Run length distribution of two-sided CUSUM procedures for continuous random variables

Citation

Tan, Y. F. and Pooi, A. H. (2005) Run length distribution of two-sided CUSUM procedures for continuous random variables. Bulletin of the Institute of Mathematics Academia Sinica, 33 (2). pp. 91-113. ISSN 2304-7909

[img] Text
4.pdf
Restricted to Repository staff only

Download (468kB)

Abstract

Cumulative sum (CUSUM) control charts are very effective in detecting special causes. In judging the perfor- mance of a CUSUM procedure, it is important to know its run length distribution. Presently iterative formulas are derived for finding the run length distribution of two-sided CUSUM. The ap- plication of the iterative formulas is illustrated in the normal two- sided CUSUM. 1. Introduction. Cumulative sum (CUSUM) procedures are widely used to monitor the quality of products from manufacturing processes. There are two main types of CUSUM procedures, namely the one-sided and two- sided CUSUM procedures. The one-sided CUSUM scheme is further clas- sified into two types, namely the lower-sided CUSUM and upper-sided CUSUM. The lower-sided CUSUM is intended to detect an upward shift in the process mean while the upper-sided CUSUM is intended to detect a downward shift in the process mean. The two-sided CUSUM procedure is intended to detect a shift which could be downward or upward in the process mean. The run length of the CUSUM procedure is the time elapsed before the process is declared to be out of control. The run length distribution and its parameters measure the performance of a CUSUM procedure. The average

Item Type: Article
Subjects: T Technology > TA Engineering (General). Civil engineering (General)
Divisions: Faculty of Engineering (FOE)
Depositing User: Ms Rosnani Abd Wahab
Date Deposited: 28 Jan 2014 02:59
Last Modified: 28 Jan 2014 02:59
URII: http://shdl.mmu.edu.my/id/eprint/5022

Downloads

Downloads per month over past year

View ItemEdit (login required)