In this paper, we show that concept of Statistical Process Control tools was thoroughly examined and the definitions of quality control concepts were presented. This is significant because of it is anticipated that this study will contribute to the literature as an exemplary application that demonstrates the role of statistical process control (SPC) tools in quality improvement in the evaluation and decision-making phase.

This is significant because of this study is to investigate applications of quality control, to clarify statistical control methods and problem-solving procedures, to generate proposals for problem-solving approaches, and to disseminate improvement studies in the ready-to-wear industry. The basic Statistical Process Control tools used in the study, the most repetitive faults were detected and these faults were divided into sub-headings for more detailed analysis. In this way, it was tried to prevent the repetition of faults by going down to the root causes of any detected fault. With this different perspective, it is expected that the study will contribute to other fields.

We give consent for the publication of identifiable details, which can include photograph(s) and case history and details within the text (“Material”) to be published in the Journal of Quality Technology. We confirm that have seen and been given the opportunity to read both the Material and the Article (as attached) to be published by Taylor & Francis.

Data for benchmarking SPC against other process monitoring methods. The data consist of a one-dimensional timeseries of floats (x.csv). Addititionally information whether the data are within the specifications are provided as another time series (y.csv). The data are generated by solving an optimization problem for each time to generate a mixture distribution of different probability distributions. Then for each timestep one record is sampled. Inputs for the optimization problem are the given probability distributions, the lower and upper limit of the tolerance interval as well as the desired median of the data. Additionally weights of the different probability distributions can be given as boundary condions for the different time steps. Metadata generated from the solving are stored in k_matrix.csv (wheights at each time step) and distribs (probability distribution objects). The data consists of phases with data from a stable mixture distribution and phases with data from a mixture distribution that do not fulfill the stability criteria.

Because of the “curse of dimensionality,” high-dimensional processes present challenges to traditional multivariate statistical process monitoring (SPM) techniques. In addition, the unknown underlying distribution of and complicated dependency among variables such as heteroscedasticity increase the uncertainty of estimated parameters and decrease the effectiveness of control charts. In addition, the requirement of sufficient reference samples limits the application of traditional charts in high-dimension, low-sample-size scenarios (small n, large p). More difficulties appear when detecting and diagnosing abnormal behaviors caused by a small set of variables (i.e., sparse changes). In this article, we propose two change-point–based control charts to detect sparse shifts in the mean vector of high-dimensional heteroscedastic processes. Our proposed methods can start monitoring when the number of observations is a lot smaller than the dimensionality. The simulation results show that the proposed methods are robust to nonnormality and heteroscedasticity. Two real data examples are used to illustrate the effectiveness of the proposed control charts in high-dimensional applications. The R codes are provided online.

We present a new method for statistical process control (SPC) of a discrete part manufacturing system based on intrinsic geometrical properties of the parts, estimated from three-dimensional sensor data. An intrinsic method has the computational advantage of avoiding the difficult part registration problem, necessary in previous SPC approaches of three-dimensional geometrical data, but inadequate if noncontact sensors are used. The approach estimates the spectrum of the Laplace–Beltrami (LB) operator of the scanned parts and uses a multivariate nonparametric control chart for online process control. Our proposal brings SPC closer to computer vision and computer graphics methods aimed to detect large differences in shape (but not in size). However, the SPC problem differs in that small changes in either shape or size of the parts need to be detected, keeping a controllable false alarm rate and without completely filtering noise. An online or “Phase II” method and a scheme for starting up in the absence of prior data (“Phase I”) are presented. Comparison with earlier approaches that require registration shows the LB spectrum method to be more sensitive to rapidly detect small changes in shape and size, including the practical case when the sequence of part datasets is in the form of large, unequal size meshes. A post-alarm diagnostic method to investigate the location of defects on the surface of a part is also presented. While we focus in this article on surface (triangulation) data, the methods can also be applied to point cloud and voxel metrology data.

Datasets to the planned publication "Generalized Statistical Process Control via 1D-ResNet Pretraining" by Tobias Schulze, Louis Huebser, Sebastian Beckschulte and Robert H. Schmitt (Chair for Intelligence in Quality Sensing, Laboratory for Machine Tools and Production Engineering, WZL of RWTH Aachen University)

Data for benchmarking SPC against other process monitoring methods. The data consist of a one-dimensional timeseries of floats (x.csv). Addititionally information whether the data are within the specifications are provided as another time series (y.csv). The data are generated by solving an optimization problem for each time to generate a mixture distribution of different probability distributions. Then for each timestep one record is sampled. Inputs for the optimization problem are the given probability distributions, the lower and upper limit of the tolerance interval as well as the desired median of the data. Additionally weights of the different probability distributions can be given as boundary condions for the different time steps. Metadata generated from the solving are stored in k_matrix.csv (wheights at each time step) and distribs (probability distribution objects according to https://doi.org/10.5281/zenodo.8249487). The data consists of phases with data from a stable mixture distribution and phases with data from a mixture distribution that do not fulfill the stability criteria.

