Statistical Process Control Scheme Research Articles

An artificial neural network approach for out‐of‐control stream identification in multiple stream processes

AbstractA multiple stream process (MSP) is a process at a point in time that generates several streams of output with quality variables and specifications that are identical in all streams. Statistical process control (SPC) schemes for MSPs are designed to issue an out‐of‐control (OC) signal when a possible change occurs from the in‐control (IC) state of the process. However, the SPM literature for MSPs lacks contributions regarding post‐hoc identification of the stream or group of streams responsible for an OC signal. In this article, we propose an artificial neural network (NN) specifically trained for this purpose and, through a Monte Carlo simulation, show its superiority in correctly identifying OC streams, since an OC signal has been correctly issued. The research is motivated by the need for a post hoc diagnosis of the heating, ventilation, and air conditioning (HVAC) systems installed on modern train coaches, generating multiple streams of quality variables of interest. A case study in this area based on the HVAC data (openly available online at https://github.com/unina‐sfere/NN4OCMSP) illustrates the practical applicability of the proposed approach, which is implemented in the Python package NN4OCMSP and published on the PyPI software repository.

Multivariate SPC via sequential multiblock-PLS

The sequential multi-block partial least squares (SMB-PLS) is proposed for implementing a multivariate statistical process control scheme. This is of interest when the system is composed of several blocks following a sequential order and presenting correlated information, for instance, a raw material properties block followed by a process variables block that is manipulated according to raw material properties. The SMB-PLS uses orthogonalization to separate correlated information between blocks from orthogonal variations. This allows monitoring the system in different stages considering only the remaining orthogonal part in each block. Thus, the SMB-PLS increases the interpretability and process understanding in the model building (Phase I), since it provides a deep insight about the nature of the system variations. Besides, it prevents any special cause from propagating to subsequent blocks enabling their use in the model exploitation (Phase II). The methodology is applied to a real case study from a food manufacturing process.

A two-stage fault diagnosis strategy for air handling units via a backpropagation multidimensional Taylor network fitter and a novel statistical process control

Faults in Heating, Ventilation, and Air Conditioning Air Handling Units can lead to increased energy consumption and failure to meet human comfort standards. Accurate and prompt fault diagnosis of Air Handling Units is therefore critical. Although widely-used deep learning methods have shown high precision in fault diagnosis, they present two main drawbacks: 1) deep learning approaches are often “black box” models lacking interpretability and rely on labeled data; 2) their complex structures entail substantial computational resources, which may not be conducive to timely fault diagnosis. To overcome these challenges, this paper introduces a method for diagnosing faults in air handling units that combines a novel statistical process control strategy, a backpropagation multidimensional Taylor network fitter, and a D-matrix approach. The backpropagation multidimensional Taylor network, using a polynomial network with only one hidden layer, approximates nonlinear functions. Compared to deep learning models, it maintains high accuracy with a simpler structure, enabling rapid fault diagnosis. Additionally, the novel statistical process control strategy employs the Markov property for fault detection, characterized by low false alarm rates, interpretability, robustness, minimal labeled data requirements, and suitability for real-time monitoring. Experimental results validate the effectiveness of our method in diagnosing faults in air handling units, achieving an average precision of 99.9%, surpassing several commonly used techniques.

A stage-wise non-Gaussian statistical process control strategy

The article delves into the development of a Non-Gaussian Process Monitoring Strategy for a Copper Cathode Manufacturing Unit (CCMU). The monitoring strategy being devised highlighted the issue of multi-stage process monitoring via the usage of Multi-block Independent Component Analysis (MBICA) techniques. MBICA is the multi-block variant of ICA technique which is prevalently used for process laden with non-Gaussian or non-normal data. Development of the monitoring strategy involved detection of fault(s) and their subsequent diagnosis. Detection of fault(s) was carried out by employment of I2 control chart whose control limit was established via Bootstrap procedure. The diagnosis of the detected fault was carried out by employment of fault diagnostic statistic. An amalgamation of MBICA and Multivariate Exponentially Weighted Moving Average (MEWMA) are also known as MBICA-MEWMA approach was also proposed for detection of incipient fault(s). The monitoring strategy thus developed was showcased for a CCMU which specialised in the manufacture of copper cathode which has got varied practical applications. The monitoring strategy thus devised was able to detect and diagnose the faults with appreciable accuracy.

