MSA - Measurement Systems Analysis

32,344 views

Published on

MSA - Measurement Systems Analysis

Published in: Business
3 Comments
23 Likes
Statistics
Notes
No Downloads
Views
Total views
32,344
On SlideShare
0
From Embeds
0
Number of Embeds
89
Actions
Shares
0
Downloads
1,976
Comments
3
Likes
23
Embeds 0
No embeds

No notes for slide
  • MSA - Measurement Systems Analysis

    1. 1. Measurement Systems Analysis <ul><li>Champion Training </li></ul>Six Sigma Champion Training Improve Analyze Measure Control
    2. 2. Learning Objectives <ul><li>Understand the language of Measurement </li></ul><ul><li>Show the importance of Measurement </li></ul><ul><li>Walk away knowing how to perform a Gage R&R and how to interpret results </li></ul><ul><li>Share some lessons learned </li></ul>Six Sigma Champion Training
    3. 3. Since Measurement systems represent a sub-process within a process... Six Sigma Champion Training <ul><li>They are subject to Variation. </li></ul><ul><li>What could be the source of this variation? </li></ul><ul><li>Why do Measurements Vary? </li></ul><ul><li>CLASS EXERCISE : Break into teams. Do a cause-effect Diagram to determine the causes of variation in Measurement. </li></ul>
    4. 4. Sources of Measurement Variation Six Sigma Champion Training
    5. 5. Possible Sources of Process Variation Six Sigma Champion Training Long-term Process Variation Short-term Process Variation Variation w/i sample Actual Process Variation Stability Linearity Repeatability Accuracy Variation due to gage Variation due to operators Measurement Variation Observed Process Variation We will look at “repeatability” and “reproducibility” as these are the primary contributors to measurement error. Reproducibility
    6. 6. Knowledge to be obtained <ul><li>How big is the measurement error? </li></ul><ul><li>What are the sources of measurement error? </li></ul><ul><li>Is the gage stable over time? </li></ul><ul><li>Is the gage capable for this process? </li></ul><ul><li>How do we improve the measurement system? </li></ul>Six Sigma Champion Training
    7. 7. Sources of Variation Six Sigma Champion Training Product Variability (Actual variability) Measurement Variability Total Variability (Observed variability)
    8. 8. Effects of Measurement Error Six Sigma Champion Training Averages Variability Measurement System Bias — Determined through “Accuracy Study ” Measurement System Variability — Determined through “R&R Study”
    9. 9. Work Around Gage Error Six Sigma Champion Training If you want to decrease your gage error take advantage of the standard error square root of the sample: Example: Gage error of 50% can be cut in half if your point estimate is a sample of 4 data points. THIS IS USED AS A SHORT TERM APPROACH TO PERFORM A STUDY, BUT YOU MUST FIX THE GAGE. = sample size
    10. 10. Terminology <ul><li>Location related terms: </li></ul><ul><ul><li>True value </li></ul></ul><ul><ul><li>Bias </li></ul></ul><ul><ul><li>Linearity </li></ul></ul><ul><li>Stability (over time) </li></ul><ul><li>Variation related terms </li></ul><ul><ul><li>Repeatability </li></ul></ul><ul><ul><li>Reproducibility </li></ul></ul><ul><ul><li>Linearity </li></ul></ul>Six Sigma Champion Training
    11. 11. <ul><li>True value: </li></ul><ul><ul><li>Theoretically correct value – unknown and unknowable </li></ul></ul><ul><ul><li>Reference standards </li></ul></ul><ul><ul><li>NIST standards </li></ul></ul><ul><li>Bias </li></ul><ul><ul><li>Distance between average value of all measurements and true value </li></ul></ul><ul><ul><li>Amount gage is consistently off target </li></ul></ul><ul><ul><li>Systematic error or offset </li></ul></ul>Six Sigma Champion Training
    12. 12. BIAS — Is the difference between the observed average of the measurement and the reference value. The reference-value is the value that serves as an agreed-upon reference. The reference value can be determined by averaging several measurements with a higher level (e.g., metrology lab) of measuring equipment. Warning: Don’t assume your metrology reference is gospel. Six Sigma Champion Training Observed Average Value Reference Value ACCURACY IS THE SAME AS BIAS BIAS Definition
    13. 13. Linearity <ul><li>Difference in the accuracy values of a gage through the expected operating range of the gage </li></ul>Six Sigma Champion Training Good Linearity Bad Linearity
