Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

NG BB 21 Intro to Minitab


Published on

Published in: Education, Technology
  • hi there would mind to send this presentation to my email address

    Are you sure you want to  Yes  No
    Your message goes here

NG BB 21 Intro to Minitab

  1. 1. UNCLASSIFIED / FOUO UNCLASSIFIED / FOUO National Guard Black Belt Training Module 21 Intro to Minitab UNCLASSIFIED / FOUO UNCLASSIFIED / FOUO
  2. 2. UNCLASSIFIED / FOUOCPI Roadmap – Measure 8-STEP PROCESS 6. See 1.Validate 2. Identify 3. Set 4. Determine 5. Develop 7. Confirm 8. Standardize Counter- the Performance Improvement Root Counter- Results Successful Measures Problem Gaps Targets Cause Measures & Process Processes Through Define Measure Analyze Improve Control TOOLS •Process Mapping ACTIVITIES • Map Current Process / Go & See •Process Cycle Efficiency/TOC • Identify Key Input, Process, Output Metrics •Little’s Law • Develop Operational Definitions •Operational Definitions • Develop Data Collection Plan •Data Collection Plan • Validate Measurement System •Statistical Sampling • Collect Baseline Data •Measurement System Analysis • Identify Performance Gaps •TPM • Estimate Financial/Operational Benefits •Generic Pull • Determine Process Stability/Capability •Setup Reduction • Complete Measure Tollgate •Control Charts •Histograms •Constraint Identification •Process Capability Note: Activities and tools vary by project. Lists provided here are not necessarily all-inclusive. UNCLASSIFIED / FOUO
  3. 3. UNCLASSIFIED / FOUO Learning Objectives  Become familiar with the Minitab statistical software  Learn some of the most widely used tools within Minitab  Learn how to navigate within Minitab UNCLASSIFIED / FOUO 3
  4. 4. UNCLASSIFIED / FOUO Overview FEAR NOT!  Minitab is a good tool for beginners, as well as established statisticians  This is just an introduction – Minitab is a powerful tool that is capable of much more than what we will have time to cover here  Suggestion: For now, do not be concerned with all the bells and whistles  Minitab Help function is always available and easy to use UNCLASSIFIED / FOUO 4
  5. 5. UNCLASSIFIED / FOUO What Will We Cover?  We are going to cover just a few of Minitab tools, including:  The Minitab layout  An overview of the menus  The Minitab Calculator  Some BASIC statistical tools  Some BASIC graphing tools  The HELP menu UNCLASSIFIED / FOUO 5
  6. 6. UNCLASSIFIED / FOUO Minitab Layout  The overall lay out of Minitab is similar to Excel, with some key differences:  The worksheet looks similar to a spreadsheet, but Minitab prefers data to be in columns. An easy way to think of it is to think of each column as a variable: Time, speed, date, etc.  The Session Window is a dialog window. It is where the non-graphical results of your calculations will be displayed. UNCLASSIFIED / FOUO 6
  7. 7. UNCLASSIFIED / FOUOWhat Does It Look Like? Session Window Worksheet UNCLASSIFIED / FOUO 7
  8. 8. UNCLASSIFIED / FOUO Minitab Characteristics  Although not a Microsoft product, Minitab shares some characteristics of other programs running on the Windows platform. In particular, the menu bar running across the top of the page, and the button bar below that. We will briefly go through some of the key functionality in the menu bar. The button bar contains shortcuts to many of the same functions.  Also, you can cut and paste data from Excel. You can do this by simply using the Ctrl-C and Ctrl-V functions you might already be familiar with. UNCLASSIFIED / FOUO 8
  9. 9. UNCLASSIFIED / FOUO Minitab Characteristics  One of the key differences in Minitab is the distinction between a Project and a Worksheet  Worksheets contain your data and nothing else. It is where you can manipulate the data and store your data. Minitab will let you have several worksheets open simultaneously. In addition, worksheets can be saved independently and shared between projects.  Projects are more like an entire notebook, containing not only one or more worksheets, but also your session window (containing any analysis you have already performed) and any graphs you have created UNCLASSIFIED / FOUO 9
