The measurement system is essential when you’re measuring because the wrong measurement system means inaccurate data.
This page is structured in this way:
- Measurement Variation
In a measurement system, one thing to know is the difference between accuracy and precision.
- Accuracy is how far the measure is from the actual value.
- Precision is when I measure how far they are from each other.
So I can have a very precise system because the measurements are very near, but with low accuracy because they are very far from reality.
About Accuracy we need to know that:
- The bias is the difference between the observed average measure to the true value.
- Linearity is another definition related to bias and says how the measurement system works across a range. So you have a linear bias across all the different values or not? Maybe people have a low bias when measuring meters, but when they pass to km or millimeters, they have a high bias that isn’t linear.
- Stability is the capacity of a measurement system to produce the same value over time when you measure the same thing.
About Precision we need to asses:
- Repeatability: When the same person measures something, you want to get the same value. A way to make this happen can be, for example, the use of a master sample. In this way, you can look at the difference between what you’re measuring and the sample to assess the correct measurement.
- Reproducibility: When different people get different measures. You can assess this problem with a Standard Operation Procedure (SOP).
Gage Repeatability and Reproducibility (Gage R&R) measures the amount of variability in measurements caused by the measurement system itself.
There are three different kinds of Gage R&R Study that are:
- Crossed: We have N operators and M parts. Each of these N operators measures all the M parts;
- Nested: We have more operators and more parts to test. Each operator test a different subset of parts. This is used when the test destroys the part.
- Expanded: when you need to test more than two factors (not only parts and operator but also others).
The Gage R&R can be performed with 3 methods:
- Range method: Give an approximation of measurement variability but without Repeatability and Reproducibility measure;
- Average Range method: This method quantifies the variability and gives Repeatability and Reproducibility measures. Only Crossed Gage R&R can be performed with this method. X Bar R Control Charts is a way to perform this method.
- Analysis of variance method: It’s the most accurate method for Repeatability and Reproducibility measurement. It also quantifies the variability of the interaction between operators and parts. ANOVA is a way to perform this method.