Phase I Statistical Process Control Charts: An Overview and Some Results

Statistical process control charts for monitoring military injuries

“…The historical average (also known as phase I) is typically determined from at least 20–25 historical baseline data points, and then subsequent data points (phase II) are monitored using the parameters established during phase I 19. The charts indicate (‘signal’) statistically significant increases or decreases in quantitative data points, when at least one data point is above the upper threshold or below the lower threshold.…”

A general framework for monitoring complex processes with both in-control and out-of-control information

“…A large K may cause the amount of data in certain OC class to be very small, and then influence the subsequent training procedure. Thus, we should limit the size of OC clusters K. As to the method for clustering, since the Phase I clustering method is not the focus of this paper, some existing works (Zorriassatine, Tannock, & O'Brien, 2003;Chakraborti, Van der Laan, & Bakir, 2008;Zou, Tsung, & Liu, 2008) are suggested to achieve good clustering performance, instead.…”

Use of prior manufacturer specifications with Bayesian logic eludes preliminary phase issues in quality control

“…We base our decision regarding a on the FAP performance metric [25], which is defined as the probability of getting at least one false alarm in the m observations of the preliminary phase. So, assuming independence among successive times and setting the control limits at AEz a=2 in a finite horizon of m À 1 observations (since charting will not include the first data point), the value of a as a function of FAP and the number of data points m will be:…”