Quality Characteristic Of Interest Research Articles

Developing Bayesian EWMA chart for change detection in the shape parameter of Inverse Gaussian process.

Bayesian Control charts are emerging as the most efficient statistical tools for monitoring manufacturing processes and providing effective control over process variability. The Bayesian approach is particularly suitable for addressing parametric uncertainty in the manufacturing industry. In this study, we determine the monitoring threshold for the shape parameter of the Inverse Gaussian distribution (IGD) and design different exponentially-weighted-moving-average (EWMA) control charts based on different loss functions (LFs). The impact of hyperparameters is investigated on Bayes estimates (BEs) and posterior risks (PRs). The performance measures such as average run length (ARL), standard deviation of run length (SDRL), and median of run length (MRL) are employed to evaluate the suggested approach. The designed Bayesian charts are evaluated for different settings of smoothing constant of the EWMA chart, different sample sizes, and pre-specified false alarm rates. The simulative study demonstrates the effectiveness of the suggested Bayesian method-based EWMA charts as compared to the conventional classical setup-based EWMA charts. The proposed techniques of EWMA charts are highly efficient in detecting shifts in the shape parameter and outperform their classical counterpart in detecting faults quickly. The proposed technique is also applied to real-data case studies from the aerospace manufacturing industry. The quality characteristic of interest was selected as the monthly industrial production index of aircraft from January 1980 to December 2022. The real-data-based findings also validate the conclusions based on the simulative results.

Triple sampling X̄ control chart for gamma process

ABSTRACT Monitoring production processes to provide potential customers with quality products while utilising fewer resources and reducing the cost of inspection is a challenge to quality and industrial engineers. Aiming at developing an improved design using reduced sample sizes, this study proposes a triple sampling (TS) X ¯ chart to monitor shifts in the process mean when the quality characteristic of interest follows the gamma distribution. Optimal parameters are determined by employing three optimisation procedures. The detection ability of the TS X ¯ chart is evaluated and contrasted with competing charts via the average number of observations to signal (ANOS), average run length (ARL) and expected ANOS (EANOS) criteria. When the TS X ¯ chart is designed for detecting small and large shifts, it is noticed that the sample sizes at the first, second and third sampling stages are larger for detecting small shifts. Compared to its competitors, the TS X ¯ chart requires fewer observations on the average to signal a shift; hence, it uses resources more efficiently and incurs lower inspection costs. To aid in the implementation of the TS X ¯ chart, an illustrative example of monitoring the wafer process is provided. Overwhelming evidence based on reported results supports the TS X ¯ chart.

Estimation of the lot fraction defective in a finite lot of products with auxiliary quality characteristics

Many companies use sampling plans for the acceptance or rejection of lots of products. The final outcome of this decision-making process is based on the inspection of a sample of products selected from a lot under inspection, where a quality characteristic is observed. An important parameter of interest related to acceptance sampling for attributes is the proportion of defective items. This parameter is unknown for a given lot of products, but it can be estimated from the aforementioned sample information. Additional quality characteristics can be observed at the inspection stage. We propose to use this auxiliary information to obtain more accurate estimators of the lot fraction defective at the estimation stage. Various relevant applications of this process are described. For possible scenarios that may arise in practice, the empirical properties of the suggested estimation methods are investigated using Monte Carlo simulations, and desirable results are obtained when there is a strong relationship between the quality characteristic of interest and the auxiliary quality characteristic.

Dynamic monitoring of polynomial profiles with attribute responses and between-profile correlation

Profile monitoring is a popular statistical process control problem in recent years. In many applications, quality characteristics of interest are attribute data due to the inherent feature of processes or limitation of data collection costs. The correlation among data is becoming more significant, since data collection intervals are becoming shorter in the big data era. However, research on monitoring profiles with attribute responses in the presence of Between-Profile Correlation (BPC) has received relatively scant attention. Motivated by a real example of automobile warranty claims, this article aims to monitor polynomial profiles with attribute responses and BPC. The generalized polynomial model and the learning curve model are adopted to characterize the profile relationship and the correlation between profiles, respectively. Then, an EWMA chart with dynamic control limits (dEWMA) is developed. Simulation studies show that ignoring the BPC does not affect the in-control performance of the chart with dynamic control limits, but does have devastating effects on the out-of-control performance. The proposed dEWMA chart can address the impact of correlation and provide superior monitoring performance compared with some competitors. Finally, a real example of warranty claims data is presented to illustrate the implementation of the proposed chart.

