Distribution‐Free Adaptive Step‐Down Procedure for Fault Identification

Turkoz, Mehmet; Kim, Sangahn; Jeong, Young-Seon; Al-Khalifa, Khalifa N.

doi:10.1002/qre.2096

Cited by 8 publications

(1 citation statement)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section, we assess the performance of the proposed chart in identifying the faulty variables. To evaluate the performance of fault identification, we utilize two metrics, correctness ratio (CR) and the expected error rate (EER) outlined by (Zou et al, 2011;Turkoz et al, 2016;2019). where γ ∈ , and are index sets of truly shifted variables and identified variables, respectively.…”

Section: Performance In Process Diagnosticsmentioning

confidence: 99%

Monitoring of group-structured high-dimensional processes via sparse group LASSO

Kim,

Turkoz,

Jeong

et al. 2024

Ann Oper Res

Self Cite

View full text Add to dashboard Cite

In a general high-dimensional process, a large number of process parameters or quality characteristics is found to be featured through their dependencies and relevance. The features that have similar characteristics or behaviors in the process operation can be categorized into multiple groups. Thus, when a few quality characteristics in the process change, it is highly probable that the process shift would have occurred in a few relevant groups. Recently, several advanced statistical process control techniques are developed to monitor the changes in high-dimensional processes under sparsity. However, monitoring schemes that utilize the grouped pattern of the quality characteristics are sparse. This paper proposes a new method to monitor the high-dimensional process when the grouped structure of the process data is observed. The proposed method identifies the potentially changed groups and individual variables within the groups based on a modified sparse group LASSO (MSGL) model. Then, a monitoring statistic is obtained using MSGL-based likelihood function to test abnormality of the process. Extensive numerical studies are conducted to demonstrate the effectiveness and efficiency of the proposed method. In addition, a real-life application of a liquefied natural gas process is presented to illustrate the proposed method.

show abstract

Section: Performance In Process Diagnosticsmentioning

confidence: 99%

Monitoring of group-structured high-dimensional processes via sparse group LASSO

Kim,

Turkoz,

Jeong

et al. 2024

Ann Oper Res

Self Cite

View full text Add to dashboard Cite

show abstract

Special Issue on INFORMS 2015 Annual Meeting

Chen

Jeong

2016

Quality & Reliability Eng

View full text Add to dashboard Cite

E ach year, the Institute for Operations Research and the Management Sciences (INFORMS) Annual Meeting provides an excellent platform for researchers world-wide to present their state-of-the-art research in quality and reliability. The objective of this Special Issue is to introduce the recent advancements in quality and reliability methodological studies as well as applications in a form of full-length papers expanded from the presentations in the INFORMS 2015 Annual Meeting (November 1-4, Philadelphia, PA).We are very grateful for the contributions of the authors who submitted papers. After two to three rounds of rigorous reviews, 10 papers were accepted for publication. Among them, five papers focus on process monitoring and fault identification. The paper by Turkoz et al. proposes a new fault identification method when a high dimensional process is out-of-control. The proposed method combines the support vector data description-based test statistic with an adaptive step-down procedure to identify the faults. By using a nonparametric one-class classification method, the proposed approach does not rely on any distribution assumption. Compared with the existing distribution free methods, the proposed method has much more stable performance when the number of faults is more than one. The paper by Choe et al. focuses on the off-line change-point detection problem for time-series data. It adopts the Thresholded Least Absolute Shrinkage and Selection Operator (LASSO) techniques to control the false positives. The authors demonstrated the superior performance of the proposed method by comparing with several benchmark methods based on both simulations and a case study related to solar panel performance. The paper proposed by Zang, Wang, and Jin is handling with the issue of monitoring of processes based on unaligned profiles. While existing works focus on monitoring of well-aligned profiles, this paper develops new algorithms for monitoring unaligned profiles with varying sampling points. For this, they propose a robust dynamic time warping algorithm for profile alignment that are robust to noises and shift signals. And then, they propose a penalization-based charting algorithm that gives more effective performance in shift detection. In order to illustrate their proposed framework, they applied it for unaligned profile monitoring to the ingot growth process and monitoring heating power profiles. The paper by Abdella et al. proposes the double Exponentially Weighted Moving Average based procedure to evaluate the quality of a process based on polynomial quality profiles. The simulations studies have revealed the distinctive performance of their proposed double Exponentially Weighted Moving Average based control charts in quickly detecting changes in the second-order polynomial profiles. Their extensive simulation studies are based on two shift patterns in the polynomial quality profiles: changes in the coefficients of the regression parameters and changes in the process variability. Timely detection of whether a data ...

show abstract

Multi-class Bayesian support vector data description with anomalies

Turkoz

Kim

2021

Ann Oper Res

Self Cite

View full text Add to dashboard Cite

Support vector data description (SVDD) procedure fits a spherically shaped boundary around the normal data by minimizing the volume of the description. However, the SVDD may not find an efficient boundary if the normal data consist of multiple classes. In addition to the multi-class normal data, some anomaly observations can be available. We propose a generalized SVDD procedure which finds multiple spheres around the multi-class data by incorporating the anomaly observations into the training procedure. Thus, descriptions for each class include as many as their corresponding class observations by keeping the other class and anomaly observations as far as possible. Moreover, we introduce a generalized Bayesian framework which utilizes the relationships among the classes by not only considering the prior information from normal classes but also the anomaly class. Experiments with various simulation studies and real-life applications demonstrate that the proposed approach can effectively identify the anomalies in multi-class data.

show abstract

Distribution‐Free Adaptive Step‐Down Procedure for Fault Identification

Cited by 8 publications

References 22 publications

Monitoring of group-structured high-dimensional processes via sparse group LASSO

Monitoring of group-structured high-dimensional processes via sparse group LASSO

Special Issue on INFORMS 2015 Annual Meeting

Multi-class Bayesian support vector data description with anomalies

Contact Info

Product

Resources

About