Monitoring Process Variability Research Articles

Remedial Measures to Lessen the Effect of Imprecise Measurement with Linearly Increasing Variance on the Performance of the MAX-EWMAMS Scheme

Simultaneous monitoring of process mean and variability by a single control chart has been increasingly taken into consideration in recent years. However, the effect of imprecise measurements on the performances of some existing control schemes has been neglected. In this paper, the effect of measurement errors with linearly increasing variance on the detecting and diagnosing capability of the MAX-EWMAMS control chart is first investigated in Phase II monitoring. The results obtained using simulation studies show that the measurement errors affect the two performances of the chart, significantly. Then, two remedial measures including the ranked set sampling approach and using a larger sample size are proposed. The simulation results confirm that both the remedial measures compensate for the effect of measurement errors, adequately. A real-data example is also given to illustrate the effect of measurement errors on the rate of false alarm.

Integrated Process Modeling-A Process Validation Life Cycle Companion.

During the regulatory requested process validation of pharmaceutical manufacturing processes, companies aim to identify, control, and continuously monitor process variation and its impact on critical quality attributes (CQAs) of the final product. It is difficult to directly connect the impact of single process parameters (PPs) to final product CQAs, especially in biopharmaceutical process development and production, where multiple unit operations are stacked together and interact with each other. Therefore, we want to present the application of Monte Carlo (MC) simulation using an integrated process model (IPM) that enables estimation of process capability even in early stages of process validation. Once the IPM is established, its capability in risk and criticality assessment is furthermore demonstrated. IPMs can be used to enable holistic production control strategies that take interactions of process parameters of multiple unit operations into account. Moreover, IPMs can be trained with development data, refined with qualification runs, and maintained with routine manufacturing data which underlines the lifecycle concept. These applications will be shown by means of a process characterization study recently conducted at a world-leading contract manufacturing organization (CMO). The new IPM methodology therefore allows anticipation of out of specification (OOS) events, identify critical process parameters, and take risk-based decisions on counteractions that increase process robustness and decrease the likelihood of OOS events.

A control chart for monitoring process variation using multiple dependent state sampling

ABSTRACTA control chart for monitoring process variation by using multiple dependent state (MDS) sampling is constructed in the present article. The operational formulas for in-control and out-of-control average run lengths (ARLs) are derived. Control constants are established by considering the target in-control ARL at a normal process. The extensive ARL tables are reported for various parameters and shifted values of process parameters. The performance of the proposed control chart has been evaluated with several existing charts in regard of ARLs, which empowered the presented chart and proved far better for timely detection of assignable causes. The application of the proposed concept is illustrated with a real-life industrial example and a simulation-based study to elaborate strength of the proposed chart over the existing concepts.

New cumulative sum control charts for monitoring process variability

ABSTRACTIn this article, we propose new cumulative sum (CUSUM) control charts using the ordered ranked set sampling (RSS) and ordered double RSS schemes, with the perfect and imperfect rankings, for monitoring the variability of a normally distributed process. The run length characteristics of the proposed CUSUM charts are computed using the Monte Carlo simulations. The proposed CUSUM charts are compared in terms of the average and standard deviation of run lengths with their existing competitor CUSUM charts based on simple random sampling. It turns out that the proposed CUSUM charts with the perfect and imperfect rankings are more sensitive than the existing CUSUM charts based on the sample range and standard deviation. A similar trend is present when these CUSUM charts are compared with the fast initial response features. An example is also used to demonstrate the implementation and working of the proposed CUSUM charts.

