Statistical monitoring of multiple profiles simultaneously using Gaussian processes

Jahani, Salman; Kontar, Raed Al; Veeramani, Dharmaraj; Zhou, Shiyu

doi:10.1002/qre.2326

Cited by 19 publications

(6 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the manufacturing processes are often affected by time or space. For example, both the correlations within profiles and between profiles for multivariate profile data based on a multivariate Gaussian process model were considered by Jahani et al Thus, the correlations between multivariate nonlinear profiles for p < m 0 need to be further studied in future research. Moreover, when the number of historical data set ( m 0 ) is smaller than the number of key quality characteristics ( p ), the correlation problem for p ≥ m 0 needs to be further studied too.…”

Section: Conclusion and Future Research Areasmentioning

confidence: 99%

Detecting the process changes for multivariate nonlinear profile data

Pan

Meng

2019

Quality & Reliability Eng

View full text Add to dashboard Cite

In profile monitoring for a multivariate manufacturing process, the functional relationship of the multivariate profiles rarely occurs in linear form, and the real data usually do not follow a multivariate normal distribution. Thus, in this paper, the functional relationship of multivariate nonlinear profile data is described via a nonparametric regression model. We first fit the multivariate nonlinear profile data and obtain the reference profiles through support vector regression (SVR) model. The differences between the observed multivariate nonlinear profiles and the reference profiles are used to calculate the vector of metrics. Then, a nonparametric revised spatial rank exponential weighted moving average (RSREWMA) control chart is proposed in the phase II monitoring. Moreover, a simulation study is conducted to evaluate the detecting performance of our proposed nonparametric RSREWMA control chart under various process shifts using out-of-control average run length (ARL 1 ). The simulation results indicate that the SREWMA control chart coupled with the metric of mean absolute deviation (MAD) can be used to monitor the multivariate nonlinear profile data when a common fixed design (CFD) is not applicable in the phase II study. Finally, a realistic multivariate nonlinear profile example is used to demonstrate the usefulness of our proposed RSREWMA control chart and its monitoring schemes. KEYWORDS common fixed design (CFD), mean absolute deviation (MAD), multivariate nonlinear profile data, spatial rank exponential weighted moving average (SREWMA) control chart, support vector regression (SVR)

show abstract

Section: Conclusion and Future Research Areasmentioning

confidence: 99%

Detecting the process changes for multivariate nonlinear profile data

Pan

Meng

2019

Quality & Reliability Eng

View full text Add to dashboard Cite

show abstract

“…Walker and Wright [31] examined the density of the wood board. Recently, Jahani et al [33] studied the performance of the ice-making process and characterized it over different temperature signals. Thus, a study for the nonlinear profile is required.…”

Section: Supplier Selection Via Process Yield Indexmentioning

confidence: 99%

Multiple Comparisons With the Best and Difference Test Statistic for Supplier Selection for Nonlinear Profiles

Negash

Kusumaningrum

2019

IEEE Access

View full text Add to dashboard Cite

It is vital to choose the right supplier to reduce cost and provide high-quality products. However, a gap remains because the supplier's process is mainly measured using qualitative and intangible criteria. Further, with technological advances, the measurement of quality characteristics is transforming through smart data sensors. With a specific time or space measurements can be done at high frequency. The functional relationships between the measures or profiles can be established. The profiles indicate the pattern in the data. The literature focused on the case when quality characteristics are described by linear profile and consider symmetric tolerance. However, in a real-world application, nonlinear profiles and asymmetric tolerance is frequently found. This study proposed multiple comparisons with the best and the difference test statistic methods to select the best suppliers when the quality characteristics are described by nonlinear profile with asymmetric tolerances. A Monte Carlo simulation study is conducted, computer programs are written in the R programming language. The result indicated in terms of rejecting inferior suppliers, the multiple comparisons with the best method perform better than the difference test statistics. With the proposed methods, managers can make decisions using a single, easy-to-understand index. Also, these methods can handle any number of suppliers. For the convenience of a decision-maker, critical values, and profile size requirements are provided. An illustrative example is included to give a better insight into the proposed methods.INDEX TERMS Supplier selection, nonlinear profile, multiple comparisons with the best, difference test statistics, process yield indices.

show abstract

“…Using a multivariate Gaussian process model (MGP), Jahani et al 17 . proposed a model for monitoring multivariate linear profiles to deal with within‐profile and between‐profile autocorrelations.…”

Section: Introductionmentioning

confidence: 99%

Simultaneous monitoring of mean vector and covariance matrix of auto‐correlated multivariate multiple linear profiles

Rahimi

Amiri

Khoo

et al. 2022

Quality & Reliability Eng

View full text Add to dashboard Cite

The aim of this paper is to develop control charts for a simultaneous monitoring of the mean vector and covariance matrix of multivariate multiple linear regression profiles in phase II, when the independence assumption of the observations within each profile is violated, and there is multivariate autoregressive moving average (MARMA)(1,1) autocorrelation structure within each profile. A transformation method is applied to eliminate the autocorrelation, and six control charts for a simultaneous monitoring of the mean vector and covariance matrix of auto‐correlated multivariate multiple linear regression profiles are proposed. Then, through simulation runs, in terms of the average run length (ARL) criterion, it is shown that the proposed methods have better performance in comparison with the competing methods in detecting shifts in the parameters of regression profiles. Finally, the application of the proposed control charts is shown using real data in a machining process.

show abstract

Statistical monitoring of multiple profiles simultaneously using Gaussian processes

Cited by 19 publications

References 25 publications

Detecting the process changes for multivariate nonlinear profile data

Detecting the process changes for multivariate nonlinear profile data

Multiple Comparisons With the Best and Difference Test Statistic for Supplier Selection for Nonlinear Profiles

Simultaneous monitoring of mean vector and covariance matrix of auto‐correlated multivariate multiple linear profiles

Contact Info

Product

Resources

About