For Additional Information... The materials in this handbook were developed by Master Black Belts at General Electric Medical Systems to assist Black Belts and Green Belts in completing Minitab Analyses. It is assumed that the user has a basic understanding of statistical tools. Please feel free to contact us if you would like additional copies of this or other Six Sigma material, or have any questions, comments, or suggestions. Stephanie Spencer 414.548.4673 8* 320.4673 GEMS-Am Six Sigma Office 414.548.5111 8* 320.5111 1 Rev 1
Overview of Key Changes from Minitab Version 10 Added Features: More Info? Added Six Sigma 밚 1?Analysis Six Sigma 밚 1?& 밚 2?as Drop Down Menu Items Gage R&R Studies as Drop Down Menu Items New DoE Interface Improved Mathematical Manipulations with 밅 alculator Graphing Options Added to Many Analysis Dialog Boxes New Probability Plot for Lognormal, Weibull and Exponential Distributions Fitted Line Plots Include Options for Quadratic and Cubic as well as Linear Box-Cox Transformations Toolbars (Windows 95 only) Insert and Delete Columns Launch Minitab and Open Files from File Manager or Explorer Session Window Does Not Display Command Language Unless Selected Longer File Names (Windows 95 only) 18 18,22 28 76 10 x 62 x 58 x 90 x 24 7 Minitab Help 2 Rev 1
Table of Contents The Basics -- PAGE Minitab Windows 6 Toolbars 7 Performing Calculations 10 Working with Data -- Changing Data Type 14 Stacking Data / Data Blocks 16 Creating Patterned Data 20 Re-coding Data 22 Transforming Data 24 Gather Tools -- Gage R&R 28 The 밚1?-- Product Report 32 The 밚2?-- Process Report 34 Graphing -- Statistical Problem Description 40 Basic Plot 46 Graph Brushing & Editing 48 Copying to Other Applications 51 Grouping Variables 52 Multi-vari Plots 54 Box Plots 56 Normal Probability Plots 58 Statistical Tests -- One Sample t-tests 62 Two Sample t-tests 64 Homogeneity of Variance 66 Analysis of Variance 68 Chi Squared 70 DoE -- Create Factorial DoE Design 76 Analyze Factorial DoE Design 78 Analyze Custom Factorial DoE Design 80 Main Effects Plot 82 Interaction Plots 84 Regression -- Regression 88 Fitted Line Plots 90 Residuals Analysis 92 3 Rev 1
The Basics -- s Minitab Windows Toolbars Performing Calculations 5 Rev 1
Minitab Basics Minitab Windows Menu Bar Session Window: Analytical Output Data Window: A Worksheet, not a Spreadsheet Column names are above first row Everything in a column is considered to be the same variable Info Window: Synopsis of worksheet History Window: Stores Commands Four Interactive Windows. Only One Open at a Time. Windows Saved Separately. 6 Rev 1
Toolbars -- Windows 95 The Data Window Toolbar Open File Save File Print Window Manage Graphs Insert Cells Insert Rows Close Graphs Insert Columns Cancel Move Columns Clear Cells Cut Copy Paste Undo Previous Brushed Row Next Brushed Row Last Dialog Box Session Window Data Window Help These commands can also be found in drop down menus, or accessed with shortcut keys. 7 Rev 1
Toolbars -- Windows 95 The Session Window Toolbar Open File Save File Print Window Previous Command Next Command Find Find Next Manage Graphs Close Graphs Cancel Cut Copy Paste Undo Last Dialog Box Session Window Data Window Help These commands can also be found in drop down menus, or accessed with shortcut keys. 8 Rev 1
Toolbars -- Windows 95 The Graph Window Toolbar Open File View Mode Manage Graphs Save File Print Window Edit Mode Close Graphs Brush Mode Cancel Cut Copy Paste Undo Last Dialog Box Session Window Help Data Window These commands can also be found in drop down menus, or accessed with shortcut keys. 9 Rev 1