    14. 14. Stability <ul><li>The distribution of measurements remains constant and predictable over time for both mean and standard deviation </li></ul><ul><li>Total variation in the measurements obtained with a gage, on the same master or master parts, when measuring a single characteristic over an extended time period. </li></ul><ul><li>Evaluated using a trend chart or multiple measurement analysis studies over time </li></ul>Six Sigma Champion Training
    15. 15. Six Sigma Champion Training Stability (drift) Definition Stability — Is the total variation in the measurement obtained with a measurement system (test / gage ) on the same master parts when measuring a single characteristic over an extended time period. Time-1 Time-2 time Magnitude Stability Points to the frequency of Mean center Calibration
    16. 16. <ul><li>Total variation in the measurement system </li></ul><ul><li>Measure of natural variation of repeated measurements </li></ul><ul><li>Terms: Random Error, Spread, Test/Retest error </li></ul><ul><li>Repeatability and Reproducibility </li></ul>Six Sigma Champion Training    MS G O 2 2 2  
    17. 17. Repeatability <ul><li>The inherent variability of the measurement system </li></ul><ul><li>Variation in measurements obtained with a gage when used several times by one operator while measuring a characteristic on one part. </li></ul><ul><li>Estimated by the pooled standard deviation of the distribution of repeated measurements </li></ul><ul><li>Repeatability is less than the total variation of the measurement system </li></ul>Six Sigma Champion Training
    18. 18. Six Sigma Champion Training Repeatability Definition REPEATABILITY Repeatability — The variation in measurements obtained with one measurement instrument when used several times by one appraiser while measuring the identical characteristic on same part.
    19. 19. Reproducibility <ul><li>Operator variability of the measurement system </li></ul><ul><li>Variation in the average of the measurements made by different operators using the same gage when measuring a characteristic on one part </li></ul><ul><li>Must be adjusted for gage variation </li></ul><ul><li>Reproducibility is less than the total variation of the measurement system </li></ul>Six Sigma Champion Training
    20. 20. Reproducibility Definition Six Sigma Champion Training Reproducibility — Is the variation in the average of the measurements made by different appraisers using the same measuring instrument when measuring the identical characteristic on the same part. Reproducibility Operator-A Operator-C Operator-B
    21. 21. Six Sigma Champion Training The Nature of Process Variation 4 3 2 1 Precise but not Accurate Accurate but not Precise 5 4 3 2 1 . . . . . .Test equipment MUST be a least 10 times more accurate & precise then what’s being tested Rule of thumb:
    22. 22. Measurement System Discrimination <ul><li>Least count should be at most one-tenth of the total process capability or tolerance (6 sigma) </li></ul><ul><ul><li>Process capability 10 Max Least count 1 </li></ul></ul><ul><li>Part to Part variation must be greater than the smallest unit of measure </li></ul><ul><li>Range control chart provides best indication of inadequate discrimination </li></ul><ul><ul><li>Occurs when only 1,2, or 3 possible values for the range within the control limits exists </li></ul></ul><ul><li>Number of Distinct Categories equals part sigma/ total </li></ul><ul><li>gage sigma  1.41. </li></ul>Six Sigma Champion Training
    23. 23. Measurement System Capability <ul><li>Gage Capability </li></ul><ul><li>Addresses what percent of the tolerance or process capability is taken up by measurement error . </li></ul><ul><li>Best case: 10% Acceptable: 30% </li></ul><ul><li>Includes both repeatability and reproducibility </li></ul><ul><ul><li>Operator  Unit  Trial Experiment </li></ul></ul>Six Sigma Champion Training Usually expressed as percent
    24. 24. Measurement Error Effect on Capability Index <ul><li>We know that </li></ul><ul><li>Therefore: </li></ul>Six Sigma Champion Training where
    25. 25. R&R Effect on Capability Six Sigma Champion Training 70% 60% 50% 40% 30% 10%
    26. 26. <ul><li>Variable Gage R&R </li></ul><ul><ul><li>Numbers </li></ul></ul><ul><ul><li>Units of measure </li></ul></ul><ul><li>Attribute Gage R&R </li></ul><ul><ul><li>Subjective (cosmetic defects) </li></ul></ul><ul><ul><li>Scatter of defects </li></ul></ul><ul><ul><li>feel/visual </li></ul></ul>Six Sigma Champion Training Types of R&R Studies
