Monitoring Correlated Profile and Multivariate Quality Characteristics

Amiri, Amirhossein; Zou, Changliang; Doroudyan, Mohammad Hadi

doi:10.1002/qre.1483

Cited by 30 publications

(19 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They applied the Gaussian process models. Amiri et al evaluated the correlated simple linear profiles and multivariate normal quality characteristics. They proposed a method based on MEWMA control chart to monitor the correlated profiles and multivariate quality characteristics simultaneously.…”

Section: Introductionmentioning

confidence: 99%

Monitoring simple linear profiles in the presence of generalized autoregressive conditional heteroscedasticity

Hadizadeh

Soleimani

2017

Quality & Reliability Eng

View full text Add to dashboard Cite

In this paper, monitoring of simple linear profiles is investigated in the presence of nonequality of variances or heteroscedasticity, ie, generalized autoregressive conditional heteroscedasticity. In this condition, using of the common methods regardless of the heteroscedasticity leads to the fault interpretations. We consider a simple linear profile and assume that there is a generalized autoregressive conditional heteroscedasticity (GARCH) (1,1) model within the profiles. Here, we particularly focus on Phase II monitoring of simple linear regression. We studied the generalized autoregressive conditional heteroscedasticity effect, briefly GARCH effect, on the average run length criterion. As the remedial measures, the weighted least squares method to estimate the regression parameters and the heteroscedasticity‐consistent approaches to estimate the covariance matrix of regression parameters, are used to extract the GARCH effect. Two control chart methods namely T2 and exponentially weighted moving average 3 are discussed to monitor the simple linear profiles. Their performances are evaluated by using the average run length criterion. Finally, a real case from an industry field is studied.

show abstract

Section: Introductionmentioning

confidence: 99%

Monitoring simple linear profiles in the presence of generalized autoregressive conditional heteroscedasticity

Hadizadeh

Soleimani

2017

Quality & Reliability Eng

View full text Add to dashboard Cite

show abstract

“…Among them, Kim et al (2006) proposed a multi-channel profile monitoring method based on Principal Curves to monitor multiple homogeneous signals in forging processes. Amiri et al (2013) investigated the problem of integrated monitoring of mixed-type data, i.e. profile data and multivariate quality characteristics, possibly coming from multiple sources.…”

Section: Introductionmentioning

confidence: 99%

Profile monitoring via sensor fusion: the use of PCA methods for multi-channel data

Grasso

Colosimo

Pacella

2014

International Journal of Production Research

View full text Add to dashboard Cite

Continuous advances of sensor technology and real-time computational capability is leading to data rich environments to improve industrial automation and machine intelligence. When multiple signals are acquired from different sources (i.e., multi-channel signal data), two main issues must be faced: i) reduce data dimensionality, to make the overall signal analysis system efficient and actually implementable in industrial environment, and ii) fuse together all the sensor outputs to achieve a better comprehension of the process. In this frame, Multi-way Principal Component Analysis (PCA) represents a multivariate technique to perform both the tasks. The paper investigates two main multi-way extensions of the traditional PCA to deal with multi-channel signals, one based on unfolding the original datasets, and one based on multi-linear analysis of data in their tensorial form. The approaches proposed for data modelling are combined with appropriate control charting to achieve multi-channel profile data monitoring.The developed methodologies are demonstrated with both simulated and real data. IntroductionThe development of low-cost, non-intrusive and smart sensors on one hand, and the continuous improvement of real-time computational capability on the other hand, make a large amount of data potentially available in industry. In this frame, sensor signals acquired during the process provide a suitable source of information to develop an inprocess quality control and to allow a faster implementation of corrective actions. In several applications, the acquired signals present cyclically repeating patterns; in those cases the suite of profile monitoring techniques (Woodall et al., 2004;Williams et al., 2007) provides the natural framework to evaluate the stability over time of processquality. An overview of parametric and nonparametric approaches for profile data as well as application domains investigated at this time can be found in the recent book edited by Noorossana et al. (2012).This paper focuses on the specific case of monitoring profiles that are signal data.On this topic, the first seminal paper on signal profile monitoring is due to Jin and Shi (1999), who suggested using wavelet analysis to monitor tonnage signals in stamping The largest portion of profile monitoring literature focuses on single signal analysis, regardless the strong industrial interest for multi-signal applications. Data-rich environments in industry, in fact, are leading to an increasing demand for multi-sensor data fusion methods to solve quality-related problems. The most widely studied applications in literature include stability analysis and chatter detection (Kuljanic et al., 2009;Inasaki, 1999) and tool condition monitoring (Cho et al., 2010;Wang et al., 2007;Chen and Jen, 2000;Bahr et al., 1997;Bhattacharyya and Sengupta, 2009;Lezanski, 2001;Shi and Gindy, 2007;Aliustaoglu et al. 2009). However, only few authors studied profile monitoring approaches in the field of sensor fusion. Among them, Kim et al. In this study, we co...

