Non Random Patterns - Part 2
Shift in the Process: A shift in the process is defined as a pattern in which seven consecutive data points appear on only one side of the mean. Statistical studies have shown that if the number of points on one side of the mean is less than 7 then it is not a non-random pattern and could have occurred due to random causes.
A shift in the process is very important for the management to detect. This is because it signifies that the average of the process has moved to a different level. Therefore, the control chart with the mean and control specifications based on old measures is redundant and will give wrong information. It is for this reason that it is important for the organization to detect this pattern as early as possible and take corresponding measures. The most common causes of the appearance of non- random shift patterns is as follows:
Process Improvement: Shifts in process means usually happen as a result of a deliberate attempt through a process improvement initiative. The best practises to do any task are continuously evolving. Introduction of new technologies enables newer and better ways of performing the work. This pattern can be detected on the control charts and the process improvement can be institutionalised.
Introduction of New Inputs: Another common reason for shift in the process mean is the introduction of new inputs. These new inputs could be markedly better or worse than the inputs used earlier and as a result change the performance of the entire process
Change in Quality Control Measures: The shift could also be the result of a change in quality control measures. The thrust may have changed from cure to prevention and as a result the quality checks may have dramatically improved the process, shifting its mean.
Two out of three points at Two Sigma or Beyond
As the name suggests, this may be quite straightforward. However the logic behind this pattern is interesting. The normal curve states that 95% of all data points will lie between 2 standard deviations. However, if 2 out of 3 consecutive points are lying on or above the standard deviation, we can be rest assured that the 95% criteria is not being met, at least in that point in time.
This is therefore an early sign to the management that the process may be about to go out of control. A thoughtful management will make corrective action plans that will look into the process, understand the issue and correct it as soon as possible.
❮❮ Previous | Next ❯❯ |
Related Articles
- Step 3A: Control Chart - Discrete Data
- Step 3B: Control Chart - Continuous Data
- Non Random Patterns - Part 1
- Closing a Six Sigma Project
- Recording Results of Six Sigma Project
Authorship/Referencing - About the Author(s)
The article is Written and Reviewed by Management Study Guide Content Team. MSG Content Team comprises experienced Faculty Member, Professionals and Subject Matter Experts. We are a ISO 2001:2015 Certified Education Provider. To Know more, click on About Us. The use of this material is free for learning and education purpose. Please reference authorship of content used, including link(s) to ManagementStudyGuide.com and the content page url.