Enhancing R control chart performance in monitoring process dispersion using scaled weighted variance method for skewed populations

• Main Authors: Atta, Abd, Shoraim, Majed, Syed Yahaya, Sharipah Soaad, Abuzaid, Ali, Mahdi, Esam
• Format: Article
• Language: English
• Published: Medwell Publishing 2020
• Subjects: QA75 Electronic computers. Computer science
• Online Access: http://repo.uum.edu.my/27546/1/JEAS%2015%206%202020%201508-1514.pdf; http://repo.uum.edu.my/27546/; http://doi.org/10.36478/jeasci.2020.1508.1514

This study improves the performance of R control chart for monitoring process dispersion of skewed populations using scaled weighted variance method. This control chart, called Scaled Weighted Variance R control chart (SWV-R) hereafter, the SWV-R control chart compared with Skewness Correction R chart (SC-R) and Weighted Variance R chart (WV-R) in terms of false alarm. In terms of probability of detection rates the proposed SWV-R chart is compared with R chart of the exact method, SC-R and WV-R control charts. The proposed SWV-R control chart reduces to the Shewhart R control chart when the underlying distribution is symmetric. An illustrative example is given to show how the proposed SWV-R control chart is constructed and works simulations study show that the proposed SWV-R control chart has the lower false alarm rates than the SC-R and WV-R control charts, when the underlying distributions are Weibull and gamma. In terms of the probability of detection rates, the proposed SWV-R control chart is closer to R control chart with the exact method than WV-R and almost the same performance as SC-R chart.