Gather Tools -- s Gage R&R The 밚1?-- Product Report The 밚2?-- Process Report 11 Rev 1
Gage R&R Setting it up -- Select: Stat > Quality Tools > Gage R & R Studies Enter columns containing data for : Enter tolerance range. Click OK (twice) to get results. 12 Rev 1
Gage R&R The Graphical Output -- How much variation is coming from the parts? How do the average readings for each operator compare? How do the distribution of readings for each operator compare? Gage R&R (ANOVA) for Measure Gage name: Date of study: Reported by: Tolerance: Misc: Sample Mean 1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0 Xbar Chart by Operator 1 2 3 3.0SL=0.8796 X=0.8075-3.0SL=0.7354 Average 1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 Part ID 1 Operator*Part Interaction 2 3 4 5 6 7 8 9 10 Operator 1 2 3 How much difference does each operator see between 1st and 2nd readings? Sample Range Percent 0.15 0.10 0.05 0.00 100 90 80 70 60 50 40 30 20 10 0 R Chart by Operator 1 2 3 0 Components of Variation Gage R&R Repeat Reprod Part-to-Part 3.0SL=0.1252 R=0.03833-3.0SL=0.000 %Total Var %Study Var %Toler 1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 Oper ID 1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 Part ID 1 1 2 By Operator 2 By Part 3 4 5 6 7 8 9 3 10 What percentage of the total variation is coming from the gage? Is repeatability or reproducibility the issue? How much variation do we see in readings for the same part? 13 Rev 1
Gage R&R What Contained in the Session Window Output? First Table» ANOVA table Shows whether part, operator or part*operator interaction are major contributors to variations in the data. Look for p-values <.05. Second Table» Variance components» Standard deviations» A constant multiple of the standard deviations, usually 5.15*sigma Third Table 99% of the area under a curve is within an interval 5.15 standard deviations wide. This number is also called the study variation and used to estimate how wide an interval one would need to capture 99% of the measurements from a process.» % Contribution to total variation made by each variance component Each component is divided by the total variation then multiplied by 100.» % Study variation Standard Deviation of each component is divided by the total Standard Deviation. Total WILL NOT sum to 100.» % Tolerance Enter tolerance range (Upper limit - Lower limit) under options, if desired. 14 Rev 1
Gage R&R The Session Window Output -- Gage R&R Study - ANOVA Method ANOVA Table With Operator*Part Interaction Source DF SS MS F P Parts 9 2.05871 0.228745 39.7178 0.00000 Operators 2 0.04800 0.024000 4.1672 0.03256 Oper*Part 18 0.10367 0.005759 4.4588 0.00016 Repeatability 30 0.03875 0.001292 Total 59 2.24912 Gage R&R Source VarComp StdDev 5.15*Sigma Total Gage R&R 0.004437 0.066615 0.34306 Repeatability 0.001292 0.035940 0.18509 Reproducibility 0.003146 0.056088 0.28885 Operator 0.000912 0.030200 0.15553 Oper*Part 0.002234 0.047263 0.24340 Part-To-Part 0.037164 0.192781 0.99282 Total Variation 0.041602 0.203965 1.05042 What are major contributors? Look for p-values <.05. Estimate of interval needed to capture 99% of measurements. Source %Contribution %Study Var %Tolerance Total Gage R&R 10.67 32.66 34.31 Repeatability 3.10 17.62 18.51 Reproducibility 7.56 27.50 28.89 Operator 2.19 14.81 15.55 Oper*Part 5.37 23.17 24.34 Part-To-Part 89.33 94.52 99.28 Total Variation 100.00 100.00 105.04 % Gage R&R. Ideal is < 10 % of tolerance. Number of Distinct Categories = 4 Distinct Categories the measuring system can distinguish. If less than 2, measurement system can 뭪 distinguish. Two is go / no go. Need 4 for good system. 15 Rev 1