Sei sulla pagina 1di 33

Series A

TQM Training Module on


Scatter Diagram
Doc. No. A-04.02.20180603
Revision 02; 03rd June 2018
Authors: Pankaj Lochan

For further clarifications, write to pankaj.lochan@jsw.in © Total Quality Management, JSW Group
TQM Training Series: 6 series with 66 training modules
This training module: Scatter Diagram ( A-04)
Series-A Series-C Series-D Series-F
Basic Problem Solving Tools Quality Management Basics Productivity & Efficiency Tools Advanced Statistical Tools
A-01 Flow Charts C-01 Quality Mgmt. Basics D-01 Value Stream Mapping (VSM) F-01 Sampling & Distribution
A-02 Cause & Effects Diagram C-02 Basic Statistics D-02 Time & Motion Study F-02 Hypothesis Testing
A-03 Stratification C-03 Statistical Process Control D-03 SMED F-03 Regression
A-04 Scatter Diagram C-04 KPI Drill Down D-04 Wrench Time Analysis F-04 Basics of DoE
A-05 Control Charts C-05 KPI Benchmarking D-05 Queuing Theory F-05 Factorial DoE
A-06 Check Sheets C-06 Strategic Analysis Tools D-06 Inventory Management F-06 Principal Component Analysis
A-07 Histogram C-07 Policy Management D-07 Linear Programming Problem F-07 Cluster Analysis
C-08 Policy Diagnosis F-08 Conjoint Analysis
A-08 Pareto Charts D-08 Game Theory
C-09 Daily Management F-09 Discriminant Analysis
A-09 Graphs D-09 OEE
C-10 Daily Mgmt. in Maintenance F-10 Factor Analysis
D-10 PERT & CPM
C-11 Cross Functional Mgmt.
Series-B F-11 Response Surface Method
C-12 Quality Assurance Basics
Basic Management Tools F-12 Taguchi DoE
C-13 MSA
B-01 Brainstorming Series-E F-13 Weibull Analysis
C-14 PFD, FMEA, Control Plan
B-02 Affinity Diagram Decision-making Tools
C-15 Cost of Poor Quality (COPQ)
B-03 Arrow Diagram C-16 Improvement Fundamentals E-01 Quality Function Deployment
B-04 Tree Diagram C-17 4i Methodology E-02 Fault Tree Analysis
B-05 PDPC C-18 5S E-03 AHP & Paired Analysis
B-06 Matrix Diagram C-19 Quality Circles E-04 Pugh Matrix
B-07 Matrix Data Analysis C-20 QC Story Approach E-05 Time Series Analysis
B-08 Relation Diagram C-21 Kaizen, OPL, Poka Yoke
3

FLOW OF DOCUMENT

Introduction to
1
 What is Scatter ?
Scatter Diagram  Types of Scatter
Diagrams

How to make a
2
 What is Correlation
Coefficient – r ?
Scatter Diagram

Examples from our


3 Plant

How to make a
4 Scatter in MS-Excel
(c) TQM Department, JSW Vijayanagar
4

SCATTER DIAGRAMS: USES & BENEFITS

Uses / Applications Benefits

‾ Used for studying ‾ Helps understand type of correlation


CORRELATION (relation between (positive, negative)
two inter-related data variables) ‾ Helps understand degree of correlation
OR, in other words, (none, weak, strong, perfect)

‾ Used for study the relationship ‾ Helps establish countermeasures on basis of


between dependent (x) and the relationship
independent (y) ‾ Also, help establish cause & effect relationship

Caution:
‾ Don’t confuse between correlation and a cause-effect relationship, that will
need further testing even after being well correlated !

(c) TQM Department, JSW Vijayanagar


5

SCATTER DIAGRAM: AN INTRODUCTION

 A Graphical Technique - can be used in teams to


establish Correlation between two interdependent
variables.

Weight of a
 Used as one of the basic problem solving tools mainly for

person 
defining relationship between two variables; esp when
there is a suspicion on two parameters to be correlated
 First used by Prof. Kaoru Ishikawa in 1943.
 For more than 2 variables, multivariate analysis /
hypothesis testing is used Height of a
person 

Differentiation of Scatter Plots can be made on basis of:

1. Type of Correlation – Positive, negative, curvi-linear

2. Extent of Correlation – None, Low, High, Perfect

(c) TQM Department, JSW Vijayanagar


6

TYPES OF SCATTER DIAGRAMS

Type of Correlation No correlation Positive Negative


correlation correlation

consumption
Bus timings

Ductility
5 major types:
1. No correlation

Lime
2. Positive
Train timings in HM silicon Carbon content %
correlation Bellary

3. Negative Curvi-linear correlation Partial correlation


correlation

consumption
4. Curvi-linear
Stamina

Income
correlation

Fuel
5. Partial
correlation
Age Speed of vehicle Education

(c) TQM Department, JSW Vijayanagar


7

TYPES OF SCATTER DIAGRAMS

Extent of Correlation No correlation Weak Strong


Correlation correlation

4 major types:
1. No correlation
2. Weak
Correlation
3. Strong
correlation
4. Perfect
correlation

Perfect correlation
(c) TQM Department, JSW Vijayanagar
8

WHAT IS CORRELATION ?

