Search

Multivariate autoregressive (MAR) models are an attractive choice for applications in the processes related to finance, medical, and industry. For the monitoring of such processes, control chart is the most important and widely used tool of statistical process control tool kit. Moreover, the presence of auxiliary information helps in better estimation of different process parameters. The literature on use of auxiliary variables in control charts assumes independence of observations. In practice, we may come across processes dealing with autocorrelated outcomes. In such situations, a control chart usually produces high false alarms and exhibits slow detection of shifts when the process is out-of-control. This study intends to suggest some auxiliary information-based Shewhart charts for autocorrelated univariate and bivariate AR(1) processes. The proposed structures take into account the autocorrelation structure and offer more effective designs of control charts for efficient process monitoring. The performance measures used in this study are based on run length measures such as average run length, extra quadratic loss, relative average run length and performance comparison index. A detailed performance analysis is carried out to sort out the best performing charts. In addition, we have considered an application from a manufacturing process to demonstrate the implementation of the proposed charting structures in real scenario. [ABSTRACT FROM AUTHOR]

In many industrial processes, the quality characteristic of interest has a relation (linear or non-linear) with other supporting variable(s). Simple linear profile is a well-known term used for the quality characteristic, which is linearly associated with another descriptive variable, and the monitoring of simple linear profile parameters (i.e., slope, intercept, and error variance) is known as linear profiling. In the literature, a well-known approach named as EWMA_3 chart is used for the simultaneous monitoring of intercept, slope, and error variance. This approach is very efficient as compared to EWMA/R, Hotelling T2, and Shewhart_3 charts but it is a tedious method, since distinct pair of control limits require individual charting constant for each parameter. In this study, new methods are designed for the simultaneous monitoring of simple linear profile parameters, which requires single charting constant and have several advantages such as simplicity, efficiency, and ease of applicability. The findings of this study reveal that newly designed control charts such as the Max − EWMA − 3 and SS − EWMA − 3 have almost similar performance with EWMA_3 chart. Specifically, Max − EWMA − 3 − C chart shows superiority among all other control charts. Further, importance of the stated proposals is highlighted by the real example from the field of chemical engineering. [ABSTRACT FROM AUTHOR]

In many manufacturing processes, study variable is not the only quality characteristic, but there may exist some explanatory variable(s) that are associated with the study variable. This association may be linear or nonlinear depending on the nature of variables. The term profiling is used for such relationships among study and explanatory variables. Linear profiles are more common options because of their simplicity and coverage of more common scenarios. A popular proposal for the monitoring of linear profiles is EWMA − 3 chart that detect shifts in the profile parameters including slope, intercept, and error variance. The said chart is designed under simple random sampling. In this study, we have designed and investigated EWMA − 3 chart under the different ranked set sampling strategies ( τ) such as ranked set sampling (RSS), median ranked set sampling (MRSS), extreme ranked set sampling (ERSS), double ranked set sampling (DRSS), double median ranked set sampling (DMRSS), and double extreme ranked set sampling (DERSS). We have used average run length (ARL), extra quadratic loss (EQL), relative average run length (RARL), and performance comparison index (PCI) as performance measures for the aforementioned designs of EWMA − 3 chart under different sampling schemes. The computational results of run length properties revealed that the ranked set based EWMA − 3 chart offers better detection ability. More specifically, EWMA − 3 and EWMA − 3 serve the purpose in a more efficient way. [ABSTRACT FROM AUTHOR]

Exponentially weighted moving average (EWMA) and cumulative sum (CUSUM) charts are used to detect smaller shifts in process parameters. The usual EWMA and CUSUM charts depend on the normality assumption for a better detection ability. This study proposes an efficient EWMA control chart based on the spirit of Tukey control chart, especially designed for skewed distributions. The performance of the proposed and the competing charts is measured using different length properties such as average run length (ARL), standard deviation of run length (SDRL), and median run length (MDRL). We have observed that the proposed chart is quite efficient at detecting process shifts of smaller magnitude, especially for skewed distributions. For practical considerations, the proposed chart is implemented at aerospace manufacturing data on industrial production index. [ABSTRACT FROM AUTHOR]

Control charts are popular tools in statistical process control (SPC) and artificial neural network (ANN) technique is an attractive alternative for efficient monitoring of process parameters. This study uses the artificial neural network technique with back propagation method to process control system for dispersion parameter. We have trained an artificial neural network to be used in statistical control charts using varying runs rules schemes. By investigating the performance of trained artificial neural network under normal and bootstrapping environments we have made comparisons of the usual ANN and three runs rules-based schemes for ANN to gain the precision of process. We have used average run length (ARL), extra quadratic loss (EQL), relative ARL (RARL), and performance comparison index (PCI) measures and explored the said structures of trained artificial neural network under bootstrapping by implementing runs rules schemes. We have also suggested a modification in the trained ANN for variance change detection. An example with real data is also given for practical considerations. [ABSTRACT FROM AUTHOR]

Control charts act as the most effective statistical process control (SPC) tools for the monitoring of manufacturing processes. In this study, we propose and investigate a set Shewhart-type variability control chart based on the utilization of auxiliary information for efficient phase II process monitoring. The design parameters of the proposals are derived under correlated setups for the monitoring of variability parameter. The properties of these charting structures are evaluated in terms of average run length and some other related measures. The performance abilities of these charts are compared with each other and also with some existing counterparts. The comparisons revealed that the proposed charts are very efficient at detecting shifts in the variability parameter and have the ability to perform better than the competing charts in terms of run length characteristics. We have also used real datasets to illustrate the application of the proposed structures in practical situations. [ABSTRACT FROM AUTHOR]

Statistical process control (SPC) has its own importance in the field of quality control. In SPC, control charts are significant tools to monitor process parameters, and exponentially weighted moving average (EWMA) control chart is one such tool. It is a memory-type chart, which is used to target mainly the smaller shifts in the process parameters. Adaptive EWMA (AEWMA) scheme is used to identify small as well as large shifts. EWMA and AEWMA are based on the assumption of normality, which is quite hard to find in practice, and there are many situations where outliers are occasionally present. In the current study, we have proposed four robust adaptive EWMA schemes for monitoring process location parameter. We have investigated their performance under uncontaminated normal and contaminated normal environments. We have carried out comparisons amongst different competing charts based on average run length (ARL), standard deviation of run length (SDRL) and different percentiles of run length distribution. Two examples related to manufacturing processes are also provided for practical implementation of the proposed schemes. [ABSTRACT FROM AUTHOR]

The ordinary CUSUM chart is based on normality assumption. But in real-life phenomenon such as monitoring of lifetime variable, this assumption is not always valid. Hence, a variant of the CUSUM chart, the VCUSUM chart, has been constructed to monitor small shift in a process that is based on a Maxwell distribution. The performance of the chart has been checked by studying the run length properties and it was found to perform better for small shifts as compared to Shewhart type V chart for Maxwell parameter. Finally, a real-life application is presented to monitor the failure rate of a vertical boring machine. [ABSTRACT FROM AUTHOR]