    27. 27. Six Sigma Champion Training The Inspection Exercise The Necessity of Training Farm Hands for First Class Farms in the Fatherly Handling of Farm Live Stock is Foremost in the Eyes of Farm Owners. Since the Forefathers of the Farm Owners Trained the Farm Hands for First Class Farms in the Fatherly Handling of Farm Live Stock, the Farm Owners Feel they should carry on with the Family Tradition of Training Farm Hands of First Class Farmers in the Fatherly Handling of Farm Live Stock Because they Believe it is the Basis of Good Fundamental Farm Management. Task: Count the number of times the 6th letter of the alphabet appears in the following text.
    28. 28. Convert Data Six Sigma Champion Training Try To Always Convert Attribute To Variables Examples: <ul><li>End Disk Height </li></ul><ul><li>Likert Scale </li></ul><ul><li>Leak Rate (go/no go) </li></ul><ul><li>Mass Spec </li></ul>
    29. 29. <ul><li>EV= Equipment Variation (Repeatability) </li></ul><ul><li>AV= Appraiser Variation (Reproducibility) </li></ul><ul><li>R&R= Repeatability & Reproducibility </li></ul><ul><li>PV= Part Variation </li></ul><ul><li>TV= Total Variation of R&R and PV </li></ul><ul><li>K1-Trial, K2-Operator, & K3-Part Constants </li></ul>Six Sigma Champion Training Basic Terms
    30. 30. <ul><li>Generally two or three operators </li></ul><ul><li>Generally 10 units to measure </li></ul><ul><li>Each unit is measured 2-3 times by each operator </li></ul>Gage R&R study Six Sigma Champion Training
    31. 31. Preparation for a Measurement Study <ul><li>Determine if reproducibility is an issue. If it is, select the number of operators to participate. </li></ul><ul><li>Operators selected should normally use the measurement system. </li></ul><ul><li>Select samples that represent the entire operating range. </li></ul><ul><li>Gage must have graduations that allow at least one-tenth of the expected process variation. </li></ul><ul><li>Insure defined gaging procedures are followed. </li></ul><ul><li>Measurements should be made in random order. </li></ul><ul><li>Study must be observed by someone who recognizes the importance of conducting a reliable study. </li></ul>Six Sigma Champion Training
    32. 32. Procedure for Performing R&R Study <ul><li>Calibrate the gage, or assure that it has been calibrated. </li></ul><ul><li>Have the first operator measure all the samples once in random order. </li></ul><ul><li>Have the second operator measure all the samples once in random order. </li></ul><ul><li>Continue until all operators have measured the samples once (this is Trial 1). </li></ul><ul><li>Repeat above steps for the required number of trials. </li></ul><ul><li>Use GR&R form to determine the statistics of the study. </li></ul><ul><ul><li>Repeatability, Reproducibility & %GR&R </li></ul></ul><ul><ul><li>Standard deviations of each of the above </li></ul></ul><ul><ul><li>% Tolerance analysis </li></ul></ul><ul><li>Analyze results and determine action, if any. </li></ul>Six Sigma Champion Training
    33. 33. Six Sigma Champion Training Guidelines % R&R Results <5% No issues  10% Gage is OK 10% – 30% Maybe acceptable based upon importance of application, and cost factor Over 30% Gage system needs improvement/corrective action Variable Gage R&R
    34. 34. <ul><li>PT1 PT2 PT3 PT4 PT5 PT6 PT7 PT8 PT9 PT10 OP/TRIAL </li></ul><ul><li>0.65 1.00 0.85 0.85 0.55 1.00 0.95 0.85 1.00 0.60 A1 </li></ul><ul><li>0.60 1.00 0.80 0.95 0.45 1.00 0.95 0.80 1.00 0.70 A2 </li></ul><ul><li>0.55 1.05 0.80 0.80 0.40 1.00 0.95 0.75 1.00 0.55 B1 </li></ul><ul><li>0.55 0.95 0.75 0.75 0.40 1.05 0.90 0.70 0.95 0.50 B2 </li></ul><ul><li>0.50 1.05 0.80 0.80 0.45 1.00 0.95 0.80 1.05 0.85 C1 </li></ul><ul><li>0.55 1.00 0.80 0.80 0.50 1.05 0.95 0.80 1.05 0.80 C2 </li></ul><ul><li>Xbar & R Minitab Example </li></ul><ul><li>Using Aiag49:mtw Data File </li></ul><ul><li>Specification: 0.6 - 1.0 mm </li></ul><ul><li>Process Variation: 1.6 mm </li></ul><ul><li>Reference QS Measurement System Analysis Manual </li></ul>Gasket Thickness Study Six Sigma Champion Training
    35. 35. Six Sigma Champion Training
    36. 36. Six Sigma Champion Training Gage R&R Study for Thickness – XBar/R Method Source Variance StdDev 5.15*Sigma Total Gage R&R 2.08E-03 0.045650 0.235099 Repeatability 1.15E-03 0.033983 0.175015 Reproducibility 9.29E-04 0.030481 0.156975 Part-to-Part 3.08E-02 0.175577 0.904219 Total Variation 3.29E-02 0.181414 0.934282 Source %Contribution %Study Var %Tol %Process Total Gage R&R 6.332 25.164 58.77 14.69 Repeatability 3.509 18.733 43.75 10.94 Reproducibility 2.823 16.802 39.24 9.81 Part-to-Part 93.668 96.782 226.05 56.51 Total Variation 100.000 100.000 233.57 58.39 Number of distinct categories = 5