  10. 10. UNCLASSIFIED / FOUOFile Menu Select File Project Commands Worksheet Commands Recently Opened Worksheets UNCLASSIFIED / FOUO 10
  11. 11. UNCLASSIFIED / FOUOEdit Menu Select Edit The most frequently used feature found under the Edit drop down menu is the Edit Last Dialog which can (and should) be quickly accessed by using the Control-E quick key combination. This feature will allow you to quickly access the last dialog box you used and either re-run, change, or review the last analysis Select you performed. Edit Last Dialog UNCLASSIFIED / FOUO 11
  12. 12. UNCLASSIFIED / FOUOData Menu Select Data The Data menu contains several Column manipulation features. Two examples are the Column Select Stack and Unstack features.Unstack Columns These features allow you toor Stack Columns combine multiple columns (or rows) into one, or break one column (or row) into multiple columns (or rows). UNCLASSIFIED / FOUO 12
  13. 13. UNCLASSIFIED / FOUO Pizza Example  Let‟s work through an example of this. Open the Minitab worksheet called Pizza Exercise.mtw.  We are going to stack the columns with the different defect types so that we can make a Pareto chart UNCLASSIFIED / FOUO 13
  14. 14. UNCLASSIFIED / FOUOPizza Example (Cont.) Step One: Open data set Pizza Exercise.mtw UNCLASSIFIED / FOUO 14
  15. 15. UNCLASSIFIED / FOUO Pizza Example (Cont.)  First, lets add a couple of headers for the columns we are about to stack  The first column we will call “Defect Type” and it will be where the defect types will be listed. We will put that header over column C11. The other column we will call “Count” and that is where we will put the count for each defect type. We will place that over column C12. UNCLASSIFIED / FOUO 15
  16. 16. UNCLASSIFIED / FOUOPizza Example (Cont.) Step 2: Type in the two new column headers C11 = Defect Type C12 = Count UNCLASSIFIED / FOUO 16
  17. 17. UNCLASSIFIED / FOUO Pizza Example (Cont.)  Now, let‟s go to the Data menu:  Data>Stack>Columns UNCLASSIFIED / FOUO 17
  18. 18. UNCLASSIFIED / FOUOPizza Example (Cont.) Data> Stack>Columns Step 3: Select Data, choose Stack and finally, select Columns UNCLASSIFIED / FOUO 18
  19. 19. UNCLASSIFIED / FOUO Pizza Example (Cont.)  We are going to stack all the columns that list the different defect types. That would include Late, Wrong Ingredients, Wrong Order, Wrong Address, and Damaged. Highlight all these columns and select them for the “Stack the following columns” box.  We have a few choices for where to put the new stacked columns. Remember we created two headers for the new columns: Defect Type and Count. Since these columns are on the current worksheet, we select the radio button labeled Column of current worksheet. This will open up two more fields for us to place the stacked columns.  In the first field, next to Column of current worksheet, select the “Count” column. Next, in the field labeled Store subscripts in, select the “Defect Type” column. Make sure that the box labeled Use variable names in subscript column is selected. Remember, Minitab often refers to columns as variables.  When done, click OK UNCLASSIFIED / FOUO 19
  20. 20. UNCLASSIFIED / FOUOPizza Example (Cont.) Data>Stack Columns Step 4 Highlight the columns you want stacked, then click “Select” to cause them to appear here Step 5 We want the data stored in the current worksheet, so select this Radio Button, enter data and click on OK UNCLASSIFIED / FOUO 20
  21. 21. UNCLASSIFIED / FOUO Pizza Example (Cont.)  There you go!  We will come back to this worksheet a little later and use these stacked columns to create a Pareto chart UNCLASSIFIED / FOUO 21
  22. 22. UNCLASSIFIED / FOUOPizza Example (Cont.) Data>Stack>Columns Your Stacked Data are now in columns C11-T & C12 UNCLASSIFIED / FOUO 22