The train data were used to train the G-SPC model. The test data were used for benchmarking purposes

Funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy – EXC-2023 Internet of Production – 390621612.

Machine learning methods have been widely used in different applications, including process control and monitoring. For handling statistical process control (SPC) problems, conventional supervised machine learning methods (e.g., artificial neural networks and support vector machines) would have some difficulties. For instance, a training dataset containing both in-control and out-of-control (OC) process observations is required by a supervised machine learning method, but it is rarely available in SPC applications. Furthermore, many machine learning methods work like black boxes. It is often difficult to interpret their learning mechanisms and the resulting decision rules in the context of an application. In the SPC literature, there have been some existing discussions on how to handle the lack of OC observations in the training data, using the one-class classification, artificial contrast, real-time contrast, and some other novel ideas. However, these approaches have their own limitations to handle SPC problems. In this article, we extend the self-starting process monitoring idea that has been employed widely in modern SPC research to a general learning framework for monitoring processes with serially correlated data. Under the new framework, process characteristics to learn are well specified in advance, and process learning is sequential in the sense that the learned process characteristics keep being updated during process monitoring. The learned process characteristics are then incorporated into a control chart for detecting process distributional shift based on all available data by the current observation time. Numerical studies show that process monitoring based on the new learning framework is more reliable and effective than some representative existing machine learning SPC approaches.

ABSTRACT Laboratory tests for technical evaluation or irrigation material testing involve the measurement of many variables, as well as monitoring and control of test conditions. This study, carried out in 2016, aimed at using statistical quality control techniques to evaluate results of dripper tests. Exponentially weighted moving average control charts were elaborated, besides capability indices for the measurement of the test pressure and water temperature; and study on repeatability and reproducibility (Gage RR) of flow measurement system using 10 replicates, in three work shifts (morning, afternoon and evening), with 25 emitters. Both the test pressure and water temperature remained stable, with “excellent” performance for the pressure adjustment process by integrative-derivative proportional controller. The variability between emitters was the component with highest contribution to the total variance of the flow measurements, with 96.77% of the total variance due to the variability between parts. The measurement system was classified as “acceptable” or “approved” by the Gage RR study; and non-random causes of significant variability were not identified in the routine of tests.

The Statistical Process Control System (SPC) market has emerged as a critical component in quality management and process optimization across various industries, significantly enhancing operational efficiency and product quality. SPC utilizes statistical methods and tools to monitor and control manufacturing process

Del Castillo and Zhao have recently proposed a new methodology for the Statistical Process Control (SPC) of discrete parts whose 3-dimensional (3D) geometrical data are acquired with non-contact sensors. The approach is based on monitoring the spectrum of the Laplace–Beltrami (LB) operator of each scanned part estimated using finite element methods (FEM). The spectrum of the LB operator is an intrinsic summary of the geometry of a part, independent of the ambient space. Hence, registration of scanned parts is unnecessary when comparing them. The primary goal of this case study paper is to demonstrate the practical implementation of the spectral SPC methodology through multiple examples using real scanned parts acquired with an industrial-grade laser scanner, including 3D-printed parts and commercial parts. We discuss the scanned mesh preprocessing needed in practice, including the type of remeshing found to be most beneficial for the FEM computations. For each part type, both the “phase I” and “phase II” stages of the spectral SPC methodology are showcased. In addition, we provide a new principled method to determine the number of eigenvalues of the LB operator to consider for efficient SPC of a given part geometry, and present an improved algorithm to automatically define a region of interest, particularly useful for large meshes. Computer codes that implement every method discussed in this paper, as well as all scanned part datasets used in the case studies, are made available and explained in the supplementary materials.

The Statistical Process Control (SPC) Software market has increasingly become a cornerstone in quality management across various industries, including manufacturing, pharmaceuticals, and food processing. This specialized software leverages statistical methods to monitor and control production processes, ensuring tha

Source code and results for exactly regenerating the data and results of the simulation study reported in (to be published). Industrial sensory data is simulated and subjected to several data-driven methods for estimating temporal delays between the sensors. A new method is proposed to estimate these delays by optimizing multivariate correlations, and is shown to be more accurate than the currently more standard method that optimizes bivariate correlations.

Updated on November 6th, 2023 after internal review by authors.