An effective statistical process control scheme for industrial environmental monitoring

One major source of environmental pollution is industrial effluents, especially the heavy metal pollutants, including zinc, in the wastewater discharged from manufacturing industries (hereinafter known as an environmental process). The safe disposal of such effluents into aquatic environments is very crucial and of great health and environmental concerns worldwide. To safeguard the quality of the receiving waters, continuous and effective monitoring of zinc in industrial effluents before discharging it into the environment is essential. However, the real-life environmental data are often skewed and do not follow the normal distribution. Therefore, classical statistical process monitoring techniques based on normality assumptions cannot be used directly to monitor the environmental processes. To overcome this limitation, this study designs an optimal X‾ scheme for monitoring non-normal zinc data in industrial effluents. This scheme minimizes the total cost, including quality and manpower costs, based on optimal deployment of manpower. The optimization process ensures that the false alarm rate of the monitoring scheme meets the allowable value. A fractional factorial experiment considering different design specifications shows that the proposed optimal X‾ scheme reduces the associated costs by about 69% and 36% compared to the basic X‾ and improved X‾ schemes, respectively. The robustness of the performance of the optimal X‾ scheme is further investigated under different non-normal, and process shift distributions. A sensitivity analysis is also conducted to explore the impact of the input parameters on the associated cost of the optimal X‾ scheme. Lastly, the design and application of the proposed optimal X‾ scheme are presented using real data on zinc content in the wastewater emitted from a steel production plant.

Real-time monitoring and diagnosis scheme for IoT-enabled devices using multivariate SPC techniques

This article is aimed at condition monitoring and fault identification for Internet of Things (IoT) devices, and proposes a multivariate statistical process control scheme. The new method aims to detect sparse mean shifts using spatial rank and an improved adaptive elastic net algorithm, which can monitor the high-dimension data stream collected by IoT devices and pinpoint faulty variables. The new method is also applicable in the presence of a non-normal distribution and insufficient reference samples. Numerical simulations verify that the proposed method has clear advantages over existing methods. The case of wind turbines shows that the method can be applied to real-time monitoring and diagnosis of real IoT devices, which could provide valuable diagnosis of root cause and optimize subsequent maintenance strategies.

A Data-Driven Method for Online Monitoring Tube Wall Thinning Process in Dynamic Noisy Environment

Tube internal erosion, which corresponds to its wall thinning process, is one of the major safety concerns for tubes. Many sensing technologies have been developed to detect a tube wall thinning process. Among them, fiber Bragg grating (FBG) sensors are the most popular ones due to their precise measurement properties. Most of the current works focus on how to design different types of FBG sensors according to certain physical laws and only test their sensors in controlled laboratory conditions. However, in practice, an industrial system usually suffers from harsh and dynamic environmental conditions, and FBG signals are affected by many unpredictable factors. Consequently, the FBG signals have more fluctuations and are polluted by noises. Hence, the signals no longer directly follow the assumed physical laws and their proposed thinning detection mechanisms no longer work. Targeting at this, this article develops a data-driven model for FBG signal feature extraction and tube wall thickness monitoring using data analytic techniques. In particular, we develop a spatiotemporal model to describe dynamic FBG signals and extract features related to thickness. By taking physical law as guideline, we trace the relationship between the extracted features and the tube wall thickness, based on which we construct an online statistical monitoring scheme for tube wall thinning process. We use both laboratory test and field trial experiment to demonstrate the efficacy and efficiency of the proposed scheme. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —This article is motivated by the real industrial needs of inner erosion detection of tubes in harsh environment. Most of the current research works focus on designing various sensing apparatuses based on fiber Bragg grating (FBG) sensors for nondestructive erosion detection. These apparatuses prove to be able to collect signals reflecting tube wall thickness in static and controllable laboratory environment qualitatively. However, in reality, the industrial environment, which is impacted by many changing factors, is dynamic and uncontrollable. Consequently, the signals collected by these FBG apparatuses would have larger variations that mask the signals related to thickness. Furthermore, current methods have neither mentioned how to process their collected data to capture the unnoticeably slow but accumulative erosion information efficiently nor constructed online monitoring algorithms to detect the tube wall thinning process based on the collected signals quantitatively. Built upon their apparatuses but targeting at their unsolved challenges, we propose a novel data-driven approach for FBG signal analysis that can remove the environmental influence and extract features only related to tube wall thickness, and using the extracted features, we construct a statistical process control scheme to monitor tube wall thickness and detect erosion in real time efficiently.