Probability‐type, Shewhart‐type and EWMA control charts for individual observations from the two‐parameter Lindley distribution

AbstractLindley distribution is a continuous distribution with various applications some of which are in medicine, genetics, epidemiology, biology, finance and actuarial sciences, ecology, meteorology, sociology, demography, agriculture, hydrology, geosciences, reliability and engineering, life testing and survival analysis, airborne systems and communications, environmental studies and modeling and describing of human mistakes, strikes, accidents, behavioural and emotional or IQ test scores and waiting times of customers in queues until service, etc. Due to its variety of applications, it appears to be important that control charts for detected shifts in a process should be constructed under the assumption that the quality characteristic of interest follows a Lindley‐related distribution. Here, we construct probability‐type, Shewhart‐type and EWMA control charts (and deal with the optimal choice of its parameters) for individual observations from the two‐parameter Lindley distribution, investigate and compare their performance and illustrate them using examples with both simulated and real data. The whole analysis reveals the superiority of using skewness correction for the construction of the control charts against not using it.

Effect of Measurement Error on the Multivariate CUSUM Control Chart for燙ompositional Data

Control charts (CCs) are one of the main tools in Statistical Process Control that have been widely adopted in manufacturing sectors as an effective strategy for malfunction detection throughout the previous decades. Measurement errors (M.E’s) are involved in the quality characteristic of interest, which can effect the CC’s performance. The authors explored the impact of a linear model with additive covariate M.E on the multivariate cumulative sum (CUSUM) CC for a specific kind of data known as compositional data (CoDa). The average run length is used to assess the performance of the proposed chart. The results indicate that M.E’s significantly affects the multivariate CUSUM-CoDa CCs. The authors have used the Markov chain method to study the impact of different involved parameters using six different cases for the variance-covariance matrix (VCM) (i.e., uncorrelated with equal variances, uncorrelated with unequal variances, positively correlated with equal variances, positively correlated with unequal variances, negatively correlated with equal variances and negatively correlated with unequal variances). The authors concluded that the error VCM has a negative impact on the performance of the multivariate CUSUM-CoDa CC, as the increases with an increase in the value of the error VCM. The subgroup size m and powering operator b positively impact the proposed CC, as the decreases with an increase in m or b. The number of variables p also has a negative impact on the performance of the proposed CC, as the values of increase with an increase in p. For the implementation of the proposal, two illustrated examples have been reported for multivariate CUSUM-CoDa CCs in the presence of M.E’s. One deals with the manufacturing process of uncoated aspirin tablets, and the other is based on monitoring the machines involved in the muesli manufacturing process.

Two‐dimensional wavelet based statistical monitoring of image data

AbstractStatistical process monitoring techniques are widely used to monitor process stability. In high volume and highly automated processes, effective methods are required to monitor several quality characteristics of interest. Advances in imaging technology have paved the grounds to capture and analyze product images from different perspectives in real time. Advantages of considering images have led to their increasing role in process monitoring. When images are used for process monitoring, complicated data structures should be analyzed. Various methods including data transformation and dimension reduction have been proposed in the literature to cope with these complexities. In this paper, a two‐dimensional (2D) wavelet based method is proposed to monitor image data. A likelihood ratio test (LRT) is developed to monitor the expected values of wavelet coefficient, and the derived likelihood ratio based statistic is monitored over time to detect out‐of‐control conditions as well as change point estimation. This information can help practitioners to diagnose the process, find the potential assignable cause(s) and return the process to the in‐control state. One of the advantages of the proposed method compared to other competing methods is that the performance of the proposed method does not depend on the fault location. Simulation studies indicate satisfactory performance for the proposed method in detecting out‐of‐control conditions and estimating change points. Moreover, the performance of the proposed method is also compared to two existing approaches. The results reveal that the proposed approach performs superior in detecting either diagonal or square defects.