 Correlation is the measure of amount of linear relationship between two variables

 Is represented by the correlation coefficient ‘r’ which is a pure number ranging


between -1 and +1;

 -1 means totally negative correlation and +1 means totally positive correlation

 NEGATIVE CORRELATION POSITIVE CORRELATION 

-1 -0.8 -0.5 0 +0.5 +0.8 +1

Strong moderate weak weak moderate Strong

(c) TQM Department, JSW Vijayanagar


9

WHAT IS CORRELATION ?

 Correlation coefficient ‘r’


between two variables ‘x’ &
‘y’ is calculated as per the
Pearson’s formula 
N(Σxy) – (Σx)(Σy)
 Where: r=
N = number of pairs of values,
Σx = sum of x values Ѵ[ NΣx
2 – (Σx)2 ] [ NΣy2 – (Σy)2 ]
Σy = sum of y values
Σxy = product of paired values
Σx2 = sum of squares of x
values
Σy2= sum of squares of y
values

(c) TQM Department, JSW Vijayanagar


10

FLOW OF DOCUMENT

Introduction to
1 Scatter Diagram

How to make a
2 Scatter Diagram

Examples from our


3 Plant

How to make a
4 Scatter in MS-Excel
(c) TQM Department, JSW Vijayanagar
1

PROCEDURE TO PLOT SCATTER DIAGRAM

1 Discuss the purpose (suspicion statement)

2 Decide the Paired Variables (x, y)

Collect pairs of data for x & y (desirable > 30


3 pairs)

4 Plot the collected data on an x-y graph

5 Interpret from the plot (Graph)

Step 6&7 are optional, can 6 Calculate the correlation coefficient ‘r’
be used to make more
precise conclusions
7 Infer from the value of ‘r’
(c) TQM Department, JSW Vijayanagar
12

FLOW OF DOCUMENT

Introduction to
1 Scatter Diagram

How to make a
2 Scatter Diagram
 Hot Metal Silicon and Specific
Lime Consumption
 Blast Furnace Slag Basicity
and Cao correlation
Examples from our
3 Plant  Tap Ladle Oxygen and Mn
correlation

How to make a
4 Scatter in MS-Excel
(c) TQM Department, JSW Vijayanagar
Lime
HM Silicon Sp.
Consm 1.02 73.08 0.84 54.63 0.66 45.80

DECIDE PURPOSE (SUSPICION 0.73 40.70 0.82 61.16 0.68 63.79 0.51 42.35
Step 1&2 0.70
0.74
46.97
48.93
1.25
1.40
80.97
80.31
0.81
0.94
48.63
61.24
0.59
1.15
77.02
42.26

STATEMENT) & 2 VARIABLES 0.92 58.95 1.18 77.48 0.91 60.79 0.86 61.72
1.05 64.62 1.18 69.86 1.03 58.21 1.25 77.61
1.05 70.60 0.75 48.16 0.82 58.95 0.80 61.14
0.97 67.96 0.99 65.89 0.90 62.69 0.91 62.70
0.67 42.58 0.99 55.38 1.08 66.09 0.92 49.53
PLANT 0.99 67.98 0.72
0.89
58.98
56.65
1.12
1.38
65.58
69.01
0.64
0.90
45.89
50.59
79.86
EXAMPLE – 1
1.09
1.16 60.74
0.75 46.21 0.76 58.57 0.45 41.04
0.69 46.12
0.79 60.90 0.59 46.53 0.95 71.19
0.75 56.04
1.32 69.84 1.30 68.74 0.75 46.12