A double sampling scheme for process variability monitoring

AbstractControl charts are effective tools for signal detection in both manufacturing processes and service processes. Much of the data in service industries come from processes exhibiting nonnormal or unknown distributions. The commonly used Shewhart variable control charts, which depend heavily on the normality assumption, are not appropriately used here. This paper thus proposes a standardized asymmetric exponentially weighted moving average (EWMA) variance chart with a double sampling scheme (SDS EWMA‐AV chart) for monitoring process variability. We further explore the sampling properties of the new monitoring statistics and calculate the average run lengths when using the proposed SDS EWMA‐AV chart. The performance of the SDS EWMA‐AV chart and that of the single sampling EWMA variance (SS EWMA‐V) chart are then compared, with the former showing superior out‐of‐control detection performance versus the latter. We also compare the out‐of‐control variance detection performance of the proposed chart with those of nonparametric variance charts, the nonparametric Mood variance chart (NP‐M chart) with runs rules, and the nonparametric likelihood ratio‐based distribution‐free EWMA (NLE) chart and the combination of traditional EWMA (CEW) and the SS EWMA‐V control charts by considering cases in which the critical quality characteristic presents normal, double exponential, uniform, chi‐square, and exponential distributions. Comparison results show that the proposed chart always outperforms the NP‐M with runs rules, the NLE, CEW, and the SS EWMA‐V control charts. We hence recommend employing the SDS EWMA‐AV chart. Finally, a numerical example of a service system for a bank branch in Taiwan is used to illustrate the application of the proposed variability control chart.

Between the Poles of Data‐Driven and Mechanistic Modeling for Process Operation

AbstractThe best method for process control is the use of model‐based solutions, based on process analytical technology for online monitoring of critical process variables, product quality attributes, or a holistic process state estimation. Mechanistic models as well as data‐driven techniques are essential for real‐time process monitoring. Their main characteristics, advantages and disadvantages, and the link between both are discussed as well as the synergetic effects, benefits, and drawbacks resulting from their combination. Aspects and differences of the computational model life cycle management are highlighted.

Exponentially weighted control charts to monitor multivariate process variability for high dimensions

Multivariate monitoring of industrial or clinical procedures often involves more than three correlated quality characteristics and the status of the process is judged using a sample of size one. Majority of existing control charts for monitoring process variability for individual observations are capable of monitoring up to three characteristics. One of the hurdles in designing optimal control charts for large dimension data is the enormous computing resources and time that is required by simulation algorithm to estimate the charts parameters. This paper proposes a novel algorithm based on Parallelised Monte Carlo simulation to improve the ability of the Multivariate Exponentially Weighted Mean Squared Deviation and Multivariate Exponentially Weighted Moving Variance charts to monitor process variability for high dimensions in a computationally efficient way. Different techniques have been deployed to reduce computing space and execution time. The optimal control limits (L) to detect small, medium and large shifts in the covariance matrix of up to 15 characteristics are provided. Furthermore, utilising the large number of optimal L values generated by the algorithm enabled authors to develop exponential decay functions to predict L values. This eliminates the need for further execution of the parallelised Monte Carlo simulation.

Monitoring Pump Parameters in Small Modular Reactors Using Electric Motor Signatures

Abstract Small modular nuclear reactors (SMRs) are designed for long-term operation with minimum outages and for possible deployment in remote locations. To achieve this operational goal, the SMRs may require remote and continuous monitoring of performance parameters that contribute to operation and maintenance. This feature is also important in monitoring critical parameters during severe accidents and for postaccident recovery. Small integral light water reactors have in-vessel space constraints, and many of the traditional instrumentation are not practical in these systems. To investigate this issue, analytical and experimental researches were carried out using a flow test loop to characterize the relationship among process variables (flow rate and pressure) and pump motor signatures. The findings of this research are presented, with implications in relating electrical signatures to pump parameters. The relationship between the electrical signatures and the process variables is discussed with reference to the experimental results. The results of this work may be used for monitoring process variables in small modular reactor systems.

Monitoring process variability: a hybrid Taguchi loss and multiobjective genetic algorithm approach

The common consideration on economic model is that there is knowledge about the risk of occurrence of an assignable cause and the various cost parameters that does not always adequately describe what happens in practice. Hence, there is a need for more realistic assumptions to be incorporated. In order to reduce cost penalties for not knowing the true values of some parameters, this paper aims to develop a bi-objective model of the economic-statistical design of the S control chart to minimize the mean hourly loss cost while minimizing out-of-control average run length and maintaining reasonable in-control average run length considering Taguchi loss function. The purpose of Taguchi loss function is to reflect the economic loss associated with variation in, and deviations from, the process target or the target value of a product characteristic. In contrast to the existing modeling approaches, the proposed model and given Pareto-optimal solution sets enables the chart designer to obtain solutions that is effective even for control chart design problems in uncertain environments. A comparison study with a traditional economic design model reveals that the proposed chart presents a better approach for quality system costs and the power of control chart in detecting the assignable cause.

