An efficient nonparametric double progressive mean chart for monitoring of the process location

There is growing literature on new versions of “memory‐type” control charts, where deceptively good zero‐state average run‐length (ARL) performance is misleading. Using steady‐state run‐length analysis in combination with the conditional expected delay (CED) metric, we show that the increasingly discussed progressive mean (PM) and homogeneously weighted moving average (HWMA) control charts should not be used in practice. Previously reported performance of methods based on these two approaches is misleading, as we found that performance is good only when a process change occurs at the very start of monitoring. Traditional alternatives, such as exponentially weighted moving average (EWMA) and cumulative sum (CUSUM) charts, not only have more consistent detection behavior over a range of different change points, they can also lead to better out‐of‐control zero‐state ARL performance when properly designed.

Section: Other Pm Researchmentioning

confidence: 99%

The impracticality of homogeneously weighted moving average and progressive mean control chart approaches

Knoth

Tercero-Gómez

Khakifirooz

et al. 2021

“…) × 100% (cf. Nazir et al 47 and Abbas et al 48 ). The ACUSUM chart under the CN distribution at 10% contamination yields at least 1.6% to 49.8% more false alarms (cf.…”

Section: Robustness Against the Non-normality Of The Proposed Npacusu...mentioning

confidence: 96%

“…Hou and Yu 14 developed an NPCUSUM shift detection chart for joint monitoring of the process parameters. Liang et al., 15 Song et al., 16 Graham, 17 Malela–Majika and Rapoo, 18 Li et al., 19 Abbas et al., 20 Erem and Mahmood 21 and the references therein may also be seen for more literature on these types of charting structures.…”

Section: Introductionmentioning

confidence: 99%

Nonparametric adaptive cumulative sum charting scheme for monitoring process location

Abbas,

Nazir,

Xiang

et al. 2024

Self Cite

For effective monitoring of the range of sustainable changes in the process median, a nonparametric adaptive cumulative sum signed‐rank (NPACUSUM‐SR) control chart has been suggested in this study. Through the use of Monte Carlo simulation, the numerical analysis of the suggested NPACUSUM‐SR scheme has been calculated for zero and steady states. Run length (RL) profiles have been used to explore and compare the proposed NPACUSUM‐SR chart's in‐control resilience under various symmetrical heavy‐tailed, light‐tailed, and contaminated normal distributions. The suggested NPACUSUM‐SR chart has undergone a thorough investigation of the competition using Wilcoxon signed‐rank statistics and sign statistics under steady‐state and zero‐state conditions. The comparison shows that the suggested NPACUSUM‐SR design is superior to its competitors, especially when considering steady‐state performance over a wide range of shifts. This study examines the viability of the suggested design through a manufacturing‐related application.

“…The PD‐ARL is calculated as:

( {\frac{{AR{L_0} - AR{L_1}}}{{AR{L_0}}}} ) \times 100\%

(cf. Abbas et al 39 …”

Section: Performance Evaluationmentioning

confidence: 99%

Design and analysis of exponentially weighted moving average control charts for monitoring the variability of log‐normal processes with estimated parameters

Akhtar

Abid

Amir

et al. 2021

Self Cite

In statistical process control (SPC), the control chart is a quite popular technique to monitor the process efficacy. From a statistical point of view, the control chart is considered superior if it has an effective structure withholding property of the resistance against infrequent situations in a practical environment. The current study is designed for the same purpose for observing the dispersion parameter of log‐normal distribution by using the structure of an exponentially weighted moving average (EWMA) chart. For the extensive study, EWMA range, EWMA standard deviation, EWMA Qn,${Q_n},$ EWMA Sn,${S_n},$ EWMA mean absolute deviation, and EWMA transform standard deviation control charts are proposed. Properties of the run‐length profile of the proposed designed structures based on different existing estimators as well as a newly transformed dispersion estimator for log‐normal standard deviation are evaluated. The results indicate that the newly developed scheme outperforms the competitors when the dispersion parameter of the log‐normal distribution attains a large value. A real‐life application is also provided to validate how the developed design can be used in practice.