  23. 23. UNCLASSIFIED / FOUO Calc Menu  The Calc menu contains a wealth of tools that you will be using on a regular basis. We will touch on them only briefly here.  Two such items are the Column Statistics and Row Statistics features. These items allow you to run simple calculations such as the sum, mean, median, and standard deviation on a row or column of data. The result will be displayed in the session window.  Two other commonly used items are the Make Patterned Data, and Random Data functions  Make Patterned Data will allow creating series, such as 1,2,3,1,2,3 etc. This is good for coding items such as days of the week.  Random Data can do a couple of things for you. Not only can you create a set of random numbers, but this function can also be used to randomize a column of data. UNCLASSIFIED / FOUO 23
  24. 24. UNCLASSIFIED / FOUOCalc Menu (Cont.) Select Calc To Get The Following Drop-Down Menu UNCLASSIFIED / FOUO 24
  25. 25. UNCLASSIFIED / FOUO Calc Menu (Cont.)  The Calculator is a very commonly used tool. The power behind the calculator is that it allows us to perform a wide range of mathematical and statistical functions not only on a column of data, but between columns of data. UNCLASSIFIED / FOUO 25
  26. 26. UNCLASSIFIED / FOUOThe Calculator The Calculator Dialog Box UNCLASSIFIED / FOUO 26
  27. 27. UNCLASSIFIED / FOUO The Calculator – Example Let’s Look at an Example:  Anthony‟s Pizza delivery company was interested in customer service problems with pizzas that they delivered. Over a 12-day period, they recorded how many pizzas they delivered, and how many had defects that fell into the following categories: Late, Wrong Ingredients, Wrong Order, Wrong Address, and Damaged. In addition, they recorded the “Pizza Stylist” who was making the pizzas each day.  When they were done collecting the data, they keyed it into the Minitab worksheet in Columns C4 thru C9. UNCLASSIFIED / FOUO 27
  28. 28. UNCLASSIFIED / FOUO The Calculator – Example (Cont.)  For this example, let‟s say that we want to know how many total defects we had each day. Since we know how many of each type of defect we have, all we need to do is add all of the defect columns together.  We will use the Calculator to calculate this result, and put the result in column C10  Let‟s start by typing the header for column C10 and calling it “Total Defects” UNCLASSIFIED / FOUO 28
  29. 29. UNCLASSIFIED / FOUOThe Calculator – Example (Cont.) Type in the newcolumn header for “Total Defects” in column C-10 UNCLASSIFIED / FOUO 29
  30. 30. UNCLASSIFIED / FOUO The Calculator – Example (Cont.)  Now we can have some fun!  We want to store the result in our new variable “Total Defects”. We can select this variable in two ways. We can select the column from the list on the left, then click the Select button. Or we can simply double click the column from the list.  It is important to note that to select an item for a particular function, whether it be the Store the Result in Variable box, or Expression box, the cursor MUST be in the box you are selecting (or else you might not see the variables you want to choose).  Next, we want to create the equation. We would do this exactly the same as we would using a calculator.  In the expression box, select Late, then the “+” sign, then Wrong Ingredients, “+” Wrong Order, etc., until you have selected all the variables you want to add together. We have already told Minitab where we want the result stored in the previous step. When you are done, hit OK. UNCLASSIFIED / FOUO 30
  31. 31. UNCLASSIFIED / FOUOThe Calculator – Example (Cont.) Tell Minitab where to store the results C10 or Total Defects Put the formula to use here: Late + Wrong Ingr. + Wrong Order + Wrong Address + Damaged Click On OK UNCLASSIFIED / FOUO 31
  32. 32. UNCLASSIFIED / FOUO The Calculator – Example (Cont.)  Voila!  Column C10 (Total Defects) is now populated with the data you wanted UNCLASSIFIED / FOUO 32
  33. 33. UNCLASSIFIED / FOUOThe Calculator – Example (Cont.) UNCLASSIFIED / FOUO 33
