Graphing using Minitab

45

DOT SIZE

40

35

5 6 7 8
LINE

L. Goch – February 2011

AGENDA
 Why Graph Data?

 Under STAT
All Minitab Tutorial
 Run Chart
Worksheets are located
 Pareto Chart in the folder ‘C:\Program
 Multi-Vari Chart Files\Minitab
 Under GRAPH 16\English\Sample Data’
 Scatterplot
 Histogram
 Boxplot
 Individual Value Plot
 Bar Chart
 Pie Chart
 3D Scatterplot
WHY GRAPH THE DATA?
 Graphs help us understand the nature of variation
 Graphs make the nature of data more accessible to
the human mind
 Graphs help display the context of the data

 Graphs should be the primary presentation tool in

data analysis
 If you can’t show it graphically, you probably
don’t have a good conclusion
 Graphs help separate the signal from the noise

Graphical Analysis is also Called

DATA MINING!
Source: Donald Wheeler: Understanding Variation
 RULES FOR EFFECTIVE DATA COLLECTION

 Team must follow sampling plan consistently

 Do a short Pilot Run to test your procedures
 Note changes in operating conditions that are not
part of the normal or initial operating conditions
 Maintain monitors on gauges for key process
inputs
 Record any events that are out of the ordinary
 Log data into database quickly
 Keep a log book
AVAILABLE GRAPH TOOLS
RUN CHART:

STAT > QUALITY TOOLS >

RUN CHART
RUN CHART: STAT > QUALITY TOOLS > RUN CHART
 Tests for Process Stability by applying some
statistical diagnostic tests to the series

Open worksheet Radon.mtw

 RUN CHART

Run Chart of Membrane

45

40
Membrane

35

30

25

20
1 2 3 4 5 6 7 8 9 10
Sample

Number of runs about median: 3 Number of runs up or down: 5

Expected number of runs: 6.0 Expected number of runs: 6.3
Longest run about median: 5 Longest run up or down: 3
Approx P-Value for Clustering: 0.022 Approx P-Value for Trends: 0.135
Approx P-Value for Mixtures: 0.978 Approx P-Value for Oscillation: 0.865
PARETO CHART:

STAT > QUALITY TOOLS >

PARETO CHART
PARETO CHART: STAT > QUALITY TOOLS >
PARETO CHART  Pareto Charts are an essential
tool to help prioritize
Open worksheet
improvement targets
EXH_QC.MTW
 Pareto’s allow us to focus on
the 20% of the problems that
cause 80% of the poor
performance
 Defects Counts
Missing Screws 274
PARETO CHART Missing Clips 59
Defective Housing 19
Leaky Gasket 43
Pareto Chart of Defects Scrap 4
Unconnected Wire 8
300
Missing Studs 6
250
Incomplete Part 10

200
C ounts

150

100

50

0
Defects
Missing Screws

Missing Clips

Missing Studs
Leak y Gask et

Incomplete Part
Defectiv e Housi

Unconnected Wir

Scrap
Counts 274 59 43 19 10 8 6 4
Percent 64.8 13.9 10.2 4.5 2.4 1.9 1.4 0.9
Cum % 64.8 78.7 88.9 93.4 95.7 97.6 99.1 100.0
SECOND LEVEL PARETOS
 We can generate a second level Pareto using the By
statement
 This breaks down the overall Pareto by time of day
 SECOND LEVEL PARETO
Flaws Period
Scratch Day
Scratch Day
Peel Day
Peel Day
Smudge Day
Scratch Day
Other Day Pareto Chart of Flaws by Period
Other Evening
Peel Evening Period = Day Period = Evening F law s
Peel Evening 8 P eel
Peel Evening S cratch
Peel Evening O ther
Scratch Evening 6 S mudge
Scratch Evening
Peel Night 4
4
Scratch Night 3
Smudge Night 2 2
Scratch Night 2
Peel Night 1 1 1
Count

