An Fuzzy Scaled Weighted Variance S Control Chart for Skewed Populations

Kanittha Yimnak, Rungsarit Intaramo

Abstract

A control chart is one of the most important tools in statistical process control (SPC) that leads to improving quality of processes and ensures the required quality levels. The usual assumption for designing a control chart is the data or measurement should have a normal distribution. However, this assumption may not be true for some processes, there are some factors that cause an uncertainty data, such as human, measurement device or environmental conditions. Therefore, the purposes of this paper are to study a new S control chart for monitoring process dispersion of skewed populations, which has a non-normal distribution as Weibull, gamma and lognormal. This control chart, called fuzzy scaled weighted variance S control chart (FSWV-S), that is an improvement of the scaled weighted variance S control chart (SWV-S) proposed Atta et al [17].  Finally, the efficiency of the FSWV S control charts is compared with the average run length ( ) by Monte Carlo (MC) simulation technique.