  34. 34. UNCLASSIFIED / FOUO Stat Menu  The Stat menu is where the majority of Minitab‟s functionality resides (such as Basic Statistics, Regression, and Control Charts)  We are not going to get into a lot of this at this time, but we will be coming back to visit Basic Statistics and Pareto charts shortly UNCLASSIFIED / FOUO 34
  35. 35. UNCLASSIFIED / FOUO Stat Menu (Cont.) Select Stat UNCLASSIFIED / FOUO 35
  36. 36. UNCLASSIFIED / FOUO Graph Menu  The Graph menu contains many, many graphs that can be quite helpful, including Time Series plots, Histograms, and Boxplots. It should be noted however, that these graphs are also available in other places while you are using other tools.  After selecting a graph type, for many of the graphs, you will get a Dialog Box which illustrates the various styles available for the specific graph you have chosen. (There are some types of graphs that have only a single style.) UNCLASSIFIED / FOUO 36
  37. 37. UNCLASSIFIED / FOUO Graph Menu (Cont.) Select Graph drop down menu If you choose Dotplots you would then get another dialog box showing the various types of Dotplots Select Cancel UNCLASSIFIED / FOUO 37
  38. 38. UNCLASSIFIED / FOUO Editor Menu  The Editor menu contains many features you might already be familiar with due to your exposure with Excel and other related products. Such features include Find and Replace. Selecting Editor UNCLASSIFIED / FOUO 38
  39. 39. UNCLASSIFIED / FOUO Window Menu  The Window menu covers several features that are common to other Windows-based programs, such as Tile, and Cascade Selecting Window This area shows currently open Windows UNCLASSIFIED / FOUO 39
  40. 40. UNCLASSIFIED / FOUO Help Menu  The Help menu is similar to other help menus you have probably seen  The main difference is that (as you may have noticed) Help is also available from all the dialog boxes  In addition, Minitab has a built in Stat Guide. The Stat Guide is more centered around the how and why you use a particular tool and assistance in interpreting the results.  The traditional Help menu is more centered around how to use the software UNCLASSIFIED / FOUO 40
  41. 41. UNCLASSIFIED / FOUO Help Menu (Cont.) Selecting Help UNCLASSIFIED / FOUO 41
  42. 42. UNCLASSIFIED / FOUO Basic Statistics  Let‟s work through our Pizza example to see how to use Minitab to get some basic information about a process UNCLASSIFIED / FOUO 42
  43. 43. UNCLASSIFIED / FOUO Basic Statistics (Cont.)  Under Stat>Basic Statistics, there is a small feature that can provide a wealth of information. We use it often to get a quick snapshot of the data underlying the process we are looking at.  Let‟s say we want to examine the number of defects our pizza company has been experiencing over the past 12 days. Now that we are back at our desk, we are excited to do something with this data to give us an idea how the process is doing. We might start by going to: Stat>Basic Statistics>Graphical Summary UNCLASSIFIED / FOUO 43
  44. 44. UNCLASSIFIED / FOUO Basic Statistics (Cont.) Stat>Basic Statistics> Graphical Summary We want the Graphical Summary UNCLASSIFIED / FOUO 44
  45. 45. UNCLASSIFIED / FOUO Basic Statistics (Cont.)  We select the variable we wish to examine – in this case we are looking at “Total Defects”  Now, since a picture is worth a hundred words, let‟s have Minitab show us some graphs along with our data analysis  Click the OK button UNCLASSIFIED / FOUO 45
  46. 46. UNCLASSIFIED / FOUO Basic Statistics (Cont.) Stat>Basic Statistics> Graphical Summary Double Click on “Total Defects” to add it to the Variables Box Click on OK UNCLASSIFIED / FOUO 46
  47. 47. UNCLASSIFIED / FOUO Basic Statistics (Cont.)  To get a good statistical overview of the data, let‟s take a look at the Graphical Summary. This summary will include several graphs along with summary statistics and other useful information. UNCLASSIFIED / FOUO 47