HM Silicon To check the 0.81


1.05
44.67
73.43
0.84
0.70
57.60
58.15
1.02
0.99
63.40
48.39
0.93
0.68
70.00
52.00
1.09 55.14 0.96 68.14 0.96 64.82
59.19
and Lime correlation of HM
0.85
0.82 60.52 0.98 64.62 0.57 42.48
0.79 53.60
0.69 54.21 1.07 53.99 0.79 52.08
0.77 49.16
54.38 55.38 69.60
Consumption
0.94 0.79 1.03
Silicon and Specific 41.05
0.43
0.64 48.09 1.09 51.73 0.56 41.60
0.72 46.60
0.80 50.72 0.67 52.38 1.17 60.80
1.14 85.30
64.75 79.43 75.38
Lime consumption.
1.04 1.49 1.20
0.95 69.86
0.77 54.68 0.76 49.83 0.68 48.48
1.06 69.88
0.68 49.19 0.70 45.28 1.30 69.90
1.29 78.67 0.51 40.99 0.85 68.46 1.06 79.32
0.53 42.70 0.91 53.36 1.10 57.66 0.97 62.16
0.42 52.10 0.61 44.20 0.73 46.36 1.12 78.52
0.93 56.19 1.20 66.15 0.80 57.61 0.52 43.65
1.36 73.38 0.94 70.83 0.88 60.54 1.30 69.90
0.59 53.41 0.75 41.99 0.99 59.53 0.77 42.99
0.79 53.56 0.64 49.73 1.24 80.91 0.72 43.80

COLLECT DATA
0.98 63.67
Step 3 0.99 55.95
1.00
0.83
68.95
52.40
1.39
0.87
80.40
54.31
0.90
0.80
68.22
43.83
0.56 42.45 0.58 53.85 1.42 81.84 0.79 60.47
0.94 60.16 0.79 63.06 0.98 67.46 0.88 58.36
0.84 51.52
Data Collection
0.71 58.16 1.01 61.28 1.17 64.25
1.09 68.14 0.73 46.11 1.09 66.20 0.45 41.21
1.07 64.48 49.69 58.75 50.73
period: Nov-Dec’14
0.80 0.78 0.85
0.66 43.83 1.01 70.24 1.26 78.20 0.67 40.74
1.01 74.41 1.32 68.71 1.42 47.41 0.82 61.42
1.13 74.08 0.86 60.16 0.52 78.33 0.60 40.48
0.39 41.82 0.93 67.68 0.68 42.79 0.44 40.07
0.48 44.67 1.05 78.41 0.78 54.22 0.97 72.02
1.17 77.56 1.20 63.30 1.34 82.95 0.73 49.44
0.62 47.94 1.08 69.08 1.42 74.84 0.70 49.57
0.85 46.58 0.89 52.11 0.61 51.52 0.90 60.40
0.49 40.11 1.02 75.71 0.71 43.26 0.75 46.02
1.13 75.61 1.02 67.33 0.65 45.49 0.81 49.09
0.78 56.84 0.88 49.53 0.67 58.02 0.50 41.20
87.19
(c) TQM Department, JSW Vijayanagar 1.41
Step 4
PLOT THE SCATTER

PLANT
EXAMPLE – 1 Draw Scatter plot by taking Hot Metal Silicon (independent variable) as X- axis and
Specific lime Consumption (Dependent Variable) as Y axis
HM Silicon
and Lime 100.00
Step 5 Infer from the plot
Consumption 90.00
Speccific Lime Consumption

80.00
By observing Scatter plot,
70.00 we confirm the hypothesis
60.00 that the two parameters
50.00 HM Silicon and Specific
40.00 Lime Consumption are
30.00 positively correlated and
20.00
they have moderate
r= positive correlation.
10.00
0.768
0.00
0.00 0.50 1.00 1.50 2.00
Hot Metal Silicon

(c) TQM Department, JSW Vijayanagar


B2 Basicity
N CaO (x)
(y)

DECIDE PURPOSE (SUSPICION


1 37.97 1.11
2 36.39 0.95
3 36.41 0.98
Step 1&2 4 36.6 0.96

STATEMENT) & 2 VARIABLES 37.42 1.03


5
6 36.24 0.95
7 37.01 1
8 36.97 1
9 36.7 0.99
10 37.81 1.09
11 37.06 1.03
PLANT
Collect Data,
12 37.06 1.02
36.39 0.98
EXAMPLE – 2
13

Step 3 14 36.09 0.95

Slag Analysis 36.59 1

Minimum 30
15
16 36.19 0.98
36.54 0.99
of Blast
17
18 36.91 1.01
During Blast Furnace #4
data points
19 36.66 1.01

