Pub Date : 2023-01-23DOI: 10.1080/00224065.2022.2151530
Mohammed Saif Ismail Hameed, José Núñez Ares, P. Goos
Abstract Experimental data are often highly structured due to the use of experimental designs. This does not only simplify the analysis, but it allows for tailored methods of analysis that extract more information from the data than generic methods. One group of three-level experimental designs that are suitable for such tailored methods are orthogonal minimally aliased response surface (OMARS) designs (Núñez Ares and Goos 2020), where all main effects are orthogonal to each other and to all second-order effects. The design based analysis method of Jones and Nachtsheim (2017) has shown significant improvement over existing methods in terms of powers to detect active effects. However, the application of their method is limited to only a small subgroup of OMARS designs known as definitive screening designs (DSDs). In our work, we not only improve upon the Jones and Nachtsheim method for DSDs, but we also generalize their analysis framework to the entire family of OMARS designs. Using extensive simulations, we show that our customized method for analyzing data from OMARS designs is highly effective in identifying active effects when compared to other modern (non-design based) analysis methods, especially in cases where the true model is complex and involves many second-order effects.
{"title":"Analysis of data from orthogonal minimally aliased response surface designs","authors":"Mohammed Saif Ismail Hameed, José Núñez Ares, P. Goos","doi":"10.1080/00224065.2022.2151530","DOIUrl":"https://doi.org/10.1080/00224065.2022.2151530","url":null,"abstract":"Abstract Experimental data are often highly structured due to the use of experimental designs. This does not only simplify the analysis, but it allows for tailored methods of analysis that extract more information from the data than generic methods. One group of three-level experimental designs that are suitable for such tailored methods are orthogonal minimally aliased response surface (OMARS) designs (Núñez Ares and Goos 2020), where all main effects are orthogonal to each other and to all second-order effects. The design based analysis method of Jones and Nachtsheim (2017) has shown significant improvement over existing methods in terms of powers to detect active effects. However, the application of their method is limited to only a small subgroup of OMARS designs known as definitive screening designs (DSDs). In our work, we not only improve upon the Jones and Nachtsheim method for DSDs, but we also generalize their analysis framework to the entire family of OMARS designs. Using extensive simulations, we show that our customized method for analyzing data from OMARS designs is highly effective in identifying active effects when compared to other modern (non-design based) analysis methods, especially in cases where the true model is complex and involves many second-order effects.","PeriodicalId":54769,"journal":{"name":"Journal of Quality Technology","volume":"1 1","pages":"366 - 384"},"PeriodicalIF":2.5,"publicationDate":"2023-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76126148","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-20DOI: 10.1080/00224065.2022.2147884
S. Kloppenborg
Abstract In the aerospace composites industry, new material lots are tested to determine if they are suitable for use. It is common to accept or reject the material lot by comparing the sample mean and lower extremum to reference values that are established based on an initial (qualification) sample of material property measurements. Current industry practices assume that the samples are drawn from a normal distribution with known parameters equal to the mean and standard deviation of the qualification sample: this assumption yields a producer’s risk that is too high. This article presents a two-sample method of setting these reference values, considering the sampling distribution of the qualification sample. This new method is validated through simulation which shows that it produces the correct probability of Type I error. Simulation is also used to investigate the statistical power of the new method and it is compared to others commonly used. A case study is presented to demonstrate the use of the new method using composite material data from industry.
{"title":"Lot acceptance testing using sample mean and extremum with finite qualification samples","authors":"S. Kloppenborg","doi":"10.1080/00224065.2022.2147884","DOIUrl":"https://doi.org/10.1080/00224065.2022.2147884","url":null,"abstract":"Abstract In the aerospace composites industry, new material lots are tested to determine if they are suitable for use. It is common to accept or reject the material lot by comparing the sample mean and lower extremum to reference values that are established based on an initial (qualification) sample of material property measurements. Current industry practices assume that the samples are drawn from a normal distribution with known parameters equal to the mean and standard deviation of the qualification sample: this assumption yields a producer’s risk that is too high. This article presents a two-sample method of setting these reference values, considering the sampling distribution of the qualification sample. This new method is validated through simulation which shows that it produces the correct probability of Type I error. Simulation is also used to investigate the statistical power of the new method and it is compared to others commonly used. A case study is presented to demonstrate the use of the new method using composite material data from industry.","PeriodicalId":54769,"journal":{"name":"Journal of Quality Technology","volume":"264 1","pages":"336 - 348"},"PeriodicalIF":2.5,"publicationDate":"2023-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79728458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-12DOI: 10.1080/00224065.2022.2148590
S. Grimshaw
Abstract SPC with positive autocorrelation is well known to result in frequent false alarms if the autocorrelation is ignored. The autocorrelation is a nuisance and not a feature that merits modeling and understanding. This paper proposes exhaustive systematic sampling, which is similar to Bayesian thinning except no observations are dropped, to create a pooled variance estimator that can be used in Shewhart control charts with competitive performance. The expected value and variance are derived using quadratic forms that is nonparametric in the sense no distribution or time series model is assumed. Practical guidance for choosing the systematic sampling interval is offered to choose a value large enough to be approximately unbiased and not too big to inflate variance. The proposed control charts are compared to time series residual control charts in a simulation study that validates using the empirical reference distribution control limits to preserve stated in-control false alarm probability and demonstrates similar performance.