The triple exponentially weighted moving average control chart for monitoring Poisson processes

AbstractIn many real applications, the quality characteristics of interest, such as the number of nonconformities in a production unit or the annual incidence rate, may follow the Poisson distribution. In the present article, we propose a triple exponentially weighted moving average control chart for monitoring Poisson processes (regarded as PTEWMA chart). One‐ and two‐sided schemes of the proposed chart are designed and are compared with the existing PEWMA and PDEWMA charts in terms of the average run‐length (ARL). The results show that for small values of the smoothing parameter , the PTEWMA chart is more effective than its competitors in detecting small shifts while its superiority increases as the value of increases. Moreover, the two‐sided PTEWMA chart is less ARL‐biased for small downward shifts than the other two charts. Two illustrative examples are provided to demonstrate the application of the proposed chart.

Empirical likelihood ratio charts for profiles with attribute data and random predictors in the presence of within‐profile correlation

AbstractProfile monitoring has been an increasingly popular statistical process control (SPC) problem recently. In many applications, quality characteristics of interest are attribute data, which can be adequately characterized by generalized linear models. Most existing control charts for monitoring such profiles in the literature assume that observations within each profile are independent, which is often violated in practice. This paper aims to provide a unified framework for Phase II monitoring of profiles with attribute data and random predictors in the presence of within‐profile correlation. To this end, a working correlation matrix is introduced to take the dependency into account. Then, a new control scheme is designed based on the empirical likelihood ratio test. The advantages of the proposed method are that both changes in the mean and the correlation can be detected, and the correlation structure is assumption‐free. Simulation studies show that the proposed method outperforms the existing schemes in nearly all cases, and it is more effective and robust. A real example of automobile warranty claims is demonstrated to illustrate the implementation of the proposed control chart.

Functional regression control chart for monitoring ship CO2$_{2}$ emissions

AbstractOn modern ships, the quick development in data acquisition technologies is producing data‐rich environments where variable measurements are continuously streamed and stored during navigation and thus can be naturally modelled as functional data or profiles. Then, both the CO emissions (i.e. the quality characteristic of interest) and the variable profiles that have an impact on them (i.e. the covariates) are called to be explored in the light of the new worldwide and European regulations on the monitoring, reporting and verification of harmful emissions. In this paper, we show an application of the functional regression control chart (FRCC) with the ultimate goal of answering, at the end of each ship voyage, the question: given the value of the covariates, is the observed CO emission profile as expected? To this aim, the FRCC focuses on the monitoring of residuals obtained from a multivariate functional linear regression of the CO emission profiles on the functional covariates. The applicability of the FRCC is demonstrated through a real‐case study of a Ro‐Pax ship operating in the Mediterranean Sea. The proposed FRCC is also compared with other alternatives available in the literature and its advantages are discussed over some practical examples.

An extended nonparametric homogeneously weighted moving average sign control chart

AbstractParametric (or traditional) control charts are based on the assumption that the quality characteristic of interest follows a specific distribution. However, in many applications, there is a lack of knowledge about the underlying distribution. To this end, nonparametric (or distribution‐free) control charts have been developed in recent years. In this article, a nonparametric double homogeneously weighted moving average (DHWMA) control chart based on the sign statistic is proposed for monitoring the location parameter of an unknown and continuous distribution. The performance of the proposed chart is measured through the run‐length distribution and its associated characteristics by performing Monte Carlo simulations. The DHWMA sign chart is compared with other nonparametric sign charts, such as the homogeneously weighted moving average, generally weighted moving average (GWMA), double GWMA, and triple exponentially weighted moving average sign charts, as well as the traditional DHWMA chart. The results indicate that the proposed chart performs just as well as and in some cases better than its competitors, especially for small shifts. Finally, two examples are provided to show the application and implementation of the proposed chart.