Resistive RAM variability monitoring using a ring oscillator based test chip

Common problems with Oxide-based Resistive Random Access Memory (so-called OxRRAM) are related to high variability in operating conditions and low yield. Although research has taken steps to resolve these issues, variability remains an important characteristic for OxRRAMs. In this paper, a test structure consisting of an OxRRAM matrix where each memory cell can be configured as a ring oscillator is introduced. The oscillation frequency of each memory cell is function of the cell resistance. Thus, the test structure provides within-die accurate information regarding OxRRAM cells variability. The test structure can be used as a powerful tool for process variability monitoring during a new process technology introduction but also for marginal cells detection during process maturity.

Present Chemical Production Simulator Model Validation Method

The developed new validation method of the simulator model in standard and emergency situations for chemical production management skills training at minimal amount of monitored process variables is presented by the example of fluorine electrochemical production process. The method consists of two stages. At the first stage the validation quality of separate process variables is evaluated. It allows verifying the developed model accuracy and identification of its inputs or indicated the necessity to correct some development phase. At the second stage the model operation and its applicability for training purposes is generally evaluated. The simulation quality of the monitored process variables which enable cause-effect analysis to detect emergency reasons is checked.

Monitoring multivariate process variability when sub-group size is small

ABSTRACTThe current practice of multivariate process variability monitoring, when sub-group size is small, has nothing to do with probability of false alarm (PFA). Consequently, the reliability of the existing control charts remains undetermined. In this article, we propose a control chart which is reliable, very sensitive to the change in variance for small or moderate correlation, and provides a root causes analysis of an out-of-control signal. To illustrate these advantages, an industrial example is presented and the results are compared with those issued from the existing methods.

Combining attribute and variable data to monitor process variability: MIX S 2 control chart

The necessity to monitor process variability to maintain similar, but more economical, performance to that of the S2 control chart in terms of ARL1 is the motivation of this paper. In general, a sample of four to six units is used to build an S2 control chart. Reducing the sample size to two or three units is a possibility, but the performance will be poor (in terms of ARL1), and the project specification may not be met. Additional cost would be required to remove the nonconforming units before shipment to the customers. This would compromise the quality of products in an environment of high worldwide competitiveness. The aim of this paper is to present the MIX S2 control chart, which employs a two-phase control chart to monitor the variability. At every interval of h hours, each unit is sequentially classified as approved or rejected by employing a Go/No Go ring gauge. A unit is approved if its dimension lies in [LDL; UDL], where L(U)DL is the lower(upper) discriminant limit; otherwise, it is rejected. If A1-approved items are first observed, then the production continues. However, if B1-rejected items are first observed, then n (n < 5) complementary units are taken, and the quality characteristics are measured to calculate the sample variance S2. If S2 is greater than the control limit, then the process is stopped for adjustment; otherwise, the production continues. The MIX S2 chart employs mixed control charts (attribute and variable) with the aim to find optimized parameters—LDL, UDL, A1, B1, and CL—such that similar values to ARL1 (and average sample size) are provided at a lower cost than the S2 control chart.

Effect of Measurement Error on Joint Monitoring of Process Mean and Variability under Ranked Set Sampling

In this paper first we apply ranked set sampling (RSS) procedure in joint monitoring of both process mean and variance. Then, we study the effect of measurement error on joint monitoring of process mean and variance when simple random sampling (SRS) as well as RSS procedure is used. The results prove that the measurement error can seriously deteriorate the ability of the control chart in detecting all out‐of‐control scenarios. The results also represent that using RSS procedure can improve the ability of the monitoring scheme in detecting mean shifts, variance shifts as well as joint shifts either measurement error exists or not. In other words, RSS procedure can reduce the adverse effect of measurement error on detecting ability of joint monitoring scheme. After that, we utilize multiple measurements at each sample point when both SRS and RSS procedures are used. We investigate the effect of parameters in the covariate model thorough a sensitivity analysis. Finally, the applicability of the proposed control charts is illustrated using a real case of the piston rings in an automotive engine manufactured by a forging process. Copyright © 2016 John Wiley &amp; Sons, Ltd.