SPC scheme to monitor surgery duration

AbstractA unified statistical process control (SPC) scheme is developed to monitor work‐process duration of repetitive processes (constant work content), semi‐repetitive (work content somewhat varies between cycles) and memoryless (no typical work content). As a basic paradigm to develop the new scheme, we monitor surgery duration (SD), known to represent durations of all three types of work processes. Recently, a new bi‐variate model for SD was developed and empirically validated. Based on the new model, we derive an approximation to the distribution of SD sample average and use it to construct a two‐stage SPC scheme (for a medically specified subcategory of surgeries). In the first stage, SD distributional shape, as reflected in third and fourth moments, is monitored to determine that no shape‐changing assignable causes are active (e.g., work‐content variation is stable); in the second stage, location and scale parameters are separately monitored via a traditional Shewhart scheme. Using an existent database of surgery times, application of the new scheme is demonstrated, and average‐run‐length properties explored. Extensions (like incorporating co‐variates) and generalizations are addressed.

A SPC strategy for decision making in manufacturing processes

Tapping is an extensively employed manufacturing process by which a multi-teeth tool, known as tap, cuts a mating thread when driven into a hole. When taps are new or slightly worn the process is usually in control and the geometry of the resulting threads on the work piece is correct. But as the tap wear increases the thread geometry deviates progressively from the correct one and eventually the screw threads become unacceptable.The aim of this paper consists on a development of statistical process control strategy for decision making according to data coming from the current signal of the tap spindle for assessing thread quality. It could operate on line and indicates when the tap wear is so critical that, if the process were continued, it would result in unacceptable screw threads. The system would be very cost-effective since the tapping process could be run without any operator intervention.

Monitoring location and scale of multivariate processes subject to a multiplicity of assignable causes

ABSTRACT This paper presents a fully adaptive multivariate statistical process control scheme for processes where multiple assignable causes may occur. The assignable causes affect both the mean vector and the covariance matrix, monitored by a T 2 chart and a control chart based on entropy, respectively. Markov chain theory is employed to model the stochastic operation of the proposed scheme and optimize its economic-statistical operation. A numerical investigation and a sensitivity analysis are conducted in order to evaluate the performance and investigate the robustness of the model. A real example is employed to illustrate the operation of the scheme.

Wafer quality monitoring using spatial Dirichlet process based mixed-effect profile modeling scheme

The objective of this work is to develop a profile-based statistical process control scheme for efficaciously monitoring wafer thickness profiles with non-normality in an industrial wafer slicing process. This is an important research area because the geometric quality of semiconductor wafers in a slicing process has a direct impact on the functional integrity of semiconductor parts and the efficiency of the production. Since non-normality in profiles indicates the existence of inter-cluster variations (i.e., the profiles cannot be represented by a single mean profile with normally-distributed random noise), it deteriorates the effectiveness of many traditional statistical process control methods with normality assumption. To realize the objective of this work, a mixed-effect profile monitoring (MEPM) scheme is proposed. The MEPM scheme adaptively groups profile data into clusters and models the inter-cluster variations, consequently, enabling a robust statistical process control scheme for detecting deviant profile data. Capturing the clustering information of the profile data leads to a deep understanding and an accurate modeling of the spatial data. It is a significant improvement over the current practice of monitoring the geometric product quality by summary quality features (such as total thickness variation) or by profiles neglecting inter-cluster variations. In this paper, the MEPM scheme is tested for detecting the out-of-control wafers from a wafer slicing process. Based on wafer thickness profiles, the MEPM scheme outperforms other benchmark methods and identifies the deviant wafers with low average type II error (missed detection rate) as 0.039. This profile monitoring scheme is extensible to geometrical quality assurance for products in other processes.