Improved simple linear profiling method with application to chemical gas sensors

AbstractIn industrial applications, a functional relationship between the quality characteristics of interest is well represented by profile functions. These are often characterized by linear or nonlinear relationships that depend on the nature of the experimental study. A monitoring method based on linear profiles is known as linear profiling, which is commonly used because of its simplicity. In this article, we propose a new linear profiling method based on a homogenously weighted moving average (HWMA) chart to monitor the linear profile parameters. The HWMA chart is based on the idea of allocating a specific weight to the current estimate of the profiling parameter, and the remaining weight is circulated equally among the prior estimates. The performance of the proposed control chart was evaluated using the average run length criterion and further compared with existing charts for the monitoring of the linear profile parameter. Our simulation results revealed the superiority of the newly proposed chart in terms of its quick detection ability. An example concerning gas sensors from the chemical industry is also presented to illustrate the application of the proposed chart.

Intelligent parameters optimization for laser cutting of highly reflective and thermally conductive materials using artificial neural network

High energy density laser beam is a best approach to cut difficult- to- laser cut materials. Laser is most suitable to cut highly reflective and thermal conductive materials i.e. Aluminium and its alloy which is difficult to laser cut material. The optical and thermal characteristics of material are the main proficiency of laser beam machining. The industries as automobiles and aircraft which need complicated design or difficult portrait forward with inflexible model condition adopt that types of material extensively. The quality of laser cutting mainly possible with the appropriate selection of input process parameters. For optimal setting of process parameters, the effect of their variation on different quality characteristics of interest is required to be investigated. In this paper the simulation test results show that the simulated values of material removal rate (MRR) using neural network model are quite close to the values calculated with theoretical formulae. To express the relation between cutting quality and cutting specification the artificial neural network (ANN) approach is established for vertical laser cutting position. For optimizing the parameters, concluding cutting results and surmising new cutting information the artificial neural network (ANN) provides outstanding outcomes.

Retraction Note to: Development of censored data-based robust design for pharmaceutical quality by design

Robust design techniques, which are based on the concept of building quality into products or processes, are increasingly popular in many manufacturing industries. In this paper, we propose a new robust design model in the context of pharmaceutical production research and development. Traditional robust design principles have often been applied to situations in which the quality characteristics of interest are typically time insensitive. In pharmaceutical manufacturing processes, time-oriented quality characteristics, such as the degradation of a drug, are often of interest. As a result, current robust design models for quality improvement which have been studied in the literature may not be effective in finding robust design solutions. To address such practical needs, this paper develops a robust design model using censored data, which is perhaps the first attempt in the robust design field. We then study estimation methods, such as the expectation–maximization algorithm and the maximum likelihood method, in the robust design context. Finally, comparative studies are discussed for model verification via a numerical example.

Design of experiments in the service industry: results from a global survey and directions for further research

PurposeThe purpose of this study is to critically evaluate the current status of Design of experiments (DoE) in the service industry or non-manufacturing processes in the manufacturing sector. This involves the identification of critical success factors and challenges in the use of DoE in the context of the service industry as well as the key skills needed for its successful applications.Design/methodology/approachIn order to identify the critical success factors and challenges, an online survey instrument was developed and distributed across Six Sigma Black Belts (SSBB), Six Sigma Master Black Belts (SSMBB), process improvement champions, continuous improvement professionals, DoE experts and specialists as well as academics who are involved in DoE research and applications.FindingsA total of 109 participants from over 20 countries responded to the study. One of the first findings of the study is that over 40% of the process improvement professionals in the service sector do not have any hands-on experience with the application of DoE. Further, this study highlights the following critical factors for successful DoE applications: dedicating adequate resources in terms of time, budget and people for the experiment; getting a clear understanding of the problem at hand; securing a reliable and validated measurement system for measuring the response or quality characteristics of interest; and identifying input independent variables, which influence the quality characteristics during the brainstorming stage. Some respondents also mention the importance of having training or education in DoE as a critical success factor. At the same time, the lack of adequate planning to experimentation, mathematical knowledge and statistical reasoning remains as challenge to be solved.Research limitations/implicationsAs data were collected at an individual level, inter-reliability, which measures the consistency of judgements on a particular survey item, could not be gauged. Moreover, due to low sample size across some continents, an intercontinental comparative study on the extent of the application of DoE across various continents could not be performed.Originality/valueTo the best of the authors’ knowledge, this is the first global survey carried out to understand the status of DoE in the service industry. The results of the global study can be used to encourage quality, continuous improvement and process improvement professionals in the service sector to drive wider acceptance of one of the most powerful tools in the Six Sigma toolkit for problem-solving scenarios. Moreover, senior managers in the service industry can take away some of the key benefits, challenges and success factors in the implementation of DoE.

