Preparatory Year Mathematics Program, King Fahd University of Petroleum and Minerals, Dhahran 31261, Saudi Arabia
Department of Mathematical Sciences, Faculty of Science, Universiti Teknologi Malaysia, 81310 UTM Skudai, Johor, Malaysia
Department of Mathematics and Statistics, King Fahd University of Petroleum and Minerals, Dhahran 31261, Saudi Arabia
The exponentially weighted moving average (EWMA) control chart is an effective tool for the detection of small shifts in the process variability. This research studied the properties of EWMA charts based on unbiased sample variance for monitoring of changes in the process dispersion. However, since an increase in process variance could lead to an increased number of defective products, we consider only upward shifts in the process variance. The proposed schemes are based on simple random sampling and extreme variations of ranked set sampling technique for efficient monitoring. Using Monte Carlo simulations, we compare the relative performance of EWMA charts based on unbiased sample variance and its logarithmic transformation as well as some existing schemes for monitoring increases in variability of a normal process. It is found that the proposed schemes significantly outperform several other procedures for detecting increases in the process dispersion. Numerical example is given to illustrate the practical application of the proposed schemes using real industrial data.