Statistical tool to detect small hearing threshold shifts

Objective: The aim of this study is to present a new tool that can be used in the prevention of noise-induced hearing loss. Even in most countries where noise levels are well regulated, many workers are still exposed to high levels of noise and suffer from permanent threshold shifts. It is necessary to develop a new strategy to prevent such damage. Method: A statistical process control (SPC) scheme is presented that is able to detect both large and small hearing threshold shifts. Monte Carlo simulations were used to assess the performance of this hearing monitoring scheme. Different hearing threshold shifts were simulated to evaluate the performance of a variety of hearing development scenarios. Results: It is possible to detect hearing threshold shifts smaller than the standard deviation of the hearing tests performed. This means that permanent hearing threshold shifts smaller than 5 dB can be detected and acted on. Outliers can also be automatically detected and treated, increasing the robustness of the monitoring scheme. Conclusion: The proposed statistical framework can be used as an early warning indicator of noise-induced hearing loss with the aim of improving workers’ safety. Individual counteractions can be implemented, reducing the risk of further damage.

Statistical Process Control Based on Optimum Gages

Classical statistical process control (SPC) by attributes is based on counts of nonconformities. However, process quality has greatly improved with respect to past decades, and the vast majority of samples taken from high‐quality processes do not exhibit defective units. Therefore, control charts by variables are the standard monitoring scheme employed. However, it is still possible to design an effective SPC scheme by attributes for such processes if the sample units are classified into categories such as ‘large’, ‘normal’, or ‘small’ according to limits that are different from the specification limits. Units classified as ‘large’ or ‘small’ will most likely still be conforming (within the specifications), but such a classification allows monitoring the process with attributes charts. In the case of dimensional quality characteristics, gages can be built for this purpose, making inspection quick and easy and reducing the risk of errors. We propose such a control chart, optimize it, compare its performance with the traditional and S charts and with another chart in the literature that is also based in classifying observations of continuous variables through gaging, and present a brief sensitivity analysis of its performance. The new chart is shown to be competitive with the use of –S charts, with the operational advantage of simpler, faster, and less costly inspection. Copyright © 2017 John Wiley &amp; Sons, Ltd.

Large scale adoption of statistical process control (SPC) for volumetric modulated ARC therapy patient-specific quality assurance: A retrospective analysis on 1400 patients