  48. 48. UNCLASSIFIED / FOUOBasic Statistics (Cont.) Stat>Basic Statistics>Graphical Summary Summary for Total Defects A nderson-D arling N ormality Test A -S quared 0.49 P -V alue 0.174 M ean 10.667 S tD ev 6.110 V ariance 37.333 S kew ness 1.00674 Kurtosis 0.40899 N 12 M inimum 4.000 1st Q uartile 6.000 M edian 8.500 3rd Q uartile 14.750 5 10 15 20 25 M aximum 24.000 95% C onfidence Interv al for M ean 6.784 14.549 95% C onfidence Interv al for M edian 6.000 14.737 95% C onfidence Interv al for S tD ev 95% Confidence Intervals 4.328 10.374 Mean Median 5.0 7.5 10.0 12.5 15.0 UNCLASSIFIED / FOUO 48
  49. 49. UNCLASSIFIED / FOUO Basic Statistics (Cont.)  First, let‟s look at the graphs that Minitab gives us  The left side of the graph includes the “pictures” including (from top to bottom):  Histogram – Sorts the data into groups, then plots the frequency of each group along an X-axis and overlays a „Normal‟ distribution  Boxplot of the data – Breaks the data into quartiles (i.e., 50% of our data is inside the box)  95% Confidence Intervals – For our Mean and Median  The right side of the page contains the calculated data. The main items you will be using include: Mean Median Standard Deviation Minimum Maximum  In addition there is also a Normality Test and Quartile Information UNCLASSIFIED / FOUO 49
  50. 50. UNCLASSIFIED / FOUO Pizza Example Analysis But what does this mean for our Pizza example? Let’s look at it piece by piece:  Histogram – This is showing us that our number of defects vary between 4 and 24 per day. This tells us that there is a great deal of variation in our process.  Boxplot of the Data – This shows us the same variation, but with some additional information. For example, we can see that 50% of the pizzas examined in this time period had between 6 and 14.75 defects, but with an additional 25% having more than 14.75 defects. The right side of the page contains some detail about our process as well. For example:  The Mean (or average) number of defects on a given day is 10.7 defects, but the standard deviation is over 6 defects. In other words, on average we create over 10 defects per day, and the variation (as measured by the Standard Deviation) further amplifies the problem.  Our Median (the middle value in our data set) is 8.5 defects and the range of our data (another measure of variation calculated as Max - Min) is a whopping 20 defects! UNCLASSIFIED / FOUO 50
  51. 51. UNCLASSIFIED / FOUO Pizza Example Analysis (Cont.)  What if you do not need the graphs and you just want the data analysis? You can choose just the data analysis by selecting Display Descriptive Statistics. This analysis will show up in your session window. Highlights include:  N = Sample size  Mean = Statistical average of the sample  Median = The “middle point” of the sample  Standard Deviation = The “average” amount that each data point deviates from the mean  Maximum = The largest data point in the sample  Minimum = The smallest data point in the sample UNCLASSIFIED / FOUO 51
  52. 52. UNCLASSIFIED / FOUOBasic Statistics (Cont.) Stat>Basic Statistics> Display Descriptive Statistics Select Display Descriptive Statistics for data analysis without the graphs UNCLASSIFIED / FOUO 52
  53. 53. UNCLASSIFIED / FOUOBasic Statistics (Cont.) Stat>Basic Statistics> Display Descriptive Statistics Double click on C-10, “Total Defects” to add it to the Variables box Click on OK to get your analysis in the Session Window UNCLASSIFIED / FOUO 53
  54. 54. UNCLASSIFIED / FOUOBasic Statistics (Cont.) Stat>Basic Statistics>Display Descriptive Statistics Here is the analysis in the Session Window UNCLASSIFIED / FOUO 54
  55. 55. UNCLASSIFIED / FOUOBasic Statistics (Cont.) Stat>Basic Statistics> You can choose which statistics are Display Descriptive Statistics included by selecting the statistics tab and checking just the statistics you want Select Cancel UNCLASSIFIED / FOUO 55
