The article continues to consider the possibility of maintaining the technological process in a statistical control state with the use of cumulative sum charts. A numerical method for estimating the statistical controllability of a process using a scheme of decision intervals based on a computational algorithm is considered.
Cumulative sum charts can provide an out-of-control process detection and, unlike a regular Shewhart charts, allow determine quickly and accurately the process change area, as well as when corrective action is required. Unlike the graphical method using the V-mask, when the mandatory presence of an operator is required to monitor the state of the process, the numerical estimation method using the decision interval scheme can be automated and provide appropriate signals without operator involvement.
The method under consideration reveals a point in time when a change in process cannot be caused solely by the influence of random causes, that is, there is a significant systematic influence, a disorder of the process and a shift in the primary level of adjustment. The use of the same V-mask allows to solve the opposite problem: when fixing the moment of disorder of the technological process to determine what time this event took place.
When constructing a V-mask, its «opening angle» is directly related to the maximum permissible displacement of the process and, based on the requirements for the quality of the technological process, it is necessary to have an aggregate (set) of such V-masks. When applying the numerical method, this problem does not occur.
The article shows the equivalence of graphical and numerical methods theoretically and numerically, which, given the certain undoubted advantages of the latter, proves the prospect of its application.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.