Peel Night 0
0
Peel Night Period = Night Period = Weekend
Peel Night 8
8
Other Night
Other Night
6
Scratch Night 6
Scratch Night
Peel Night
4
Scratch Night 3 3 3
Smudge Night
2
Scratch Night 2
Other Night 1
Scratch Night 0
Scratch Night 0
P eel S cratch O ther S mudge P eel S cratch O ther S mudge
Peel Weekend
Peel Weekend Flaws
Peel Weekend
Smudge Weekend
Smudge Weekend
Smudge Weekend
Other Weekend
MULTI-VARI CHART:

STAT > QUALITY TOOLS >

MULTI-VARI CHART
MULTI-VARI CHART: STAT > QUALITY TOOLS >
MULTI-VARI CHART
 Multi-vari charts are a way
Open worksheet of presenting analysis of
Sinter.MTW variance data in a graphical
form. The chart displays the
means at each factor level
for every factor.
MULTI-VARI CHART
Multi-Vari Chart for Strength by SinterTime - MetalType
24 S interTime
100
150
23 200

22

21
Strength

20

19

18

17
1 2 3
MetalType
SCATTERPLOT:

GRAPH > SCATTERPLOT

SCATTERPLOT: STAT > SCATTERPLOT
 Scatterplots study the relationship between
two variables
Open worksheet
Batteries.MTW
 SCATTERPLOT
Scatterplot of FlashRecov vs VoltsAfter
7.5

7.0

6.5

6.0
FlashRecov

5.5
5.25
5.0

4.5

4.0

3.5
0.9 1.0 1.1 1.2 1.3 1.4 1.5
VoltsAfter
 SCATTERPLOT – BY A VARIABLE
Scatterplot of FlashRecov vs VoltsAfter
7.5 F ormulation
N ew
O ld
7.0

6.5

6.0
FlashRecov

5.5
5.25
5.0

4.5

4.0

3.5
0.9 1.0 1.1 1.2 1.3 1.4 1.5
VoltsAfter
HISTOGRAM:

GRAPH > HISTOGRAM

CREATING A HISTOGRAM WITH A NORMAL CURVE
 Graph > Histogram > With Fit
 Histograms examine the shape and
spread of data

Open worksheet
Camshaft.MTW
 SMOOTHED (NORMAL) DISTRIBUTION
Histogram of Length
Normal
Mean 600.1
25 StDev 1.335
N 100

20
Frequency

15

10

0
597 598 599 600 601 602 603
Length

We can view the data as a smoothed distribution (red line), in this example
using the “normal distribution” assumption. It provides an approximation
of how the data might look if we were to collect an infinite number of data
points. DOES THE DATA FIT THE CURVE??? If not, does another type of
distribution fit the data?
 SMOOTHED (SKEWED) DISTRIBUTION
Histogram of Length
Smallest Extreme Value
Loc 600.7
25 Scale 1.068
N 100

20
Frequency

15

10

0
596 597 598 599 600 601 602
Length

We can view the data as a smoothed distribution (red line), in this example
using the “skewed distribution” assumption. It provides an approximation
of how the data might look if we were to collect an infinite number of data
points. DOES THE DATA FIT THE CURVE??? If not, look for groups that
may explain the shape of the data?
CREATING A HISTOGRAM WITH GROUPS
 Graph > Histogram > With Outline
and Groups
 Data for the 2 different suppliers is
available.
Still using worksheet
Camshaft.MTW
 SMOOTHED (SKEWED) DISTRIBUTION
Histogram of Camshaft Lengths
Camparison of Supplier 1 vs Supplier 2
35 V ariable
S upp1
S upp2
30

25
Frequency

20

15

10

0
597.0 598.5 600.0 601.5 603.0
Data
100 P arts P lotted for E ach S upplier
 SMOOTHED (SKEWED) DISTRIBUTION
Histogram of Camshaft Lengths
Camparison of Supplier 1 vs Supplier 2

Supp1 Supp2
35

30

25
Frequency

20

15

10

0
597.0 598.5 600.0 601.5 603.0 597.0 598.5 600.0 601.5 603.0
100 P arts P lotted for E ach S upplier
BOXPLOT:

GRAPH > BOXPLOT

 BOXPLOTS: GRAPH > BOXPLOT
 There is another method of looking at the data that may
be easier to see differences in the distributions
 Boxplots show the spread and center of the data

 BE CAREFUL!
 The center of the Boxplot is the MEDIAN, not the MEAN
Open worksheet
Carpet.MTW
 BOXPLOTS
Boxplot of Durability

22.5

75% to 75th
20.0
100% Percentile
17.5

15.0
Durability

Average
12.5
50th
Percentile
10.0
or Median
NOTE: 0% to
7.5
Outliers will be 25%
displayed as * 25th
5.0 Percentile

We can also generate boxplots by a variable to

look at the variation due to that variable
 BOXPLOTS W/ GROUPS
 We can also generate boxplots by a variable to look at
the variation due to that variable
 Data for 4 Experimental Carpet types is available.