{"title":"Constructing control charts for autocorrelated data using an exhaustive systematic samples pooled variance estimator","authors":"S. Grimshaw","doi":"10.1080/00224065.2022.2148590","DOIUrl":"https://doi.org/10.1080/00224065.2022.2148590","url":null,"abstract":"Abstract SPC with positive autocorrelation is well known to result in frequent false alarms if the autocorrelation is ignored. The autocorrelation is a nuisance and not a feature that merits modeling and understanding. This paper proposes exhaustive systematic sampling, which is similar to Bayesian thinning except no observations are dropped, to create a pooled variance estimator that can be used in Shewhart control charts with competitive performance. The expected value and variance are derived using quadratic forms that is nonparametric in the sense no distribution or time series model is assumed. Practical guidance for choosing the systematic sampling interval is offered to choose a value large enough to be approximately unbiased and not too big to inflate variance. The proposed control charts are compared to time series residual control charts in a simulation study that validates using the empirical reference distribution control limits to preserve stated in-control false alarm probability and demonstrates similar performance.","PeriodicalId":54769,"journal":{"name":"Journal of Quality Technology","volume":"7 1","pages":"349 - 365"},"PeriodicalIF":2.5,"publicationDate":"2023-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74910736","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-12DOI: 10.1080/00224065.2022.2147108
C. Y. Hong, D. Fletcher, Jiaxu Zeng, C. McGraw, C. Cornwall, V. Cummings, N. Barr, Emily J. Frost, P. Dillingham
Abstract Split-plot experimental data are often analyzed as if the data came from a completely randomized design. As is well known, ignoring the different levels of randomization and replication can lead to serious inferential errors. However, in some experiments, including many of the ocean global change experiments that motivated this research, variation between whole-plot experimental units may be small relative to variation between subplot units. Even though a factorial analysis will often perform poorly in general, in this special case it outperforms a split-plot analysis, providing narrower confidence intervals for treatment means and differences with coverage rates close to the desired level. The performance of the proposed model-averaged analysis was compared to a classical split-plot analysis via a simulation study, and its utility demonstrated for an ocean global change experiment examining growth and condition of a juvenile mussel species. In our simulation study, model-averaged confidence intervals for whole-plot treatment means or comparisons of means were up to 40% narrower than split-plot confidence intervals while maintaining close to nominal coverage rates. In our example experiment, we observed narrowing of up to 25%. We recommend model averaging as a preferred approach when variation between whole-plot experimental units is expected to be less than between subplot units, with a few caveats for studies with very few replicates.
