The presence of asymmetric data in production processes or service operations has prompted the development of new monitoring schemes. In this article, an adapted version of the exponentially weighted moving averages (EWMA) control chart with dynamic limits is proposed to monitor the mean of samples from the skew-normal distribution. The detection ability of the proposed control chart in online monitoring was investigated by simulating the average run length (ARL) performance for different out-of-control scenarios. The results of the simulation study suggest that the proposed scheme overcomes the main drawback of the recently developed Shewhart-type control scheme. As shown in this article, the existing Shewhart-type procedure exhibits the undesirable property of taking longer to detect changes in the mean value of skewed normal observations due to increases in the shape parameter of the basic distribution than in stable conditions. The proposed control chart was shown to work fairly acceptably in all considered out-of-control scenarios.
4 November 2018 in Communications in Statistics - Simulation and Computation
In this paper a methodology for monitoring the slope of the linear specification for the Weibull regression model (WRM) scale parameter is presented. It is assumed that the common shape parameter γ of the model is known and a right-censoring mechanism is operating. The monitoring is completely possible by designing a specific likelihood ratio based cumulative sum (CUSUM) chart for the Weibull-distributed response. It was found out that the performance of the resulting control schemes depends on the known value of the shape parameter of the basic model, the dataset dimension, the theoretical censoring rate, the desired shift to be detected and the fixed covariate pattern. Although the monitoring is focused on detecting decreases in the slope that lead to unwanted decreases in the mean response, it is also possible to use these charts for detecting increases in the slope or shifts (increases or decreases) in the intercept of the model as well. Chart performance was compared with that of an adapted version of the multivariate exponentially weighted moving average (MEWMA) method. It was found out that the proposed schemes outperform the MEWMA chart in most of the studied cases.
28 June 2017 in Journal of Statistical Computation and Simulation
In most of the existing specialized literature, monitoring regression models are a special case of profile monitoring. However, not every regression model always represents appropriately a profile data structure. This is clearly the case of the Weibull regression model (WRM) with common shape parameter γ. Even though it might be thought that existing methodologies (especially likelihood-ratio (LRT)-based methods) for monitoring generalized linear profiles can also be successfully applied to monitoring regression models with time-to-event response, it will be shown in this paper that those methodologies work fairly acceptable just for data structures with 1000 observations at least approximately. It was found out that some corrections, often referred to as Bartlett's adjustments, are needed to be implemented in order to improve the accuracy of using the asymptotic distributional properties of the LRT statistic for carrying out the monitoring of WRM with relatively small and moderate dimensions of the available datasets. Simulation studies suggest that the use of the aforementioned corrections make the resulting charts work quite acceptable when available data structures contain 30 observations at least. Detection abilities of the proposed schemes improve as dataset dimension increases.
This paper provides a comprehensive review of the Screening Robust Estimation Method (SREM) for normal processes in Phase I. Particular emphasis is placed on the exponentially weighted moving average (EWMA) control chart, which is employed in the retrospective monitoring stage. A central concern addressed in this discussion is that the EWMA chart is said to be based on the false alarm rate (FAR) design criterion, but it is accurately implemented. It is well established that FAR-based methods for Phase I monitoring assume that the distribution of the monitoring statistic remains invariant at each sampling instance. However, the EWMA statistic inherently introduces a sequential dependence among all sampling moments, which contradicts this assumption. This paper shows that a modified EWMA chart, incorporating probability-based control limits rather than the conventional formulation utilized in the SREM, enhances the monitoring of normal processes in Phase I. Through simulations, it is established that the new design proposal of an EWMA chart for Phase I monitoring has a slightly narrower decision threshold among the control methodologies included in the study. The new chart is as effective as the traditional charts in detecting localized increases in the target value of the mean of a normal process, and outperforms them when other types of anomalies are present in the available preliminary samples.