  56. 56. UNCLASSIFIED / FOUO Pizza Example Analysis (Cont.)  Now lets say we want to compare Total Defects by Pizza Stylist  We can do this in the same way we ran Basic Statistics> Graphical Summary on one variable, but now we will be examining Multiple Variables all in one step  The only difference here is that we need is to select a “By Variable” that we would like to compare by. In this case, that would be by “Pizza Stylist.”  Once again, click OK UNCLASSIFIED / FOUO 56
  57. 57. UNCLASSIFIED / FOUOPizza Example Analysis (Cont.) Stat>Basic Statistics> Graphical Summary Select “Total Defects” from the list and double click on it to place it in the Variables box Select “Pizza Stylist” for the By Variables box Click On OK UNCLASSIFIED / FOUO 57
  58. 58. UNCLASSIFIED / FOUO Pizza Example Analysis (Cont.) Stat> Basic Statistics> Graphical Summary This gives you the Graphical Summary for all three Pizza Stylists in One Analysis UNCLASSIFIED / FOUO 58
  59. 59. UNCLASSIFIED / FOUO Pizza Example Analysis (Cont.)  Minitab will give us the same graphs we requested before, but this time, we will get one for each pizza stylist (which we have tiled for this screen shot). This allows us the opportunity to compare each stylist side by side.  Again, by looking at the graphs of the pizza stylists, we learn a lot very quickly, such as:  Pizza Stylist Bob has much less variation in his process. This can be seen in the histogram (which is not as spread out as the other stylists), box plot (which is also much tighter than the other stylists), and the range of the data (8 defects for Bob, 10 Defects for Sally, and 18 defects for Frank).  At this point, we can tell that stylist Bob appears to be more consistent in minimizing defects. But one thing that is important to note is that we do not know why. Are they following the same process? Is Bob cutting corners?  Also, it should be noted here, that our sample size for this example is too small to draw statistically significant conclusions UNCLASSIFIED / FOUO 59
  60. 60. UNCLASSIFIED / FOUO Pizza Example Analysis (Cont.)  Similar to the Graphical Summaries, Minitab will layout the results for each Pizza Stylist side by side in the session window. For this we have to select the option Display Descriptive Statistics. Once again, we can compare the means, standard deviations, and ranges for these two associates in fairly short order.  In addition to the difference in the ranges between the stylists, one sure sign that Bob has less variation in his process is his standard deviation – it is much less than that of the other stylists. UNCLASSIFIED / FOUO 60
  61. 61. UNCLASSIFIED / FOUO Pizza Example Analysis (Cont.) Stat>Basic Statistics> Display Descriptive Statistics Multiple Group Data Output In Session Window Note: Remember, you get to choose what statistics are included in this analysis UNCLASSIFIED / FOUO 61
  62. 62. UNCLASSIFIED / FOUO Boxplots  Another useful item that we can have Minitab create are Boxplots  In addition to being a part of several of Minitabs other tools, you can find Boxplots under the Graph menu  The next slide illustrates the Nomenclature and Various Parts of a Boxplot UNCLASSIFIED / FOUO 62
  63. 63. UNCLASSIFIED / FOUO Boxplot Terminology 4.5 Whisker shows range for the top 25% of the data points. Outlier 3.5 Third Quartile Line Median Line 2.5 First Quartile Line 1.5 1 2 Whisker shows range for the Inter Quartile Range lower 25% of the data points. (IQR or Box Length) Inter Quartile Range (IQR) = The distance between the Third Quartile Line and the First Quartile Line. This includes 50% of your data. Simply stated: IQR = Q3 – Q1. Outlier = A data point is considered an outlier if it is more than 1.5 x IQR above the Third Quartile Line or below the First Quartile Line UNCLASSIFIED / FOUO 63
  64. 64. UNCLASSIFIED / FOUO Boxplots (Cont.)  Once again, Boxplot is an area with lots of bells and whistles, we are just going to create a simple boxplot  We do this by first selecting Graph>Boxplot  For the type of Boxplot, select One Y and With Groups, since we have three groups (or boxes) on the same Y axis  In the final dialog box select “Total Defects” for the Graph Variable and choose “Pizza Stylist” for the By Variable  Click OK when done UNCLASSIFIED / FOUO 64
