1. Measurement Systems Analysis
Sebastian Fishman
April 2019

2. About the presenter
Flir Systems – Engineering Manager – current… (  )
Raytheon Vision Systems - Quality Manager
SpaceX - Senior Lean Facilitator
Allergan - Senior Project Engineer
Qimonda (Infineon Technologies) - Senior Process Engineer
Hynix Semiconductors of America - Process Engineer
Work Experience
Master’s, Applied Physics – University of Oregon, Oregon, USA
Bachelor’s, Chemical Engineering – Shenkar College, Ramat Gan, Israel
Education
Kepner-Tregoe ATS Instructor Certified
Lean Six Sigma Master Black Belt – Ohio State University/MoreSteam
Lean Six Sigma Black Belt – Virginia Commonwealth University
Six Sigma Black Belt – ASQ
Certified Quality Engineer – ASQ
Certified Quality Manager – ASQ
Certified CAPA Investigator – PathWise
Certified Medical Device Regulations - AAMI
Certifications
Senior member
Channel Cities Board Member
ASQ Member since 2011
ASQ

3. About Today Measurement System Analysis What is it?
Why do we do use it?
How is it done?
What issues do we encounter
About Today

4. Measurement System Analysis History
For many years different automotive companies impose different methods on suppliers about measuring
In 1990 Chrysler, Ford and General Motors agreed to a unified platform – via AIAG the 1st Edition of the Measurement System Analysis Handbook was released
The manual was developed to meet the needs of the automotive industry, in conjunction with ASQ and ASTM
Measurement System Analysis is required by ISO (ISO/TS 16949), automotive and other industries (medical devices, pharma, aerospace, semiconductor) – explicitly or by inference (test method validations)

5. MSA Processes AAIG (ANOVA, Gauge)
EMP (Evaluation the Measurement Process)
A Comparison between both methods is not part of this presentation
AIAG
EMP
Estimates EV
σpe
Standard deviation of the measurement system (repeatability) AV σo
Standard deviation between the operators (reproducibility)
GRR σe
Standard deviation of the combined repeatability and reproducibility PV σp
Standard deviation of the variation in the parts used in the study TV σx
Standard deviation of the total variation (combining GRR and PV)

6. Goal: base your decisions on correct data
What is MSA?
Measurement systems analysis assesses the adequacy of a measurement system for a given application
Source: Minitab
An experimental and mathematical method of determining how much the variation within the measurement process contributes to overall process variability
Source: isixsigma.com
Assesses the precision, consistency, and bias of a measurement system
Source: JMP.com
The purpose of Measurement System Analysis is to qualify a measurement system for use by quantifying its accuracy, precision, and stability
Source: MoreSteam.com
Measurement System: A collection of instruments and gages, standards, operations, methods, fixtures, software, personnel, environment, and assumptions used to quantify a unit of measure or fix assessment to the feature characteristic being measured; the complete process used to obtain measurement
Source: Measurement System Analysis Handbook, third edition, AIAG
Goal: base your decisions on correct data

7. Where is the Data coming from?
Measurement System
People measuring
(Reproducibility)
Person variability
Person and Part interaction variability
Measurement tools
(Repeateability)
Parts being measured
Gauge R&R
Gage or Gauge?
Gage is the spelling of an obsolescent word meaning a pledge, a challenge, etc. Gauge is the spelling to use when you measure measurement, estimate, or standard.

8. What is being evaluated
System
Stable
No time dependency
Accurate
No Bias
No Linearity
Precise
Repeatable
Reproducible

9. Stability
Establish reference material (can be NIST traceable or internal)
Establish measurement frequency, data gathering
Evaluate performance (trend analysis, Control Limit violation)

10. Accuracy Bias Linearity No difference between true and actual
Non linear change between measured parts and actual
Linear correlation between measured parts and actual

11. Tip On How to Remember: Machines Repeat – People Reproduce
Precision
Spread or variation of measured values
Repeatability
Variation due to the measuring device, or the variation observed when the same operator measures the same part repeatedly with the same device
Reproducibility
Variation due to the measuring system, or the variation observed when different operators measure the same part using the same device
Tip On How to Remember: Machines Repeat – People Reproduce

12. Precision (Gauge RR) Part A Operator 1 Part B Operator 2 Part C x10
Other Combinations
10 parts, measure 2-3 times
One operator, multiple instruments
Crossed vs Nested (destructive testing)
Operator 1
Part B
Operator 2
Part C
x10

13. What do we mean?

14. Conditions for Running Gauge Study
Need sufficient stats
Rule of thumb is no less than 30 measurements – why?
Replication and repetition
Break setups to capture the whole process – capture processing impact on measurement
Randomization
Parts and operators should be randomly order for measuring – avoid loading effects
Blind measurement
Operator should not know what part is measuring (or what they measured before) – avoid operator influence
Conditions for Running Gauge Study

15. Sources of Errors in Execution
Example is not limited, other more detail sources of variation and issues exist

16. Type of MSA Continues: Measurement (Gauge RR Study)
Numbers
Uses ANOVA for assessing the statistical significance of the sources of variation
Attribute: Appraiser Assessment (Attribute Gauge Study)
Assessments (pass/fail, good/bad)
Uses Kappa Statistics to assess the agreement level for the appraiser to him/herself and the Reference material (0: no agreement, 1: full agreement)
Requires more data points for statistical significance

17. Process for data gathering
Define Objectives
Attribute or Gauge
Define reference material
Nominal parts or known parts (good/bad)
Set up Process
Randomization
Data gathering sheet
Ensure measurement instrument has the correct resolution
Train operators on the process
How to measure/do the activity
Data gathering (blind)
Process data
Plug into Excel, JMP, Minitab or other statistical software
Conclusion
Is the instrument capable?
Are the Operators consistent and correctly analyzing?

18. Acceptance Guidelines
Source Automotive Industry Action Group (AIAG) (2010). Measurement Systems Analysis Reference Manual, 4th edition. Chrysler, Ford, General Motors Supplier Quality Requirements Task Force
AIAG also states that the number of distinct categories into which the measurement system divides process output should be greater than or equal to 5
Percentage of process variation
Acceptability
Less than 10%
The measurement system is acceptable.
Between 10% and 30%
The measurement system is acceptable depending on the application, the cost of the measurement device, cost of repair, or other factors.
Greater than 30%
The measurement system is not acceptable and should be improved.

19. Distinct Categories
Represents the number of non-overlapping confidence intervals that span the range of product variation
The number of distinct categories also represents the number of groups within your process data that your measurement system can discern
Example (Distinct Categories)
2 = cannot distinguish the pars
3 = can distinguish a bit (high, medium, large)
4 = can distinguish a bit more
5 = can distinguish sufficiently to come up with a conclusion
Categories is a sign of Resolution
Ability of the measurement system to detect and faithfully indicate small changes in the characteristic of the measurement result
Rule of tens (10 to 1 for measurement capabilities)

20. Sources http://reliawiki.org/index.php/Measurement_System_An alysis
ple-materials/FuelInjectorNozzles-EN.pdf
systems-analysis-platform-options.shtml
how-to/quality-and-process-improvement/measurement- system-analysis/supporting-topics/gage-r-r-analyses/is- my-measurement-system-acceptable/

21. Questions?