The Westgard Rules were originally developed by James Westgard. He published the rules in his book and discussed them extensively during his seminars on quality control. The rules are all related to statistical patterns and how likely they are to occur with just random variability. If the results of an analysis are outside of the Westgard Rules, therefore, it should indicate that there is likely something faulty in the precision or accuracy in the machine or process in question.
The rules are:
- One measurement exceeds 2 standard deviations either above or below the mean of the reference range.
- One measurement exceeds 3 standard deviations either above or below the mean of the reference range.
- Two consecutive measurements exceed 2 standard deviations of the reference range, and on the same side of the mean.
- Two measurements in the same run have a 4 standard deviation difference (such as one exceeding 2 standard deviations above the mean, and another exceeding 2 standard deviations below the mean).
- Four consecutive measurements exceed 1 standard deviation on the same side of the mean.
- 10 Consecutive measurements are on the same side of the mean.
As you can see, these are all different types of unexpected results if there were just randomness involved. These indicate that something is wrong. If the problem is with the machine itself, it may need to be fixed. If it is just with the parts being used, the run may need to be redone with different parts.
Every company that uses the Westgard Rules will need to determine what the most likely cause of deviations like this will be. The rules do not help to identify the specific problem, but rather just raise the flag that there is something wrong somewhere in the process. This is an important diagnostic tool because it will prevent a company from continuing to run more and more processes where there is something that is likely wrong, which can create a lot of waste.
Similar Glossary Terms
- Control Charts
- Quality Engineering
- Gage R&R
- Statistical Process Control (SPC)
- Taguchi Method
- APQP (Advanced Product Quality Planning)
- Variance Inflation Factor (VIF)