
Number of pages18
JournalJournal of Statistical Computation and Simulation
Volume61
Issue number4
StatePublished - 1998

Fingerprint

Statistical Process Control
Statistical process control
Random variables
Chart
Sampling
Autoregressive Moving Average
Average Run Length
Independent Random Variables
Monitor
Charts
Alternatives
Modeling
Autoregressive moving average

Keywords

  • ARMA models
  • Autocorrelation
  • Level crossing
  • Monte Carlo methods
  • Run sums
  • Shewhart chart
  • Statistical process control

ASJC Scopus subject areas

  • Applied Mathematics
  • Modeling and Simulation
  • Statistics and Probability

Cite this

Statistical process control using run sums. / Willemain, Thomas R.; Runger, George.

In: Journal of Statistical Computation and Simulation, Vol. 61, No. 4, 1998, p. 361-378.

Research output: Contribution to journalArticle

@article{5a6bd557f5ce4c1791e4c8c6752a4081,
title = "Statistical process control using run sums",
abstract = "We develop a novel and effective way to monitor quality using the large volumes of positively autocorrelated data produced by high-frequency sampling of a process. We regard the process as a sequence of runs above and below the mean. The sums of the observations in these runs behave as independent random variables suitable for charting. Using simulated data, we show that the average run length performance of charts based on run sums compares favorably to that of alternative charts based on ARMA residuals, while avoiding the need for ARMA modeling. Furthermore, we obtained the same relative performance results for iid data. Thus, run sum charts provide a powerful and comprehensive method for SPC in data-rich environments.",
keywords = "ARMA models, Autocorrelation, Level crossing, Monte Carlo methods, Run sums, Shewhart chart, Statistical process control",
author = "Willemain, {Thomas R.} and George Runger",
year = "1998",
language = "English (US)",
volume = "61",
pages = "361--378",
journal = "Journal of Statistical Computation and Simulation",
issn = "0094-9655",
publisher = "Taylor and Francis Ltd.",
number = "4",

}

TY - JOUR

T1 - Statistical process control using run sums

AU - Willemain, Thomas R.

AU - Runger, George

PY - 1998

Y1 - 1998

N2 - We develop a novel and effective way to monitor quality using the large volumes of positively autocorrelated data produced by high-frequency sampling of a process. We regard the process as a sequence of runs above and below the mean. The sums of the observations in these runs behave as independent random variables suitable for charting. Using simulated data, we show that the average run length performance of charts based on run sums compares favorably to that of alternative charts based on ARMA residuals, while avoiding the need for ARMA modeling. Furthermore, we obtained the same relative performance results for iid data. Thus, run sum charts provide a powerful and comprehensive method for SPC in data-rich environments.

AB - We develop a novel and effective way to monitor quality using the large volumes of positively autocorrelated data produced by high-frequency sampling of a process. We regard the process as a sequence of runs above and below the mean. The sums of the observations in these runs behave as independent random variables suitable for charting. Using simulated data, we show that the average run length performance of charts based on run sums compares favorably to that of alternative charts based on ARMA residuals, while avoiding the need for ARMA modeling. Furthermore, we obtained the same relative performance results for iid data. Thus, run sum charts provide a powerful and comprehensive method for SPC in data-rich environments.

KW - ARMA models

KW - Autocorrelation

KW - Level crossing

KW - Monte Carlo methods

KW - Run sums

KW - Shewhart chart

KW - Statistical process control

UR - http://www.scopus.com/inward/record.url?scp=0032280689&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032280689&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0032280689

VL - 61

SP - 361

EP - 378

JO - Journal of Statistical Computation and Simulation

JF - Journal of Statistical Computation and Simulation

SN - 0094-9655

IS - 4

ER -