Type of Document	Thesis
Author	McCulloh, Ian Andrew
URN	etd-04122004-131857
Title	Generalized Cumulative Sum Control Charts
Degree	Master of Science
Department	Industrial and Manufacturing Engineering, Department of
Advisory Committee	Advisor Name Title Joseph Pignatiello Committee Chair James Simpson Committee Member Okenwa Okoli Committee Member
Keywords	Average Run Length ARL Generalized Sequential Probability Ratio Test Quality Engineering Control Charts CUSUM
Date of Defense	2004-03-31
Availability	unrestricted
Abstract Industrial manufacturing processes can experience a variety of changes to important quality characteristics as a result of tool breakage, tool wear, introduction of new raw materials, and other factors. Statistical process control charts are often used to monitor for changes in quality characteristics for manufacturing processes. The control chart computes a statistic based on measured observations of the process and compares it to control limits. When the statistic exceeds a control limit, the control chart signals that the process is out-of-control. Quality engineers would then search for the special cause responsible for the change in the process. Rapid detection by the control chart is important to minimize the production of poor quality items as a result of an out-of-control process. Control charts that detect changes rapidly can therefore save critical process down-time and expense. This research investigates several new control charts related to the commonly used cumulative sum (CUSUM) control chart. The new control charts use control limits that change as a function of the number of process observations instead of remaining constant. The performances of these new control charts are compared to several other charts including the Shewhart chart, the CUSUM, and the exponentially weighted moving average charts. Compared to the standard CUSUM, the proposed control charts can detect a change in a process more rapidly for a given range of shifts in the mean of a process. The proposed control charts offer a flexibility to quality engineers for better optimization of monitoring schemes for manufacturing processes.
Files	Filename       Size       Approximate Download Time (Hours:Minutes:Seconds)     28.8 Modem   56K Modem   ISDN (64 Kb)   ISDN (128 Kb)   Higher-speed Access    McCullohThesis.pdf 536.83 Kb 00:02:29 00:01:16 00:01:07 00:00:33 00:00:02