How to Quickly Identify the Process Behaviour Using Control Charts
No doubt that Process behaviour charts are the life blood source for any process improvement in manufacturing industry or IT industry. there are some quick thumb rules to identify whether the process is behaving as expected or not
This article has been tagged as a stub. That means it's off to a good start, but still has plenty of room to grow before it reaches its full potential. Can you help it flourish? If you think the article offers complete and accurate instructions, feel free to remove this tag! |
Steps
-
1Remember the data points/sample data should have at least 10 to 15 drawn in particular time frame(say weekly or monthly)
-
2Identify and isolate the common cause of variation and special cause of variation
- If your process is well behaving then 90% of the time, your data points always falls with in +/- 3 sigma limits
- Data points on the mean value is very consistent
-
3If you see any trend of 8 or more data points above mean or below mean, then depend on the type of Metric that you are measuring you need to take the context of your measurement.
-
4Depend on the type of metric you can decide whether your process is with in its control or any signals existing
-
5If you see any trend upward or downward 6 points or more, then definitely it's because of special cause of variation however you need to have thorough analysis in place to confirm the same but that 6 point up or down is in desirable direction for the business then the cause may be because of stringent/dedicated/extreme focus on delivering your service/product results such 6 point up/down.
- If you see these types of trends, then "depend on" the type of Metric you need to recalculate your upper control limit and lower control limits
We could really use your help!
schoolplace antics?
rate articles?

electrical maintenance?

using sunbeds?

building muscle?

Article Info
Categories: Stub | Software | Computers and Electronics
Thanks to all authors for creating a page that has been read 3,490 times.
About this wikiHow