Simultaneous Monitoring of Multivariate Process Mean and Variability in the Presence of Measurement Error with Linearly Increasing Variance under Additive Covariate Model

Simultaneous Monitoring of Multivariate Process Mean and Variability in the Presence of Measurement Error with Linearly Increasing Variance under Additive Covariate Model

A moving average S control chart for monitoring process variability

ABSTRACTAn efficient alternative to the S control chart for detecting shifts of small magnitude in the process variability using a moving average based on the sample standard deviation s statistic is proposed. Control limit factors are derived for the chart for different values of sample size and span w. The performance of the moving average S chart is compared to the S chart in terms of average run length. The result shows that the performance of moving average S chart for varying values of w outweigh those of the S chart for small and moderate shifts in process variability.

A New Nonparametric Control Chart for Monitoring Variability

Statistical process control is widely used in industrial processes, service fields, among others. While parametric control charts are useful in certain processes, there is often a lack of enough knowledge about the process distribution. So, nonparametric control charts are needed in such situations. This paper develops a new nonparametric control chart based on the Ansari–Bradley nonparametric test and the effective change point model. Simulation results show that our proposed control chart is superior to other nonparametric control charts in monitoring process variability for most cases. Our proposed control chart is easy in computation, and powerful for monitoring process variability. Copyright © 2016 John Wiley &amp; Sons, Ltd.

Run sum control charts for the monitoring of process variability

A two-sided run sum S control chart is proposed and its average run length performance is evaluated via a Markov chain technique. The performance of the chart is compared to several well-known control-charting procedures for the monitoring of process variability. One-sided counterparts of the proposed run sum charts are also discussed. The numerical results demonstrate an improved performance of the run sum S control charts as compared with control charts with runs rules, especially in the detection of increasing shifts in process variability. A practical guidance for the selection of the appropriate charting procedure is also given.

Multivariate exponentially weighted moving sample covariance control chart for monitoring covariance matrix

In this paper, a control chart is proposed to detect changes in the covariance matrix of a multivariate normal process, when sample size is one. The proposed chart statistic is constructed based on the exponentially weighted form of sample covariance matrix given by individual observation over time. Distance between the values of variance and covariance components in this multivariate exponentially weighted moving sample covariance matrix and, the in-control corresponding elements of process variance-covariance matrix provides a basis for process variability monitoring. The statistical performance of the proposed method is evaluated through the use of a Monte Carlo simulation. The results show the superiority of the proposed control chart performance especially in the case of incremental changes in covariance matrix.

Evaluation of selective laser sintering processes by optical coherence tomography

Selective laser sintering (SLS) enables the fast, flexible and cost-efficient production of parts directly from 3D CAD data. Unlike more established machine tools, there is a marked lack of process monitoring and feedback control of key process variables. In-situ analysis techniques permit the emergence of repair techniques, in-process optimization of production parameters, and will also serve to save time and material. In this study, optical coherence tomography (OCT) is used for the first time to evaluate components produced by SLS. Using a Polyamide-PA2200, surface defects are analyzed and the limiting factors associated with the measurement technique are quantified. OCT is shown to be a useful technique for evaluating surface irregularities alongside sub-surface defects that have resulted from poor sintering or non-homogeneous powder spreading. We demonstrate detection and quantification of surface defects such as cracks, pores and voids on a ~30μm scale. Furthermore, we show that this technique can resolve ‘built-in’ fine features within a 200 to 400μm depth below the surface, covering typical layer thicknesses used by this process. This capability paves the way for real-time monitoring of the SLS process for assurance, or even dynamic correction of defects during the build.