Furnace 20
21
37.24
36.43
1.04
0.97
slag analysis, we tried 22
23
36.69
36.53
1
0.99

testing the correlation Data Collection period: 1st 24


25
36.44
36.62
0.99
0.99

January 2015 to 31st January 26 36.85 1

between Calcium oxide


27 36.39 0.96

2015 36.85 1
28
29 36.03 0.95

(CaO) and Basicity (B2) of 30


31
36.37
36.58
0.97
0.98
32 36.91 1
Slag by using a scatter 33
34
37.25
36.53
1.03
0.98
35 37.75 1.06
plot analysis of data 36
37
36.47
36.77
0.97
1.02
38 37.44 1.06
representing the process 39
40
37.44
36.91
1.07
1.02

collected over a period of


41 37.5 1.05
42 36.86 1.03
43 36.4 1.01

time to confirm suspicion 44


45
36.59
36.58
1.02
1.01
46 36.76 0.99

or hypothesis. 47
48
37.06
36.6
1.01
0.98
49 37.04 1.02
50 36.79 1.03
51 37.22 1.07
(c) TQM Department, JSW Vijayanagar 52 36.99 1.06
Step 4 PLOT THE SCATTER

PLANT
EXAMPLE – 2Draw Scatter plot by taking Calcium oxide (independent variable) as X- axis and
Slag Analysis Basicity (B2) (Dependent Variable) as Y axis
of Blast 1.12
Furnace 1.10
Step 5 Infer from the plot
1.08 By observing Scatter plot
1.06 constructed in step
number 2 we confirm the
Basicity

1.04
hypothesis that the two
1.02 parameters Calcium Oxide
1.00 and Basicity of slag are
positively correlated and
0.98
they have high positive
0.96 correlation.
0.94
35.00 36.00 37.00 38.00 39.00
Calcium Oxide
(c) TQM Department, JSW Vijayanagar
B2 Basicity
N CaO (x) xy x2 y2
(y)
1 37.97 1.11 42.1467 1441.721 1.2321
36.39 0.95 34.5705 1324.232 0.9025

CALCULATE CORRELATION
2
3 36.41 0.98 35.6818 1325.688 0.9604
Step 6 4 36.6
37.42
0.96
1.03
35.136
38.5426
1339.56
1400.256
0.9216
1.0609

COEFFICIENT ‘R’
5
6 36.24 0.95 34.428 1313.338 0.9025
7 37.01 1 37.01 1369.74 1
8 36.97 1 36.97 1366.781 1
9 36.7 0.99 36.333 1346.89 0.9801
10 37.81 1.09 41.2129 1429.596 1.1881
11 37.06 1.03 38.1718 1373.444 1.0609
37.06 1.02 37.8012 1373.444 1.0404
N(Σxy) – (Σx)(Σy)
12

r=
13 36.39 0.98 35.6622 1324.232 0.9604
14 36.09 0.95 34.2855 1302.488 0.9025
15 36.59 1 36.59 1338.828 1
36.19 0.98 35.4662 1309.716 0.9604

Ѵ[ NΣx
16
2 – (Σx)2 ] [ NΣy2 – (Σy)2
17 36.54 0.99 36.1746 1335.172 0.9801
18 36.91 1.01 37.2791 1362.348 1.0201
19 36.66 1.01 37.0266 1343.956 1.0201
20 37.24 1.04 38.7296 1386.818 1.0816
39.8277
= 21
22
36.43
36.69
0.97
1
35.3371
36.69
1327.145
1346.156
0.9409
1
Ѵ1953.183604 23
24
36.53
36.44
0.99
0.99
36.1647
36.0756
1334.441
1327.874
0.9801
0.9801
25 36.62 0.99 36.2538 1341.024 0.9801
26 36.85 1 36.85 1357.923 1

=
27 36.39 0.96 34.9344 1324.232 0.9216
0.901184 28
29
36.85
36.03
1
0.95
36.85
34.2285
1357.923
1298.161
1
0.9025

Infer from the value 30


31
36.37
36.58
0.97
0.98
35.2789
35.8484
1322.777
1338.096
0.9409
0.9604
Step 7 32 36.91 1 36.91 1362.348 1

of ‘r’ 33
34
37.25
36.53
1.03
0.98
38.3675
35.7994
1387.563
1334.441
1.0609
0.9604