    37. 37. Calculation Explanation <ul><li>5.15 Sigma = 5.15  the factor standard deviation. 5.15 was developed empirically to approximate the gage population distribution variation. </li></ul><ul><li>% Contribution = Percent contribution of each factor based upon the variance. Repeatability = 100  repeatability variance/ total variation variance. </li></ul><ul><li>% Study Variation = 5.15  the factor standard deviation divided by 5.15  the total variation standard deviation. Repeatability = 100  5.15 repeatability standard deviation/ 5.15 total variation standard deviation. </li></ul><ul><li>% Tolerance = 5.15  the factor standard deviation divided by the tolerance. Repeatability = 100  5.15 repeatability standard deviation/tolerance. </li></ul>Six Sigma Champion Training
    38. 38. Calculation Explanation <ul><li>% Process Variation = 5.15 x the factor standard deviation divided by the process variation. Repeatability = 100 x 5.15 repeatability standard deviation/ process variation. </li></ul><ul><li>Number of Distinct Categories = part standard deviation divided by the total gage R&R standard deviation times 1.41. </li></ul>Six Sigma Champion Training
    39. 39. <ul><li>Metric is the % error against known population deemed good by local experts </li></ul><ul><li>Attribute legend can be the defect codes </li></ul><ul><li>If appraiser % is less than 100% training is required, focus on area of weakness </li></ul><ul><li>100% is the target for screen effectiveness </li></ul><ul><li>Use this to prove measurement system capability prior to task assignment </li></ul><ul><li>Select the 5 vital few (80-20 rule) to conduct GR&R </li></ul>Six Sigma Champion Training Attribute Data Example Using A_grr.xls file
    40. 40. Attribute Gage R & R Effectiveness Six Sigma Champion Training Instructions: 1) The following spreadsheet is used to calculate an Attribute GR&R Effectiveness, in which up to 30 samples can be evaluated, using 2 or 3 operators. 2) In the worksheet fill in the appropriate information in the Scoring Report section and enter the type of Attributes you are evaluating in the Attribute Legend section. YOU MUST ENTER THE INFORMATION IN THE ATTRIBUTE LEGEND SECTION OR THE SPREADSHEET WILL NOT WORK . The attributes can be either alpha or numeric, e.g. Yes, No; pass, fail; go, stop; or 1, 2. You must be consistent throughout the form and spell properly, anything will work as the spreadsheet compares what is in each cell. 3) If you or an expert has selected samples to be evaluated and you know what attributes these samples are, enter this information in the Attribute sample column. This will enable you to determine how well each operator can evaluate a set of samples against a known standard. You do not need to enter information in this column for the spreadsheet to work . 4) You do not have to specify how many operators or the # of samples that you will be evaluating during the test. Simply enter the data into the spreadsheet under the specific operator. Remember the attributes must be spelled properly or the spreadsheet will not analyze the data correctly. 5) To print a copy of the report click on the Print Report icon. 6) To delete the data in the spreadsheet, click on the Delete Data icon. 7) To delete all and begin a new test, click on the Delete All icon 8) To see a Demo of the Attribute GR&R Effectiveness spreadsheet, click on the Demo icon. Move around the spread sheet to see the data. When you are finished click the Delete All icon to delete all data to begin entering your own data.
    41. 41. Six Sigma Champion Training
    42. 42. Six Sigma Champion Training
    43. 43. Calculation Six Sigma Champion Training
    44. 44. Attribute Gage R&R Workshop <ul><li>Perform attribute Gage R&R study using M&M's. </li></ul><ul><ul><li>Determine the defects by looking for consistent color in the M&M's, clear markings (M's), and roundness. </li></ul></ul><ul><ul><li>Use 3 operators/inspectors </li></ul></ul><ul><li>Complete attribute GR&R analysis and report results (30 minutes). </li></ul><ul><li>Improve inspection criteria, rerun attribute Gage R&R study/analysis and report results (30 minutes). </li></ul>Six Sigma Champion Training
    45. 45. Do You Understand? <ul><li>The language of Measurement ? </li></ul><ul><li>The importance of Measurement? </li></ul><ul><li>How to perform a Gage R&R Study and how to interpret results ? </li></ul><ul><li>Use Minitab to analyze GR&R results? </li></ul>Six Sigma Champion Training ANY QUESTIONS OR COMMENTS

    ×