On the effect of the measurement error on Shewhart t and EWMA t control charts

In the industrial practice, control charts are frequently implemented assuming that the quality characteristic of interest can be accurately measured without errors. In general, this assumption is not realistic: measurement error always exists in quality control applications and may considerably affect the performance of control charts in detecting the occurrence of an out-of-control condition. In this paper, the effect of measurement error on the statistical performance of Shewhart t and EWMA t control charts is investigated. Several tables are provided to show how the statistical performance of these control charts changes with different sources of the measurement error. The obtained results show that the measurement errors have a significant influence on the performance of both the Shewhart t and EWMA t control charts. Two examples in the analytical chemistry and food industry are presented to illustrate the use of the proposed charts.

A control chart to monitor the process mean based on inspecting attributes using control limits of the traditional X-bar chart

ABSTRACTThis paper proposes a new control chart, denoted by , for evaluating the stability of a process mean, which is based on attribute inspection. In the control chart, the mean of the quality characteristic of interest is controlled by a go-no-go gauge device that generates five categorizations. In equally-spaced times, samples of n items are collected, their categories are determined, and random values are generated according to a truncated normal distribution that corresponds to their categories. With these random values, the averages are estimated and the control charts are built. Extensive computational experiments shown that the newly introduced control charts perform similarly to the usual control chart in terms of average run length, provided that the sample sizes are raised by approximately three additional sample units. Because the control charts use attributes, they can be considered a viable alternative to the traditional control chart.

On designing an efficient control chart to monitor fraction nonconforming

AbstractProcess control measures are mostly applied in production and manufacturing industries. The most important tool used in these disciplines is control chart. In manufacturing and production processes, when the quality characteristic of interest cannot be directly measured, it becomes essential to apply attribute control charts. To monitor fraction nonconforming of the output, quality practitioners mostly prefer p‐chart. In this article, a new progressive mean (PM) control chart is being proposed for monitoring drift in proportion of nonconforming products. The design evaluations of the proposed chart are made and compared through different properties of run length distribution, such as average run length (ARL), standard deviation of run length (SDRL), and some percentile points. The performance of the proposed chart is assessed under zero‐state and steady‐state scenarios. The proposed PM chart is compared with p‐chart, moving average (MA) chart, optimal CUSUM chart, modified exponentially weighted moving average (EWMA) chart, and runs rules p‐charts for monitoring fraction nonconforming. The proposed chart spots efficiently sustained disturbances in the process as compared with their existing counterparts. Two illustrative examples are also provided; one from real‐life application of nonconforming bearing and seal assemblies data and the other from simulated data for the implementation of PM chart.

A nonparametric generally weighted moving average sign chart based on repetitive sampling

Most control charts that assume quality characteristics of interest follow a normal or specific distribution. However, in reality there is often limited or no information regarding the underlying process distribution. Therefore, in recent years, nonparametric techniques for quality control have been developed. In this paper, the nonparametric generally weighted moving average sign chart based on repetitive sampling (hereinafter RS-GWMA sign chart) is proposed to improve performance capability of existing charts in small process shifts. Simulation studies show that the nonparametric RS-GWMA sign chart with large design and adjustment parameters outperforms competing charts considered in this article. The fill volume of soft-drink beverage bottles is as industrial example used to illustrate the application of the proposed nonparametric RS-GWMA sign chart.

A comparative study on Poisson control charts

ABSTRACTIn many real applications, the quality characteristics of interest may be in the form of attribute counts which are often modeled by a Poisson distribution. For the monitoring of such data, many different control charts have been developed. In this article, we compare the performance of several widely used Poisson control charts in terms of the average run-length. Based on the comparison study, some recommendations are made in order to choose the best-performing control chart under different situations. Moreover, we apply these charts in real data relating to the airplane accidents.