{"title":"Efficient analysis of split-plot experimental designs using model averaging","authors":"C. Y. Hong, D. Fletcher, Jiaxu Zeng, C. McGraw, C. Cornwall, V. Cummings, N. Barr, Emily J. Frost, P. Dillingham","doi":"10.1080/00224065.2022.2147108","DOIUrl":"https://doi.org/10.1080/00224065.2022.2147108","url":null,"abstract":"Abstract Split-plot experimental data are often analyzed as if the data came from a completely randomized design. As is well known, ignoring the different levels of randomization and replication can lead to serious inferential errors. However, in some experiments, including many of the ocean global change experiments that motivated this research, variation between whole-plot experimental units may be small relative to variation between subplot units. Even though a factorial analysis will often perform poorly in general, in this special case it outperforms a split-plot analysis, providing narrower confidence intervals for treatment means and differences with coverage rates close to the desired level. The performance of the proposed model-averaged analysis was compared to a classical split-plot analysis via a simulation study, and its utility demonstrated for an ocean global change experiment examining growth and condition of a juvenile mussel species. In our simulation study, model-averaged confidence intervals for whole-plot treatment means or comparisons of means were up to 40% narrower than split-plot confidence intervals while maintaining close to nominal coverage rates. In our example experiment, we observed narrowing of up to 25%. We recommend model averaging as a preferred approach when variation between whole-plot experimental units is expected to be less than between subplot units, with a few caveats for studies with very few replicates.","PeriodicalId":54769,"journal":{"name":"Journal of Quality Technology","volume":"359 1","pages":"318 - 335"},"PeriodicalIF":2.5,"publicationDate":"2023-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76394776","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.1080/00224065.2023.2160575
Allison Jones-Farmer
papillomavirus
乳头瘤病毒
{"title":"Message from the Editor","authors":"Allison Jones-Farmer","doi":"10.1080/00224065.2023.2160575","DOIUrl":"https://doi.org/10.1080/00224065.2023.2160575","url":null,"abstract":"papillomavirus","PeriodicalId":54769,"journal":{"name":"Journal of Quality Technology","volume":"8 1","pages":"1 - 2"},"PeriodicalIF":2.5,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87094549","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.1080/00224065.2022.2041377
Joseph David Reviewer: Conklin
Here is the book for everyone in our field who has ever been asked about the meaning and value of statistics in the modern world. This is the book to give to your nonstatistician acquaintances the next time you get this question. For the rest of us, read this book and buckle up for an exciting adventure in the advance of datadriven knowledge and interdisciplinary collaboration. The book portrays, in admirable sweep and detail, the tireless efforts of statisticians and computer scientists under the auspices of the University of Brazil. Their goal is to create, maintain, and advance an online platform for COVID-19 pandemic prediction, one that can run on computers, notebooks, tablets, and mobile phones. Their aim is nothing less than a platform for predicting pandemic infections and deaths both in the short term—up to two weeks—and in the long term—until the end of the current wave of COVID-19 within a given state, region, and ultimately any country on the planet. The adventure plays out in 17 chapters:
这本书是为我们这个领域的每一个曾经被问到现代世界统计的意义和价值的人写的。下次你们遇到这个问题的时候可以把这本书给你们的非统计学家朋友。对于我们其余的人来说,阅读这本书,并在数据驱动的知识和跨学科合作的进步中做好准备进行令人兴奋的冒险。这本书以令人钦佩的全面和细节描绘了巴西大学(University of Brazil)赞助下统计学家和计算机科学家的不懈努力。他们的目标是创建、维护和推进一个可以在电脑、笔记本电脑、平板电脑和手机上运行的COVID-19大流行预测在线平台。他们的目标是建立一个预测大流行感染和死亡的平台,无论是在短期内(最多两周),还是在长期内(直到当前一波COVID-19在特定州、地区乃至最终在地球上任何国家结束)。这次冒险共分17章:
{"title":"Building a Platform for Data-Driven Pandemic Prediction from Data Modeling to Visualization – The CovidLP Project","authors":"Joseph David Reviewer: Conklin","doi":"10.1080/00224065.2022.2041377","DOIUrl":"https://doi.org/10.1080/00224065.2022.2041377","url":null,"abstract":"Here is the book for everyone in our field who has ever been asked about the meaning and value of statistics in the modern world. This is the book to give to your nonstatistician acquaintances the next time you get this question. For the rest of us, read this book and buckle up for an exciting adventure in the advance of datadriven knowledge and interdisciplinary collaboration. The book portrays, in admirable sweep and detail, the tireless efforts of statisticians and computer scientists under the auspices of the University of Brazil. Their goal is to create, maintain, and advance an online platform for COVID-19 pandemic prediction, one that can run on computers, notebooks, tablets, and mobile phones. Their aim is nothing less than a platform for predicting pandemic infections and deaths both in the short term—up to two weeks—and in the long term—until the end of the current wave of COVID-19 within a given state, region, and ultimately any country on the planet. The adventure plays out in 17 chapters:","PeriodicalId":54769,"journal":{"name":"Journal of Quality Technology","volume":"325 1","pages":"119 - 120"},"PeriodicalIF":2.5,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76563049","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-03DOI: 10.1080/00224065.2022.2128946
Mengmeng Liu, Robert W. Mee, Yongdao Zhou
Abstract Definitive screening designs (DSDs) have grown rapidly in popularity since their introduction by Jones and Nachtsheim (2011). Their appeal is that the second-order response surface (RS) model can be estimated in any subset of three factors, without having to perform a follow-up experiment. However, their usefulness as a one-step RS modeling strategy depends heavily on the sparsity of second-order effects and the dominance of first-order terms over pure quadratic terms. To address these limitations, we show how viewing a projection of the design region as spherical and augmenting the DSD with axial points in factors found to involve second-order effects remedies the deficiencies of a stand-alone DSD. We show that augmentation with a second design consisting of axial points is often the D s -optimal augmentation, as well as minimizing the average prediction variance. Supplemented by this strategy, DSDs are highly effective initial screening designs that support estimation of the second-order RS model in three or four factors.