  65. 65. UNCLASSIFIED / FOUO Boxplots (Cont.) Select Graph Select Boxplot UNCLASSIFIED / FOUO 65
  66. 66. UNCLASSIFIED / FOUOBoxplots (Cont.) From Next Dialog Box Select One Y and With Groups Click On OK UNCLASSIFIED / FOUO 66
  67. 67. UNCLASSIFIED / FOUOBoxplots (Cont.) Select “Total Defects” for the Group Variable Use “Pizza Stylist” for the By Variable Click On OK UNCLASSIFIED / FOUO 67
  68. 68. UNCLASSIFIED / FOUOBoxplots (Cont.) Note: There are no Outliers (asterisks) in this data set UNCLASSIFIED / FOUO 68
  69. 69. UNCLASSIFIED / FOUO Boxplots (Cont.) A boxplot is a very good tool for graphically representing variation between items you are trying to compare  The Box Portion of the graph represents where 50% of the data for that item lies, with the line in the box representing the median for that item  The lines on either side of the box are called Whiskers, and represent the range for the Upper 25% and Lower 25% of the data. An asterisk outside of the whiskers is commonly referred to as an Outlier. An outlier is a data point flagged as unusual, usually because it is significantly outside the range that Minitab would be expecting to see your data.  As we can see from these boxplots, stylists Sally and Frank have significantly more variation in their process, as well as a larger median UNCLASSIFIED / FOUO 69
  70. 70. UNCLASSIFIED / FOUO Time Series Plot  Another useful chart we often run is called a Time Series Plot (also commonly referred to as a run chart). As the name would suggest, this is a useful tool for displaying how a process is operating over time.  We find this tool under the Graph menu, labeled “Time Series Plot” UNCLASSIFIED / FOUO 70
  71. 71. UNCLASSIFIED / FOUO Time Series Plots (Cont.) Select Graph Select Time Series Plot UNCLASSIFIED / FOUO 71
  72. 72. UNCLASSIFIED / FOUO Time Series Plots (Cont.)  With Time Series Plots there are many places for us to tweak the format of the output and add enhancements to the graph, but for our example, we just want a plain and simple Run Chart  We do this by simply selecting “Total Defects” as the Series Variable. In addition, if we have a Date column (as we do here), we can place that under Time/ Scale.  The steps needed for a Time Series Plot are shown on the next few slides UNCLASSIFIED / FOUO 72
  73. 73. UNCLASSIFIED / FOUO Time Series Plots (Cont.) 1. Select Graph and Time Series Plot 2. Select Simple and Click on OK UNCLASSIFIED / FOUO 73
  74. 74. UNCLASSIFIED / FOUOTime Series Plots (Cont.) 3. Add “Total Defects” in the Series box 4. Click on Time/Scale to add Dates and then click on OK UNCLASSIFIED / FOUO 74
  75. 75. UNCLASSIFIED / FOUOTime Series Plots (Cont.) 5. Select Radio Button for Stamp 6. Add “Date” to Stamp Columns box and click on OK UNCLASSIFIED / FOUO 75
  76. 76. UNCLASSIFIED / FOUO Time Series Plots (Cont.)  And here we have it!  This shows what is happening to our process as time goes on. Of course, in order for this graph to be of any value, the data must be entered in time order.  In our pizza example here, what seems to beg questioning is “what happened in the period from 9/4 to 9/7 – particularly 9/6 & 9/7?”  Can you see any other patterns in this data? Hint: 9/1 is a Monday. UNCLASSIFIED / FOUO 76
  77. 77. UNCLASSIFIED / FOUOTime Series Plots (Cont.) UNCLASSIFIED / FOUO 77
  78. 78. UNCLASSIFIED / FOUOTime Series Plots (Cont.) If you hold your Mouse Cursor over any data point, you will get the corresponding X and Y data values for that point. You can still use the Brush feature to highlight several data points to see what Rows the data come from. UNCLASSIFIED / FOUO 78
  79. 79. UNCLASSIFIED / FOUO Pareto Charts  Another useful tool for deciding what areas for improvement should be addressed first are Pareto Charts  We can find this tool under Stat>Quality Tools>Pareto Chart  The steps needed to do a Pareto Chart for the pizza data are shown on the next few slides UNCLASSIFIED / FOUO 79
