No abstract

Exponentially weighted moving average (EWMA) control charts are mostly used to monitor the manufacturing processes. In this paper, we propose some improved EWMA control charts for detecting the random shifts in the process mean and process dispersion. These EWMA control charts are based on the best linear unbiased estimators obtained under ordered ranked set sampling (ORSS) and ordered imperfect ranked set sampling (OIRSS), named EWMA‐ORSS and EWMA‐OIRSS charts, respectively. Monte Carlo simulations are used to estimate the average run length, median run length and standard deviation of run length of the proposed EWMA control charts. It is observed that the EWMA‐ORSS mean control chart is able to detect the random shifts in the process mean substantially quicker than the Shewhart‐cumulative sum and the Shewhart‐EWMA control charts based on the RSS scheme. Both EWMA‐ORSS and EWMA‐OIRSS location charts perform better than the classical EWMA, hybrid EWMA, Shewhart‐EWMA and fast initial response‐EWMA charts. The EWMA‐ORSS dispersion control chart performs better than the simple random sampling based CS‐EWMA and other EWMA control charts in efficient detection of the random shifts that occur in the process variability. An application to real data is also given to explain the implementation of the proposed EWMA control charts. Copyright © 2013 John Wiley & Sons, Ltd.

No abstract

No abstract

No abstract

No abstract

We modify RSS to come up with new sampling method, namely, Multistage Median Ranked Set Sampling (MMRSS). The MMRSS was suggested for estimating the population median and to increase the efficiency of the estimator for specific value of the sample size. The MMRSS was compared to the Simple Random Sampling (SRS), Ranked Set Sampling (RSS) and Median Ranked Set Sampling (MRSS) methods. It is found that MMRSS gives an unbiased estimate of the population median of symmetric distributions and it is more efficient than SRS, RSS and MRSS based on the same number of measured units. Also, it was found that the efficiency of MMRSS increases in r (r is the number of stage) for specific value of the sample size. For asymmetric distributions considered in this study, MMRSS has a small bias, close to zero as r increases, especially with odd sample size. A set of real data was used to illustrate the method

No abstract