{"title":"Augmenting definitive screening designs: Going outside the box","authors":"Mengmeng Liu, Robert W. Mee, Yongdao Zhou","doi":"10.1080/00224065.2022.2128946","DOIUrl":"https://doi.org/10.1080/00224065.2022.2128946","url":null,"abstract":"Abstract Definitive screening designs (DSDs) have grown rapidly in popularity since their introduction by Jones and Nachtsheim (2011). Their appeal is that the second-order response surface (RS) model can be estimated in any subset of three factors, without having to perform a follow-up experiment. However, their usefulness as a one-step RS modeling strategy depends heavily on the sparsity of second-order effects and the dominance of first-order terms over pure quadratic terms. To address these limitations, we show how viewing a projection of the design region as spherical and augmenting the DSD with axial points in factors found to involve second-order effects remedies the deficiencies of a stand-alone DSD. We show that augmentation with a second design consisting of axial points is often the D s -optimal augmentation, as well as minimizing the average prediction variance. Supplemented by this strategy, DSDs are highly effective initial screening designs that support estimation of the second-order RS model in three or four factors.","PeriodicalId":54769,"journal":{"name":"Journal of Quality Technology","volume":"210 1","pages":"289 - 301"},"PeriodicalIF":2.5,"publicationDate":"2022-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77586878","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-22DOI: 10.1080/00224065.2022.2109533
Wei-Heng Huang, Jing Sun, A. Yeh
Abstract As data acquisition and processing technologies continue to advance rapidly, new challenges emerge for statistical process monitoring. One such challenge, especially in the era of big data analytics, is monitoring multivariate processes involving a mixture of continuous, categorical, and discrete quality variables. The existing multivariate control charts focus mostly on monitoring correlated variables of the same type. We propose a new Phase II control chart that is based on a modified Holm’s step-down multiple testing procedure (Holm 1979) which achieves two important goals at the same time: (1) it simultaneously monitors correlated variables of different types, while keeping the probability of false alarm under desirable level, and (2) when the process is determined to be out of control, it further provides, without any additional efforts, diagnostics to pinpoint which parameters are out of control. The proposed chart is shown to outperform the existing charts particularly in its ability to provide more accurate diagnostics.
{"title":"Monitoring and diagnostics of correlated quality variables of different types","authors":"Wei-Heng Huang, Jing Sun, A. Yeh","doi":"10.1080/00224065.2022.2109533","DOIUrl":"https://doi.org/10.1080/00224065.2022.2109533","url":null,"abstract":"Abstract As data acquisition and processing technologies continue to advance rapidly, new challenges emerge for statistical process monitoring. One such challenge, especially in the era of big data analytics, is monitoring multivariate processes involving a mixture of continuous, categorical, and discrete quality variables. The existing multivariate control charts focus mostly on monitoring correlated variables of the same type. We propose a new Phase II control chart that is based on a modified Holm’s step-down multiple testing procedure (Holm 1979) which achieves two important goals at the same time: (1) it simultaneously monitors correlated variables of different types, while keeping the probability of false alarm under desirable level, and (2) when the process is determined to be out of control, it further provides, without any additional efforts, diagnostics to pinpoint which parameters are out of control. The proposed chart is shown to outperform the existing charts particularly in its ability to provide more accurate diagnostics.","PeriodicalId":54769,"journal":{"name":"Journal of Quality Technology","volume":"36 1","pages":"220 - 252"},"PeriodicalIF":2.5,"publicationDate":"2022-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83716465","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-22DOI: 10.1080/00224065.2022.2110024
Shin-Fu Tsai
Abstract Dispersion effects may play a vital role, in addition to location effects, in exploring optimal addition orders of several materials in some chemical, industrial and pharmaceutical studies. Two replication-based statistical methods developed using frequentist and fiducial probability arguments are introduced in this paper to identify active dispersion effects from replicated order-of-addition experiments. Simulation results show that both approaches can maintain empirical sizes sufficiently close to the nominal level while their finite-sample performances are very similar. From a statistical perspective, the fiducial method can provide a unified probability framework to analyze dispersion effects as well as location effects. However, it is computationally more expensive than the frequentist method. Consequently, the frequentist method is recommended for real-world applications due to its low computational cost. A drug combination study is used to illustrate these two approaches. In addition, some eligible order-of-addition designs are collected in a catalogue for future work.