  80. 80. UNCLASSIFIED / FOUO Pareto Charts (Cont.) 1. Select Stat> Quality Tools> Pareto Chart UNCLASSIFIED / FOUO 80
  81. 81. UNCLASSIFIED / FOUO Pareto Charts (Cont.)  Remember, we stacked our data into the columns “Defect Type” and “Count.” These two columns make up a Defects table that can now be turned into a Pareto Chart.  To do this, we click the radio button labeled Chart defects table and fill in the two fields that are now available to us. For the first one, our Labels are stored in the “Defect Type” column. Once you have selected this for the Labels field, we need to tell Minitab where our frequency data resides. In this case it is in the column we called “Count” so select that column. When you have done that, click OK. UNCLASSIFIED / FOUO 81
  82. 82. UNCLASSIFIED / FOUOPareto Charts (Cont.) 2. Select Radio Button for Chart defects table 3. Put “Defects Type” in Labels in box and put “Count” in the Frequencies in box and click on OK UNCLASSIFIED / FOUO 82
  83. 83. UNCLASSIFIED / FOUO Pareto Charts (Cont.)  Given the limited resources available for us to address process improvements, it is important that we concentrate those resources on the largest problem areas first  From our Pareto Chart, we can see in short order that Late Deliveries make up 43.0% of the total defects we are experiencing, with Wrong Ingredients making up an additional 29.7%  This is telling me that the biggest impact on defects will be made by addressing the portion of the process resulting in Late Deliveries and (resources and scope permitting) Wrong Ingredients UNCLASSIFIED / FOUO 83
  84. 84. UNCLASSIFIED / FOUO Pareto Charts (Cont.) UNCLASSIFIED / FOUO 84
  85. 85. UNCLASSIFIED / FOUOSaving and Exiting Saving as a Project will save Saving as a Worksheet the worksheet, session window, will save only the worksheet report pad, and all graphs (Data Set) UNCLASSIFIED / FOUO 85
  86. 86. UNCLASSIFIED / FOUO Takeaways  Minitab is a powerful tool that, with just a few simple mouse clicks, can tell you a great deal about your data including: Numerical Graphical  Mean  Histogram  Median  Box Plots  Minimum  Time Series Plots  Maximum  Run Charts  Range  Pareto Charts  Standard Variation UNCLASSIFIED / FOUO 86
  87. 87. UNCLASSIFIED / FOUO What other comments or questions do you have? UNCLASSIFIED / FOUO
  89. 89. UNCLASSIFIED / FOUO Minitab Exercise  This exercise concerns a Customer Service Center. It utilizes the basic Minitab tools that you have learned and allows you to play with some basic options.  AHT = Average Handle Time (The average number of seconds required to complete a call)  Occupancy = The percentage of time that Customer Service Reps are on the phone or completing after call work relative to the amount of time they are logged into the system. Generally the target is 80-85%.  Speed of Answer = The average number of seconds it takes to answer an inbound call. Targets vary depending on the required service level but generally 30 seconds is standard.  Call Quality Score = “Your call may be recorded for quality purposes.” This really happens. Calls are recorded for each rep (usually 10-15 per month) and scored by a dedicated call monitoring team. It is a critical measure for individual reps. UNCLASSIFIED / FOUO 89
  90. 90. UNCLASSIFIED / FOUO Minitab Exercise  Using the dataset Call Center Data.xls: 1.Copy the data into Minitab 2.Display the distribution of AHT (Average Handle Time) for the business. What is the mean? 3.On a single graph, display the AHT for sites A, B, and C. Which site has the least variation? 4.Create the appropriate graph which shows the frequency of call types. Which call type occurs most often and what percentage of the overall call volume does it represent? 5.Display the relationship between Occupancy (X) and Speed Of Answer (Y). What is the nature of the relationship? 6.On a single graph, display the monthly Average Call Quality Score for Sites A, B, and C. Overall, which site has the highest scores? UNCLASSIFIED / FOUO 90