Measurement System Analysis
- Is an experimental and mathematical method of determining how much the variation
within the measurement process contributes to overall process capability.
- Assess the adequacy of a measurement system for a given application.
Two Sources of Variation:
1. Part – to – Part Variation
2. Measurement System Variation
Categories of Measurement System Error
1. Accuracy
- Difference between the part’s measured and actual value
2. Precision
- Variation when the same part is measured repeatedly with the same device
Reproducibility
Two Components of Precision:
1. Repeatability
- Variation due to the measuring device/gage; or the variation observed when
the same operator measures the same part repeatedly with the same device.
2. Reproducibility
- Variation due to the measuring system/operator; or the variation observed
when different operators measure the part using the same device
Three Components of Accuracy:
1. Stability
- Measure of how well the system performs over time; the total variation
obtained with the particular device, on the same part, when measuring a single
characteristic over time.
2. Bias
- Measure of the inaccuracy in the measurement system; the difference
between the observed average measurement and a master value
3. Linearity
- Measure of how the size of the part affects the bias of the measurement
system; the difference in the observed bias values through the expected range of
measurements
Measurement System Analysis Tools
MSA tools that help evaluate accuracy and precision in measurement system:
1. Gage R&R Study
2. Gage Run Chart
3. Gage Linearity and Bias Study
4. Type 1 Gage Study
5. Attribute Agreement Analysis
6. Attribute Gage Study (AIAG Method)
Gage R&R
- Method to evaluate measurement system to determine the amount of variation it
contributes to the total observed process variation.
- Manufacturing – gages are evaluated
- Service – appraisers are evaluated
IDEAL: small contribution of variation for Gage R&R and large contribution of variation
for Part-to-Part
Guidelines for Measurement System:
Acceptable: score < 10%
Potentially acceptable (depends on the criticality/importance of costs, risks,
measurements, etc.): 10% < score < 30%
Unacceptable: score > 30%
Types of Factors in Gage R&R Studies:
1. Crossed
- Assess the variation in the measurement system when every operator measures every
part in the study
2. Nested
Assess the variation in the measurement system when you can measure each part only
once, such as with destructive testing
3. Expanded
- Assess the variation in the measurement system when it is affected by factors in
addition to part and operator such as temperature and location
Two Methods for Assessing/Analyzing R&R:
1. X bar and R
- Can only be used by crossed
2. ANOVA
- Can be used by crossed, nested and expanded
- More accurate
Interpreting the Results
Session Window
ANOVA Table
- ANOVA is used to calculate variance components, and then uses those components to
estimate the percent variation due to the measuring system.
- Includes terms for the part, operator, and operator – part interaction.
Source
-source of the variation in the data
Degrees of Freedom (DF) in the source
-measures how much information is available to calculate each SS.
a = number of parts
b = number of operators
n = number of replicates
DFPart = a – 1
DFOperator = b – 1
DFPart*Operator = (a – 1)(b – 1)
DFRepeatability = ab*(n – 1)
DFTotal = abn - 1
Sum of Squares (SS) due to the source
-sum of squared distances, and is a measure of variability that is from
different sources
Mean Sum of Squares (MS) due to the source
-variability in the data from different sources
-accounts for the fact that different sources have different numbers of
levels or possible values
F-static (F)
-used to determine whether the effects of operator, part, or
operator*part are statistically significant
P-Value (P)
-probability of obtaining a test static (such as F) that is at least as extreme
as the value that is calculated from the sample, if the null hypothesis is
true.
*If the p-value < α, reject the null hypothesis in favour of the alternative
hypothesis; if p-value > α, fail to reject the null hypothesis. Null
hypothesis is never accepted, it can only be reject or fail to reject it.