Quality Quandaries: How to Set Up a Robust Shewhart Control Chart for Dispersion?

Self Cite

In real world scenarios, we continuously inspect the performing behavior of different features of processes. A change in the process parameters may affect the product quality and therefore needs a timely identification. For this purpose we have designed new memory structure based on Tukey control chart (TCC) under cumulative sum (CUSUM) setup. The proposed chart, namely Tukey‐CUSUM, combines the feature of TCC with CUSUM design for improved detection of shifts for normal and non‐normal data. The performance of the study proposal is evaluated through different measures based on run length including ARL, EQL, PCI and RARL. We have observed that the proposed Tukey‐CUSUM design is quite robust and efficient, especially using asymmetrical decision intervals for different process models. An application example is also provided to illustrate the study proposal. Moreover, sample selection is always of great concern for practitioners so we have included a brief description of the steps for sample selection with reference to process monitoring. Copyright © 2015 John Wiley & Sons, Ltd.

Section: The Proposed Turkey-cusum Control Chartmentioning

confidence: 93%

Robust Tukey–CUSUM Control Chart for Process Monitoring

Khaliq

Riaz

2015

Self Cite

“…Unequal subgroup sizes in financial data are common due to market closures, public holidays etc. Hence the two robust parameter estimation procedures suggested in Nazir et al and Nazir et al were adopted for unequal subgroups using the appropriate unbiasing constants. This modified stepwise procedure of robust estimation of the mean and standard deviation is summarized in Appendix .…”

Section: A Finance Applicationmentioning

confidence: 99%

“…Phase I analysis of the differences involved only a single iteration for the elimination of special cause subgroups. S d was calculated from the Phase I data following the general estimation procedure proposed by Nazir et al 27 ; see Schoonhoven et al 28 for an alternative source. We adopted the approach of Nazir et al 29 for estimating the location parameter needed for Phase II analysis.…”

Section: A Finance Applicationmentioning

confidence: 99%

Control charts for paired differences: and S_d charts

Premarathna

Godfrey

Govindaraju

2017

Control chart procedures for monitoring paired variables are sparse in the literature. After considering the average run length properties of the trued¯ chart, which monitors the mean of the differences between paired variables, we propose a new chart based on Sd, the subgroup standard deviations of the differences. Our findings show that the trued¯ chart is powerful for monitoring the changes in the means and the Sd chart is suitable for monitoring changes in the covariance structure. Furthermore, we show that the trued¯ and Sd charts perform better than existing bivariate control charts for detecting shifts in mean and variance/covariance, respectively, when standards are known. The difference charts also performed well compared to common alternatives when the standards are unknown arising from a limited amount of Phase I data. An application of these difference charts in a finance context is illustrated using the returns of Apple Inc's stock and the S&P 500 index.

“…26 Does and co-workers improved the performance of S-charts via reweighting robust estimators with respect to both subgroups and individual observations. [27][28][29][30] Although much work has been done on the application of reweighted robust estimators in SPM, a number of statistical and practical issues, which we believe need further elaboration, is discussed in the current study. Effect of trimming level used in reweighting schemes on the performance of S-Charts, to our knowledge, has not been studied in the literature, so we employed trimming levels between 0.5% and 4% to study specificity and sensitivity of the resulting control charts.…”

Section: Introductionmentioning

confidence: 99%

“…Does and co‐workers improved the performance of S‐charts via reweighting robust estimators with respect to both subgroups and individual observations. ()…”

Section: Introductionmentioning

confidence: 99%

Revisiting reweighted robust standard deviation estimators for univariate Shewhart S‐charts

Mutlu

Alakent

2018

Phase I outliers, unless screened during process parameter estimation, are known to deteriorate Phase II performance of process control charts. Reweighting estimators, ie, trimming outlier subgroups and individual observations, were suggested in the literature to improve both the robustness and efficiency of the resulting parameter estimates. In the current study, effects of various reweighted estimators at different trimming levels on the Phase II performance of S‐charts are elucidated using computer simulations including isolated and mixtures of contamination models. Outlier magnitudes in the simulations are held at a moderately low level to mimic industrial practice. Subtleties, such as varying Type I error rate among different trimming levels with respect to quantiles of dispersion estimates, prevent a single method to be revealed as the best performing one under all circumstances, and choice of estimators and trimming levels should depend on the number of subgroups in Phase I and the specifics of the process. Nevertheless, S‐chart using scale M‐estimator with logistic ρ and location M‐estimator at 2% trimming generally stands out in terms of Phase II performance, and high trimming levels are particularly recommended for high number of Phase I subgroups.