Page 201 - Special Topic Session (STS) - Volume 3

STS540 Zhi Lin C. et al.

A comparative study between the standard
deviation of the time to signal (SDTS)
performance of the GR and SSGR schemes
3
2
1
Zhi Lin Chong , Michael BC Khoo , Huay Woon You ,
5
Wei Lin Teoh , Sin Yin Teh
4
1 Universiti Tunku Abdul Rahman, Perak, Malaysia.
2,5 Universiti Sains Malaysia, Penang, Malaysia
3 Universiti Kebangsaan Malaysia, Selangor, Malaysia
4 Heriot-Watt University Malaysia, Putrajaya, Malaysia.

Abstract
The Group Runs (GR) scheme is introduced as an improved version of the
synthetic scheme; whereas the Side-Sensitive Group Runs (SSGR) scheme is
introduced as an enhancement of the GR scheme with the side-sensitive feature.
However, the average run length or average time to signal criterion is commonly
applied as a sole performance measure for both the GR and SSGR schemes.
Hence, this paper aims to reexamine the performances of the GR and SSGR
schemes using the performance measure of the standard deviation of the time
to signal (SDTS), which have not been investigated in the literature. Based on
the comprehensive simulation studies, it is discovered that although the SSGR
scheme has a better out-of-control SDTS performance compared to the GR
scheme, its in-control SDTS performance is inferior to that of the GR scheme.

Keywords
Control chart, Group Runs (GR), Side-Sensitive Group Runs (SSGR)

1. Introduction
Contemporarily, quality control is vitally viewed as high quality product will
boost the reputation and improve the sustainability of any industry. Statistical
Process Control (SPC) is commonly implemented to control and enhance the
product quality. Among the many tools in SPC, control scheme receives the
most attention due to its effectiveness in categorizing the process as in-
control (IC) or out-of-control (OC). The first control chart was introduced by
W. A. Shewhart in the 1920s and it is called the Shewhart scheme. Since then,
̅
numerous extensions on the Shewhart scheme have been proposed.
̅
Although the Shewhart scheme is the most widely applied control scheme
̅
in industries and it is effective in detecting large and abrupt mean shifts, it
receives much criticisms due to its inefficiency in detecting small to moderate
and persistent mean shifts. To solve this problem, Roberts (1959) and Page
(1954) introduced the Exponentially Weighted Moving Average (EWMA) and
Cumulative Sum (CUSUM) schemes, respectively, which are effective in
detecting small mean shifts.

190 |I S I W S C 2 0 1 9