show abstract

“…Profile practical applications have been reported by many researchers, including Stover & Brill (1998), Kang & Albin (2000), Mahmoud & Woodall (2004), Wang & Tsung (2005) and Kusiak, Zheng & Song (2009). Several control chart approaches for monitoring simple linear profiles have been developed by Kang & Albin (2000), Kim, Mahmoud & Woodall (2003), Zou, Zhang & Wang (2006), Zou, Zhou, Wang & Tsung (2007), Mahmoud, Parker, Woodall & Hawkins (2007), Soleimani, Narvand & Raissi (2013), Zhang, He, Zhang & Woodall (2013), Yeh & Zerehsaz (2013) and Amiri, Zou & Doroudyan (2014). Proposals for monitoring multivariate linear profiles (simple and/or multiple) have been developed by Mahmoud (2008), Noorossana, Eyvazian & Vaghefi (2010), Noorossana, Eyvazian, Amiri & Mahmoud (2010), Eyvazian, Noorossana, Saghaei & Amiri (2011) and Zou, Ning & Tsung (2012).…”

Section: Introductionmentioning

confidence: 99%

Profile Monitoring for Compositional Data

Guevara-González

Vargas-Navas

Linero-Segrera

2014

Rev. colomb. estad.

View full text Add to dashboard Cite

In a growing number of quality control applications, the quality of a product or process is best characterized and summarized by a functional relationship between a response variable and one or more explanatory variables. Profile monitoring is used to understand and to check the stability of this relationship over time. In some applications with compositional data, the relationship can be characterized by a Dirichlet regression model. We evaluate five T 2 control charts for monitoring these profiles in Phase I. A real example from production of concrete is given.Key words: Control chart, Dirichlet distribution, Statistical process control. ResumenEn un gran número de aplicaciones la calidad de un producto o proceso está mejor representada por una relación funcional entre una variable de respuesta y una o más variables explicatorias. El monitoreo de perfiles permite entender y chequear la estabilidad de esta relación funcional a través del tiempo. En algunas aplicaciones con datos composicionales, la relación puede ser representada por un modelo de regresión Dirichlet. En este artículo nosotros evaluamos cinco cartas de control T 2 para monitorear estos perfiles en Fase I. Un ejemplo real asociado a la producción de concreto es presentado.Palabras clave: carta de control, control estadístico de procesos, distribución Dirichlet.a Assistant professor.

show abstract

Monitoring Correlated Profile and Multivariate Quality Characteristics

Cited by 30 publications

References 34 publications

Monitoring simple linear profiles in the presence of generalized autoregressive conditional heteroscedasticity

Monitoring simple linear profiles in the presence of generalized autoregressive conditional heteroscedasticity

Profile monitoring via sensor fusion: the use of PCA methods for multi-channel data

Profile Monitoring for Compositional Data

Contact Info

Product

Resources

About