Introduction Statistical Process Control (SPC) is a tool widely used in industrial engineering for monitoring, controlling and, ideally, improving a process through statistical analysis. This technique can be applied to quality assurance (QA) practice in radiotherapy in order to characterize the treatment process variability. Purpose To apply SPC strategy to our routine VMAT pre-treatment verification QA. Materials and methods In the last years, more than 1400 patients were treated with Elekta VMAT at our institution. Plans were re-grouped according to treatment technique and disease sites: (1) 611 high-modulated complex SIB treatments (head-neck, brain, gynecological, ano-rectal; (2) 348 prostate treatments and (3) 445 liver, lung and other metastasis treated with extracranial stereotactic radiotherapy (SBRT). Groups 1 and 2–3 plans were optimized with Oncentra Masterplan (dual-arc) and Ergo++ (single arc) TPS. A total of 4030 dose measurements were performed with the PTW Seven29 array/Octavius phantom, both on coronal and sagittal planes. Doses comparison were evaluated using 3%/3 mm γ -analysis. Two metrics were evaluated: (a) points-percentage with g -value less than one (g%) and (b) mean gamma (gmean). Clinical specifications were: g % >90% and g mean Results γ pass-rate values significantly depend on plan complexity. For g %, CL and LCL were 93.9% and 89.4%, 99.1% and 96.6%, and 99.3% and 97.6%, for group 1, 2 and 3 respectively. For gmean, CL and UCL were 0.416 and 0.585, 0.361 and 0.557, and 0.304 and 0.427, for groups 1, 2 and 3 respectively. In all cases, the control limits are well within the clinical specifications. The Cpl/Cpu capability indices for g % and g mean resulted equal to 0.61 and 1.06 in group 1; 2.59 and 1.12 in group 2; 3.77 and 2.10 in group 3, respectively. g % process for group 1 was not capable at 3%/3 mm, but with 5% – 3 mm specification for g%, CL and LCL resulted 98.3% and 96.2% and Cpl was 2.84. Conclusion SPC is useful to quantifiably demonstrate the QA process conformance to clinical specifications.

SPC scheme to monitor linear predictors embedded in nonlinear profiles

Response Modeling Methodology (RMM) is a general platform to model monotone convex relationships. In this article, RMM is combined with linear regression analysis to model and estimate linear predictors (LPs) embedded in a nonlinear profile. A regression-adjusted statistical process control scheme is then implemented to monitor the LP's residuals. To model and estimate the LP, RMM defines a Taylor series expansion of an unknown response transformation and then use canonical correlation analysis to estimate the LP. A possible hindrance to the implementation of the new scheme is possible occurrence of nonnormal errors (in violation of the linear regression model). Reasons for the occurrence of this phenomenon are explored and remedies offered. The effectiveness of the new scheme is demonstrated for data generated via Monte Carlo simulation. Results from hypothesis testing clearly indicate that the type of the response distribution, its skewness and the sample size, do not affect the effectiveness of the new approach. A detailed implementation routine is expounded, accompanied by a numerical example. When interest is solely focused on the stability of the LP, and the nonlinear profile per se is of little interest, the new general RMM-based statistical process control scheme delivers an effective platform for process monitoring. Copyright © 2015 John Wiley & Sons, Ltd.

Selection of the Most Suitable Statistical Process Control Approach for Short Production Runs: A Decision-Model

DOI: 10.7763/IJIET.2015.V5.521  Abstract—Nowadays, customers are increasingly claiming not only for better quality products at the lowest possible cost, but also demanding customized solutions to satisfy their specific, sometimes unique, needs and wants. Due to this, manufacturing companies are seeking to adopt higher agile production models, such as mass customization strategies. In the quality control field, statistical process control (SPC) methods have been widely used to monitor process performance and detect abnormal situations in its behavior; however, traditional SPC approaches are usually not appropriate for small lot or batch sizes, for the start-up of a process, and for situations where a high variety of mixed products exist. Such situations are within the scope of the so called short production runs. Several SPC schemes have been proposed for short-run environments; all of them have their own advantages, shortcomings, and more suitable for certain production scenarios. This paper provides an up-to-date literature review on the topic, identifies classes of SPC short-run methods, and presents a decision-model that guides production managers in the choice of the most appropriate SPC short-run approach. The model was validated in a textile production company, and is being incorporated into a software package.

