Control Charts for the Standard Deviation

Abstract

Abstract Control charts are graphical tools to monitor the activity of a process. They are used to determine if the variation in a process is due to the presence of common causes (those that are inherent to the process) or special causes (those that are not part of the process as designed). A process is said to be in statistical control when it is under a system of common causes. A standard deviation control chart monitors process variation, as measured by the standard deviation of observations, through subgroups collected over a period of time. Typically, three lines are depicted on a control chart—the centerline (CL), the upper control limit (UCL), and the lower control limit (LCL). Rules for determining out‐of‐control conditions are defined, the most commonly used one corresponds to a point falling outside the control limits. Some parameters for the construction of the standard deviation control chart are the subgroup size, frequency of sampling, and placement of the control limits from the CL, based on the choice of an acceptable level of a false alarm rate.