Still using worksheet

Carpet.MTW
 BOXPLOTS W/ GROUPS
Boxplot of Durability

22.5

20.0

18.115
17.5
Durability

15.0
14.4825

12.8075
12.5

10.0 9.735

7.5

5.0
1 2 3 4
Carpet
INDIVIDUAL VALUE PLOT:

GRAPH > INDIVIDUAL VALUE

PLOT
 INDIVIDUAL VALUE PLOT: GRAPH > INDIVIDUAL
VALE PLOT
 Individual Value Plots also show the spread and center
of the data

Open worksheet
Billiards.MTW
INDIVIDUAL VALUE PLOT
Individual Value Plot of Elastic
90

80

70
Elastic

60
Average

50

40

30

We can also generate Individual Value Plots by a

variable to look at the variation due to that variable
 INDIVIDUAL VALUE PLOT W/ GROUPS
 We can also generate Individual Value Plots by a
variable to look at the variation due to that variable
 Data for 2 Additives is available.

Still using worksheet

Billiards.MTW
 INDIVIDUAL VALUE PLOT W/ GROUPS
Individual Value Plot of Elastic
90 Additive
0
1
2
80
75.9

70
Elastic

60

54.2

50

42.8
40

30
0 1 2
Additive
BAR CHART:

GRAPH > BAR CHART

BAR CHART: GRAPH > BAR CHART

 Bar Charts can be

created from:
1) Data that needs
to be counted
2) Functions of
data(e.g. avg,
min, max) OR
3) a Table
BAR CHART: GRAPH > BAR CHART (COUNTS OF
UNIQUE VALUES)
 Use to chart counts of unique values,
clustered by grouping variables.

Open worksheet
Exh_QC.MTW
 Count

Flaws
Period
0
1
2
3
4
5
6
7
8
9

1
Day

1
Evening
Other

3
Night

1
Weekend
UNIQUE VALUES)

2
Day
4

Evening
Peel
6

Night
3

Weekend
3

Day
2

Evening
Chart of Flaws, Period

Scratch
8

Night
0

Weekend
1

Day
0

Evening
Smudge
2

Night
3
BAR CHART: GRAPH > BAR CHART (COUNTS OF

Weekend
BAR CHART: GRAPH > BAR CHART (A FUNCTION
OF A VARIABLE)
 Use to chart counts of unique values,
clustered by grouping variables.

Still using worksheet

Exh_AOV.MTW
BAR CHART: GRAPH > BAR CHART (A
FUNCTION OF A VARIABLE)
Chart of Mean( LightOutput )
1386
1400
1313

1200
1087.33
1035 1054.67
Mean of LightOutput

1000
886.667

800

572.667 553 573.333

600

400

200

0
GlassType 1 2 3 1 2 3 1 2 3
Temperature 100 125 150
 BAR CHART: GRAPH > BAR CHART (VALUES
FROM A TABLE)
 asdfa

Open worksheet
Tires.MTW
BAR CHART: GRAPH > BAR CHART (VALUES
FROM A TABLE)
Chart of Repairs
160

140

120

100
Repairs

80

60

40

20

0
Qtr
Q1
Q2
Q3
Q4

Q1
Q2
Q3
Q4

Q1
Q2
Q3
Q4

Q1
Q2
Q3
Q4

Q1
Q2
Q3
Q4

Q1
Q2
Q3
Q4
CausesB

Leak From Seating

Damaged Sidewall
Puncture

Damaged Liner
Valve Stem Leak

Valve Core Leak

We can easily switch the X-axis so that CauseB is

plotted within Qtr.
 Repairs

Qtr
CausesB
0
20
40
60
80
100
120
140
160

Damaged Liner
Damaged Sidewall
Leak From Seating
Q1
Puncture
Valve Core Leak
Valve Stem Leak
FROM A TABLE)