A ‘r’ value of 0.901 37.75 1.06 40.015 1425.063 1.1236


35
36 36.47 0.97 35.3759 1330.061 0.9409
36.77 1.02 37.5054 1352.033 1.0404
indicates that the two
37
38 37.44 1.06 39.6864 1401.754 1.1236
39 37.44 1.07 40.0608 1401.754 1.1449
parameters Calcium Oxide 40
41
36.91
37.5
1.02
1.05
37.6482
39.375
1362.348
1406.25
1.0404
1.1025

and Basicity of slag are 42


43
36.86
36.4
1.03
1.01
37.9658
36.764
1358.66
1324.96
1.0609
1.0201

positively correlated and 36.59 1.02 37.3218 1338.828 1.0404


44
48 36.6 0.98 35.868 1339.56 0.9604
37.04 1.02 37.7808 1371.962 1.0404
they have very high
49
50 36.79 1.03 37.8937 1353.504 1.0609
51 37.22 1.07 39.8254 1385.328 1.1449
positive correlation. 52
53
36.99
37.29
1.06
1.05
39.2094
39.1545
1368.26
1390.544
1.1236
1.1025
Σ 53 1951.18 53.41 1967.026 71842.05 53.8933
(c) TQM Department, JSW Vijayanagar
PPM OXYGEN IN TAP LADLE VS % MN IN TAP
LADLE
PLANT When deoxidizing with Ferro manganese, the de-oxidation product is Mn (Fe)O, the activity of which is
EXAMPLE – 3 lowered in the presence of a calcium Aluminate slag. The change in dissolved oxygen and manganese
contents during tapping shows the trend that as % Mn increases causes decrease in Oxygen
Steel Melt % Mn in Tap PPM Oxygen in

Shop Ladle Ladle % Mn in tap Ladle


0 0 600
0.33 410
0.34 410
500
The scatter
0.35 422
0.36 450 diagram

PPM O in Tap Ladle


0.2 500 400 shows there
0.24 485
0.26 465
is negative
0.29 450
300
correlation
0.4
0.41
320
310 200
between
0.42 300 Oxygen in tap
0.43 280 100 ladle with
0.46 250
0.49 220 Manganese in
0.5 210
0
0 5 10 15 20 25 tap ladle
0.55 160
0.75 150 %Mn in Tap Ladle
0.76 80 18
0.77 70
0.79
(c) TQM Department, JSW Vijayanagar 90
19

FLOW OF DOCUMENT

Introduction to
1 Scatter Diagram

How to make a
2 Scatter Diagram

Examples from our


3 Plant

How to make a
4 Scatter in MS-Excel
(c) TQM Department, JSW Vijayanagar
20

CONSTRUCTION OF SCATTER DIAGRAM IN


EXCEL

Taking the
same Blast
Step 1: Enter Furnace
data for x & y Basicity
example

(c) TQM Department, JSW Vijayanagar


21

CONSTRUCTION OF SCATTER DIAGRAM IN


EXCEL

Step 2: Click on
‘Insert’ and then
click on “Scatter”

(c) TQM Department, JSW Vijayanagar


22

CONSTRUCTION OF SCATTER DIAGRAM IN


EXCEL

A Scatter
Plot
appears

(c) TQM Department, JSW Vijayanagar


23

CONSTRUCTION OF SCATTER DIAGRAM IN


EXCEL

Step 3: Right click on


scatter data points and
select “Add trendline”

(c) TQM Department, JSW Vijayanagar


24

CONSTRUCTION OF SCATTER DIAGRAM IN


EXCEL

A Trendline
appears

(c) TQM Department, JSW Vijayanagar


25

CONSTRUCTION OF SCATTER DIAGRAM IN


EXCEL

Step 4a:
Select Graph, then go to
“Layout” on the horizontal
bar, then to “Axis Titles”
and then to Primary
Horizontal Axis Title”

(c) TQM Department, JSW Vijayanagar


26

CONSTRUCTION OF SCATTER DIAGRAM IN


EXCEL

Step 4b:
Enter the name of the
horizontal Axis

(c) TQM Department, JSW Vijayanagar


27

CONSTRUCTION OF SCATTER DIAGRAM IN


EXCEL

Step 5a:
Select the Graph, then go to
“Layout” on the horizontal bar,
then to “Axis Titles” and then
to Primary Vertical Axis Title”.
Then choose the type

(c) TQM Department, JSW Vijayanagar


28

CONSTRUCTION OF SCATTER DIAGRAM IN


EXCEL

Step 5b:
Enter the name of
the Vertical Axis

(c) TQM Department, JSW Vijayanagar


29

CALCULATION OF CORRELATION COEFFICIENT


‘R’ IN EXCEL

Type CORRREL
Then select Array1 for data
range of x and Array 2 for
data range of y

(c) TQM Department, JSW Vijayanagar


30

(c) TQM Department, JSW Vijayanagar


31

The value of ‘r’ returned by


the formula is 0.901 which
shows a fairly high level of
positive correlation

(c) TQM Department, JSW Vijayanagar


Please login to:
tqm.jsw.in
to read the training modules
THANK YOU

Potrebbero piacerti anche