Recognition of control chart patterns using improved selection of features

Gauri, Susanta Kumar; Chakraborty, Shankar

doi:10.1016/j.cie.2008.10.006

Cited by 89 publications

(54 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It can be noted that Gauri and Chakraborty (2009) have extracted various features of CCPs without segmenting as well as by segmenting the available data plot of the observations. It is found that the features extracted without segmenting the observation window are not well capable to discriminate all the nine types of CCPs.…”

Section: Extraction Of Selected Shape Featuresmentioning

confidence: 99%

“…Considering a moving observation window of size N = 32 and assuming that a sampling interval in the control chart plot is represented by a linear distance, c = 1σ, Gauri and Chakraborty (2009) have extracted 30 shape features, and then selected a set of seven shape features that can differentiate eight types of CCPs using a CART-based systematic approach.…”

Section: Extraction Of Selected Shape Featuresmentioning

confidence: 99%

“…In this paper, using the CART-based systematic approach (Gauri and Chakraborty, 2009), a new set of seven most appropriate shape features is selected that can recognize all the nine main types of CCPs, including MIX pattern. Using these selected features, heuristic rules in the form of decision trees are developed using CART as well as QUEST algorithms for recognition of various CCPs.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Feature-based decision rules for control charts pattern recognition: A comparison between CART and QUEST algorithm

Bag¹,

Gauri²,

Chakraborty³

2012

ijiec

Self Cite

View full text Add to dashboard Cite

Control chart pattern (CCP) recognition can act as a problem identification tool in any manufacturing organization. Feature-based rules in the form of decision trees have become quite popular in recent years for CCP recognition. This is because the practitioners can clearly understand how a particular pattern has been identified by the use of relevant shape features. Moreover, since the extracted features represent the main characteristics of the original data in a condensed form, it can also facilitate efficient pattern recognition. The reported feature-based decision trees can recognize eight types of CCPs using extracted values of seven shape features. In this paper, a different set of seven most useful features is presented that can recognize nine main CCPs, including mixture pattern. Based on these features, decision trees are developed using CART (classification and regression tree) and QUEST (quick unbiased efficient statistical tree) algorithms. The relative performance of the CART and QUEST-based decision trees are extensively studied using simulated pattern data. The results show that the CART-based decision trees result in better recognition performance but lesser consistency, whereas, the QUEST-based decision trees give better consistency but lesser recognition performance.

show abstract

Section: Extraction Of Selected Shape Featuresmentioning

confidence: 99%

Section: Extraction Of Selected Shape Featuresmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Feature-based decision rules for control charts pattern recognition: A comparison between CART and QUEST algorithm

Bag¹,

Gauri²,

Chakraborty³

2012

ijiec

Self Cite

View full text Add to dashboard Cite

show abstract

“…A hybrid learning-based model was used to identify various types of unnatural CCPs in a process in [19]. A hybrid feature-based CCP classifier containing classification and regression trees, as well as NN techniques, was proposed for recognizing CCPs in [20].…”

Section: Related Workmentioning

confidence: 99%

Applying Two-Stage Neural Network Based Classifiers to the Identification of Mixture Control Chart Patterns for an SPC-EPC Process

Shao

Chang

2017

Complexity

View full text Add to dashboard Cite

The effective controlling and monitoring of an industrial process through the integration of statistical process control (SPC) and engineering process control (EPC) has been widely addressed in recent years. However, because the mixture types of disturbances are often embedded in underlying processes, mixture control chart patterns (MCCPs) are very difficult for an SPC-EPC process to identify. This can result in problems when attempting to determine the underlying root causes of process faults. Additionally, a large number of categories of disturbances may be present in a process, but typical single-stage classifiers have difficulty in identifying large numbers of categories of disturbances in an SPC-EPC process. Therefore, we propose a two-stage neural network (NN) based scheme to enhance the accurate identification rate (AIR) for MCCPs by performing dimension reduction on disturbance categories. The two-stage scheme includes a combination of a NN, support vector machine (SVM), and multivariate adaptive regression splines (MARS). Experimental results reveal that the proposed scheme achieves a satisfactory AIR for identifying MCCPs in an SPC-EPC system.

show abstract

“…Since unnatural control chart patterns (CCPs) associate a specific set of assignable causes in manufacturing processes, effective recognition of unnatural CCPs is an important issue in SPC. Five basic CCPs are commonly exhibited in control charts including normal (NOR), systematic (SYS), cyclic (CYC), trend (TRE) and shift (SHI) [2,3]. Except for the normal pattern, all other patterns are abnormal CCPs and indicate that the process being monitored is out of control and requires adjustment.…”

Section: Introductionmentioning

confidence: 99%

Recognition of Concurrent Control Chart Patterns by Integrating ICA and SVM

Lu¹,

Shao²,

Li³

2014

Appl. Math. Inf. Sci.

View full text Add to dashboard Cite

Control chart patterns (CCPs) recognition is an important issue in statistical process control, since abnormal CCPs exhibited in control charts can be associated with certain assignable causes and then help to quickly eliminate assignable causes of process variation. Most of the existing studies assume that the observed process data needs to be recognized are basic types of abnormal CCPs. However, in practical situations, the observed process data may be concurrent CCPs which are mixed together by two basic CCPs. In this study, an integrated scheme using independent component analysis (ICA) and support vector machine (SVM) is proposed for recognizing concurrent CCPs. The proposed ICA-SVM scheme initially uses ICA with concurrent patterns for generating independent components (ICs). The hidden basic patterns of the concurrent patterns can be discovered in these ICs. The ICs are then used as input variables for the SVM for building a CCP recognition model. Experimental results reveal that the proposed ICA-SVM scheme can produce accurate and stable recognition results. It outperforms the three comparison models and is able to effectively recognize concurrent control chart patterns.

show abstract

Recognition of control chart patterns using improved selection of features

Cited by 89 publications

References 17 publications

Feature-based decision rules for control charts pattern recognition: A comparison between CART and QUEST algorithm

Feature-based decision rules for control charts pattern recognition: A comparison between CART and QUEST algorithm

Applying Two-Stage Neural Network Based Classifiers to the Identification of Mixture Control Chart Patterns for an SPC-EPC Process

Recognition of Concurrent Control Chart Patterns by Integrating ICA and SVM

Contact Info

Product

Resources

About