New Single Variables Control Charts Based On The Double Ewma Statistics

• Main Author: Teh, Sin Yin
• Format: Thesis
• Language: English
• Published: 2012
• Subjects: HD28-70 Management. Industrial Management
• Online Access: http://eprints.usm.my/46264/1/TEH%20SIN%20YIN%20PhD%20Thesis24.pdf; http://eprints.usm.my/46264/
• Institution: Universiti Sains Malaysia

In Statistical Process Control (SPC) monitoring situations, there is a tendency for both the process mean and process variability to shift simultaneously. Traditionally, two separate control charts, each for the mean and variance are used concurrently to monitor the process mean and process variance. However, in many real life process monitoring situations, a simultaneous control of the process mean and process variance is necessary. This has motivated us to develop single DEWMA (called Double Exponentially Weighted Moving Average) charts which are capable of monitoring simultaneous shifts in both the process mean and process variance, when the underlying distribution of the process is normal. The DEWMA statistics are based on the approach of performing exponential smoothing twice on the original statistics of the underlying process. The objective of this study is to propose three single DEWMA charts, namely the DEWMA-Max (called the DEWMA maximum), Max-DEWMA (called the maximum DEWMA) and SS-DEWMA (called the sum of squares of DEWMA) charts.