Abstract–Statistical process control (SPC) charts are critically important for quality control and management in manufacturing industries, environmental monitoring, disease surveillance, and many other applications. Conventional SPC charts are designed for cases when process observations are independent at different observation times. In practice, however, serial data correlation almost always exists in sequential data. It has been well demonstrated in the literature that control charts designed for independent data are unstable for monitoring serially correlated data. Thus, it is important to develop control charts specifically for monitoring serially correlated data. To this end, there is some existing discussion in the SPC literature. Most existing methods are based on parametric time series modeling and residual monitoring, where the data are often assumed to be normally distributed. In applications, however, the assumed parametric time series model with a given order and the normality assumption are often invalid, resulting in unstable process monitoring. Although there is some nice discussion on robust design of such residual monitoring control charts, the suggested designs can only handle certain special cases well. In this article, we try to make another effort by proposing a novel control chart that makes use of the restarting mechanism of a CUSUM chart and the related spring length concept. Our proposed chart uses observations within the spring length of the current time point and ignores all history data that are beyond the spring length. It does not require any parametric time series model and/or a parametric process distribution. It only requires the assumption that process observation at a given time point is associated with nearby observations and independent of observations that are far away in observation times, which should be reasonable for many applications. Numerical studies show that it performs well in different cases.

Additional file 2. Ethics Statement Supplement. Contains details of MUSIC practices’ IRB status.

Batch effects are technical sources of variation introduced by the necessity of conducting gene expression analyses on different dates due to the large number of biological samples in population-based studies. The aim of this study is to evaluate the performances of linear mixed models (LMM) and Combat in batch effect removal. We also assessed the utility of adding quality control samples in the study design as technical replicates. In order to do so, we simulated gene expression data by adding “treatment” and batch effects to a real gene expression dataset. The performances of LMM and Combat, with and without quality control samples, are assessed in terms of sensitivity and specificity while correcting for the batch effect using a wide range of effect sizes, statistical noise, sample sizes and level of balanced/unbalanced designs. The simulations showed small differences among LMM and Combat. LMM identifies stronger relationships between big effect sizes and gene expression than Combat, while Combat identifies in general more true and false positives than LMM. However, these small differences can still be relevant depending on the research goal. When any of these methods are applied, quality control samples did not reduce the batch effect, showing no added value for including them in the study design.

In this study, a new distribution-free Phase I control chart for retrospectively monitoring multivariate data is developed. The suggested approach, based on the multivariate signed ranks, can be applied to individual or subgrouped data for detection of location shifts with an arbitrary pattern (e.g., isolated, transitory, sustained, progressive, etc.). The procedure is complemented with a LASSO-based post-signal diagnostic method for identification of the shifted variables. A simulation study shows that the method compares favorably with parametric control charts when the process is normally distributed, and largely outperforms other multivariate nonparametric control charts when the process distribution is skewed or heavy-tailed. An R package can be found in the supplementary material.

Replication dataset for "A quality improvement project using statistical process control methods for Type 2 diabetes control in a resource-limited setting" (doi: 10.1093/intqhc/mzx051).

corrMatrix.m: This Matlab function computes the correlation matrix of w-test statistics.

KMC.m: This Matlab function computes the critical values for max-w test statistic based on Monte Carlo method. It is needed to run corrMatrix.m before use it.

kNN.m: This Matlab function based on neural networks allows anyone to obtain the desired critical value with good control of type I error. In that case, you need to download file SBPNN.mat and save it in your folder. It is needed to run corrMatrix.m before use it.

SBPNN.mat: MATLAB's flexible network object type (called SBPNN.mat) that allows anyone to obtain the desired critical value with good control of type I error.

Examples.txt: File containing examples of both design and covariance matrices in adjustment problems of geodetic networks.

rawMC.txt: Monte-Carlo-based critical values for the following signifiance levels: α′= 0.001, α′= 0.01, α′= 0.05, α′= 0.1 and α′= 0.5. The number of the observations (n) were fixed for each α ′with n = 5 to n= 100 by a increment of 5. For each "n" the correlation between the w-tests (ρwi,wj) were also fixed from ρwi,wj = 0.00 to ρwi,wj = 1.00, by increment of 0.1, considering also taking into account the correlation ρwi,wj = 0.999. For each combination of α′,"n" and ρwi,wj, m= 5,000,000 Monte Carlo experiments were run.

In statistical process control, accurately estimating in-control (IC) parameters is crucial for effective monitoring. This typically requires a Phase I analysis to obtain estimates before monitoring commences. The traditional “fixed” estimate (FE) approach uses these estimates exclusively, while the “adaptive” estimate (AE) approach updates the estimates with each new observation. Such extreme criteria reflect the traditional bias-variance tradeoff in the framework of the sequential parameter learning schemes. This paper proposes an intermediate update rule that generalizes two ad hoc criteria for monitoring univariate Gaussian data, by giving a lower probability to parameter updates when an out-of-control (OC) situation is likely, therefore updating more frequently when there is no evidence of an OC scenario. The simulation study shows that this approach improves the detection power for small and early shifts, which are commonly regarded as a weakness of control charts based on fully online adaptive estimation. The paper also shows that the proposed method performs similarly to the fully adaptive procedure for larger or later shifts. The proposed method is illustrated by monitoring the sudden increase in ICU counts during the 2020 COVID outbreak in New York.