Bivariate quality control using two-stage intelligent monitoring scheme

In manufacturing industries, it is well known that process variation is a major source of poor quality products. As such, monitoring and diagnosis of variation is essential towards continuous quality improvement. This becomes more challenging when involving two correlated variables (bivariate), whereby selection of statistical process control (SPC) scheme becomes more critical. Nevertheless, the existing traditional SPC schemes for bivariate quality control (BQC) were mainly designed for rapid detection of unnatural variation with limited capability in avoiding false alarm, that is, imbalanced monitoring performance. Another issue is the difficulty in identifying the source of unnatural variation, that is, lack of diagnosis, especially when dealing with small shifts. In this research, a scheme to address balanced monitoring and accurate diagnosis was investigated. Design consideration involved extensive simulation experiments to select input representation based on raw data and statistical features, artificial neural network recognizer design based on synergistic model, and monitoring–diagnosis approach based on two-stage technique. The study focused on bivariate process for cross correlation function, ρ=0.1–0.9 and mean shifts, μ=±0.75–3.00 standard deviations. The proposed two-stage intelligent monitoring scheme (2S-IMS) gave superior performance, namely, average run length, ARL1=3.18–16.75 (for out-of-control process), ARL0=335.01–543.93 (for in-control process) and recognition accuracy, RA=89.5–98.5%. This scheme was validated in manufacturing of audio video device component. This research has provided a new perspective in realizing balanced monitoring and accurate diagnosis in BQC.

Uneven batch data alignment with application to the control of batch end-product quality

Batch processes are commonly characterized by uneven trajectories due to the existence of batch-to-batch variations. The batch end-product quality is usually measured at the end of these uneven trajectories. It is necessary to align the time differences for both the measured trajectories and the batch end-product quality in order to implement statistical process monitoring and control schemes. Apart from synchronizing trajectories with variable lengths using an indicator variable or dynamic time warping, this paper proposes a novel approach to align uneven batch data by identifying short-window PCA&PLS models at first and then applying these identified models to extend shorter trajectories and predict future batch end-product quality. Furthermore, uneven batch data can also be aligned to be a specified batch length using moving window estimation. The proposed approach and its application to the control of batch end-product quality are demonstrated with a simulated example of fed-batch fermentation for penicillin production.

The critical role of NIR spectroscopy and statistical process control (SPC) strategy towards captopril tablets (25 mg) manufacturing process understanding: a case study

In this work, near-infrared spectroscopy (NIRS) method was used to evaluate the uniformity of dosage units of three captopril 25 mg tablets commercial batches. The performance of the calibration method was assessed by determination of Q value (0.9986), standard error of estimation (C-set SEE = 1.956), standard error of prediction (V-set SEP = 2.076) as well as the consistency (106.1%). These results indicated the adequacy of the selected model. The method validation revealed the agreement of the reference high pressure liquid chromatography (HPLC) and NIRS methods. The process evaluation using the NIRS method showed that the variability was due to common causes and delivered predictable results consistently. Cp and Cpk values were, respectively, 2.05 and 1.80. These results revealed a non-centered process in relation to the average target (100% w/w), in the specified range (85–115%). The probability of failure was 21:100 million tablets of captopril. The NIRS in combination with the method of multivariate calibration, partial least squares (PLS) regression, allowed the development of methodology for the uniformity of dosage units evaluation of captopril tablets 25 mg. The statistical process control strategy associated with NIRS method as PAT played a critical role in understanding of the sources and degree of variation and its impact on the process. This approach led towards a better process understanding and provided the sound scientific basis for its continuous improvement.

Moving/motionless foreground object detection using fast statistical background updating

In video surveillance, the detection of foreground objects in an image sequence from a still camera is very important for object tracking, activity recognition and behaviour understanding. The conventional background subtraction cannot respond promptly to dynamic changes in the background, and temporal difference cannot accurately extract the object shapes and detect motionless objects. In this paper, we propose a fast statistical process control scheme for foreground segmentation. The proposed method can promptly calculate the exact grey-level mean and standard deviation of individual pixels in both short- and long-term image sequences by simply deleting the earliest one among the set of images and adding the current image scene in the image sequence. A short-term updating process can be highly responsive to dynamic changes of the environment, and a long-term updating process can well extract the shape of a moving object. The detection results from both the short- and long-term processes are incorporated to detect motionless objects and eliminate non-stationary background objects. Experimental results have shown that the proposed scheme can be well applied to both indoor and outdoor environments. It can effectively extract foreground objects with various moving speeds or without motion at a high process frame rate.