Damaged Liner
Damaged Sidewall
Leak From Seating
Q2
Puncture
Valve Core Leak
Valve Stem Leak

Damaged Liner
Chart of Repairs

Damaged Sidewall
Leak From Seating
Q3
Puncture
Valve Core Leak
Valve Stem Leak

the X-axis still plotted within Qtr. Damaged Liner

Damaged Sidewall
Leak From Seating
Q4
Puncture
Valve Core Leak
We can easily stack the Causes B into one bar on

Valve Stem Leak

BAR CHART: GRAPH > BAR CHART (VALUES
BAR CHART: GRAPH > BAR CHART (VALUES
FROM A TABLE)

Chart of Repairs
C ausesB
400 V alv e S tem Leak
V alv e C ore Leak
P uncture
Leak F rom S eating
Damaged S idew all
Damaged Liner
300
Repairs

200

100

0
Qtr
Q3

Q2

Q1

Q4
PIE CHART:

GRAPH > PIE CHART

PIE CHART: GRAPH > PIE CHART
 Use to display the proportion of each data
category relative to the whole data set.

Open worksheet
Tires.MTW
PIE CHART: GRAPH > PIE CHART

Pie Chart of CausesA

Leak F rom S eating

7.0%

Damaged Liner
9.2%
P uncture
28.8%

V alv e C ore Leak

12.8%

Damaged S idew all

14.6%

V alv e S tem Leak

27.6%
3D SCATTERPLOT:

GRAPH > 3D SCATTERPLOT

3D SCATTERPLOT: GRAPH > 3D SCATTERPLOT

 Use to evaluate relationships between

three variables at once by plotting data
on three axes.
Open worksheet
Reheat.MTW
3D SCATTERPLOT: GRAPH > 3D SCATTERPLOT
3D Scatterplot of Quality vs Time vs Temp
O perator
A
B

7.5

Q uality 5.0

2.5 40

35
0.0
30 Time
350
400 25
450
Temp

Us the 3D Graph Tools to Enlarge & Rotate Graph

(Check Tools >Toolbars >3D Graph Tools).
CONCENTRATION DIAGRAMS
 CANNOT BE CREATED IN MINITAB
 Concentration Diagrams provide a visual display of
occurrences to identify trends
 Usually a pictorial representation (drawing) of the
product is used as the basis
 Occurrences are marked on the drawing where they
were noticed for all units reviewed
 Take a look at the following examples…

A Concentration Diagram is a great tool to

Investigate the nature of surface defects
LOOKING FOR PAINT DEFECTS
Top View of a Cooktop
X = 1 defect

x
xxxx x x
x
x xx
xx xx
x x x
x
x

x
xxx
xxx xx xxx
ANNOTATING GRAPHS:
• To Change Title: Double click on Title, Change Font or Text,
Click ‘OK’.
• To Add Subtitle or Footnote: Left Click anywhere on Graph,
Click Add, Select Option to be added.
• To Underline Legend Title: Double Click on Legend box, Left
click on ‘Header Font’ tab, Check Underline.
• To add data labels: Right Click anywhere on graph, Left
click on ‘Add’, Left click on ‘Data Labels’, Left click on ‘OK’.
• To add Groups to data: Double Click on any Data Point,
Select Groups tab, Select column to group by
• To Delete Legend Box: Right click on Legend box, Left Click
on ‘Delete’.
• To move the position of a Label: Right Click to select the
label you want to move. You may have to Right Click more than
once. Right Click, hold and drag the label to the new position.
• To Unslant X-axis Labels: Double click X-axis, select
Alignment tab, enter 90 for text angle, Click on ‘OK’.
• To Add Jitter to Data Points: Double click any Data Point,
select the Jitter tab, Check Add jitter to direction, Click on
‘OK’.
 CONCLUSIONS

 Results need to be Supported by data

 Not based on conjecture or intuition

 Shown in 1) Graphical & 2) Statistical format

 Make sense from an 3) Engineering standpoint

Good Conclusions Require

Data and Hard Evidence!!