{"title":"Analyzing dispersion effects from replicated order-of-addition experiments","authors":"Shin-Fu Tsai","doi":"10.1080/00224065.2022.2110024","DOIUrl":"https://doi.org/10.1080/00224065.2022.2110024","url":null,"abstract":"Abstract Dispersion effects may play a vital role, in addition to location effects, in exploring optimal addition orders of several materials in some chemical, industrial and pharmaceutical studies. Two replication-based statistical methods developed using frequentist and fiducial probability arguments are introduced in this paper to identify active dispersion effects from replicated order-of-addition experiments. Simulation results show that both approaches can maintain empirical sizes sufficiently close to the nominal level while their finite-sample performances are very similar. From a statistical perspective, the fiducial method can provide a unified probability framework to analyze dispersion effects as well as location effects. However, it is computationally more expensive than the frequentist method. Consequently, the frequentist method is recommended for real-world applications due to its low computational cost. A drug combination study is used to illustrate these two approaches. In addition, some eligible order-of-addition designs are collected in a catalogue for future work.","PeriodicalId":54769,"journal":{"name":"Journal of Quality Technology","volume":"15 1","pages":"271 - 288"},"PeriodicalIF":2.5,"publicationDate":"2022-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81100640","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-16DOI: 10.1080/00224065.2022.2106912
Peiyao Liu, Juan Du, Yangyang Zang, Chen Zhang, Kaibo Wang
Abstract Nowadays advanced sensing technology enables real-time data collection of key variables during manufacturing, known as multi-channel profiles. These data facilitate in-process monitoring and anomaly detection, which have been extensively studied in recent years. However, most studies treat each profile as a whole, e.g., a high-dimensional vector or function, and construct monitoring schemes accordingly. As a result, these methods cannot be implemented until the entire profile has been obtained, leading to long detection delay especially if anomalies occur in early sensing points of the profile. In addition, they require that profiles of different samples have the same time length and feature location, yet additional time-warping operation for real misaligned samples may weaken the anomaly patterns. To address these problems, this article proposes an in-profile monitoring (INPOM) control chart, which not only gives the feasibility of detecting anomalies inside the profile, but also can handle the misalignment problem of different samples. In particular, our INPOM scheme is built upon state space model (SSM). To better describe the clustered between-profile correlation and avoid overfitting, SSM is extended to a regularized SSM (RSSM), where regularizations are imposed as prior information and expectation maximization algorithm is integrated for posterior maximization to efficiently learn the model parameters. Furthermore, a monitoring statistic based on one-step-ahead prediction error of RSSM is constructed for INPOM control chart. Thorough numerical studies and real case studies demonstrate the effectiveness and applicability of our proposed RSSM-INPOM framework.
{"title":"In-profile monitoring for cluster-correlated data in advanced manufacturing system","authors":"Peiyao Liu, Juan Du, Yangyang Zang, Chen Zhang, Kaibo Wang","doi":"10.1080/00224065.2022.2106912","DOIUrl":"https://doi.org/10.1080/00224065.2022.2106912","url":null,"abstract":"Abstract Nowadays advanced sensing technology enables real-time data collection of key variables during manufacturing, known as multi-channel profiles. These data facilitate in-process monitoring and anomaly detection, which have been extensively studied in recent years. However, most studies treat each profile as a whole, e.g., a high-dimensional vector or function, and construct monitoring schemes accordingly. As a result, these methods cannot be implemented until the entire profile has been obtained, leading to long detection delay especially if anomalies occur in early sensing points of the profile. In addition, they require that profiles of different samples have the same time length and feature location, yet additional time-warping operation for real misaligned samples may weaken the anomaly patterns. To address these problems, this article proposes an in-profile monitoring (INPOM) control chart, which not only gives the feasibility of detecting anomalies inside the profile, but also can handle the misalignment problem of different samples. In particular, our INPOM scheme is built upon state space model (SSM). To better describe the clustered between-profile correlation and avoid overfitting, SSM is extended to a regularized SSM (RSSM), where regularizations are imposed as prior information and expectation maximization algorithm is integrated for posterior maximization to efficiently learn the model parameters. Furthermore, a monitoring statistic based on one-step-ahead prediction error of RSSM is constructed for INPOM control chart. Thorough numerical studies and real case studies demonstrate the effectiveness and applicability of our proposed RSSM-INPOM framework.","PeriodicalId":54769,"journal":{"name":"Journal of Quality Technology","volume":"1 1","pages":"195 - 219"},"PeriodicalIF":2.5,"publicationDate":"2022-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83423333","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}