You are on page 1of 6

Prove that the T2 from example 11.

1 is from the general formula of :


T2=
From example 11.1 it is known that:
n=10,

= 115.59,

= 1.06,

=1.23,

=0.83,

= 0.79

the formula used in example 11.1:


[

T2=

Proof:
T2= n [

] [

=n[

] [

] [

=
=
=

, then put the given data:


[

Then it is proven that the example 11.1 use the general formula of T2 for p=2

Example 11.1 The T2 control chart


The tensile strength and diameter of a textile fiber are two important quality characteristics that are to
be jointly controlled. The quality engineering has decided to use n=10 fiber specimens in each sample.
He has taken 20 preliminary samples, and on the basis of these data he concludes that = 115.59 psi,
= 1.06 (x10-2) inch, =1.23, =0.83, and
= 0.79. set up the T2 control chart.
The statistic he will use for process-control purpose is:
[

T2=

]
Means

variances and covariances

control chart

Tensile strength
( 1)

diameter
( 2)

115.25

1.04

1.25

0.87

0.8

2.16

0.45

115.91

1.06

1.26

0.85

0.81

2.14

0.41

115.05

1.09

1.3

0.9

0.82

6.77

0.5

116.21

1.05

1.02

0.85

0.81

8.29

0.21

115.9

1.07

1.16

0.73

0.8

1.89

0.21

115.55

1.06

1.01

0.8

0.76

0.03

0.23

114.98

1.05

1.25

0.78

0.75

7.54

0.41

115.25

1.1

1.4

0.82

0.8

3.01

0.52

116.15

1.09

1.19

0.87

0.83

5.92

0.35

10

115.92

1.05

1.17

0.86

0.95

2.41

0.1

11

115.75

0.99

1.45

0.79

0.78

1.13

0.54

12

114.9

1.06

1.24

0.82

0.81

9.96

0.36

13

116.01

1.05

1.26

0.55

0.72

3.86

0.17

14

115.83

1.07

1.17

0.76

0.75

1.11

0.33

15

115.29

1.11

1.23

0.89

0.82

2.56

0.42

16

115.63

1.04

1.24

0.91

0.83

0.08

0.44

17

115.47

1.03

1.2

0.95

0.7

0.19

0.65

18

115.58

1.05

1.18

0.83

0.79

0.36

19

115.72

1.06

1.31

0.89

0.76

0.35

0.59

20

115.4

1.04

1.29

0.85

0.68

0.62

0.63

115.5875

1.058

1.229

0.8285

0.7885

Sample
Number k

2
1k

2
2k

2
12k

| |

Tk

To calculate Hotelling T2 , we first need to use the general formula of T2 that is:
T2=
Step to find the value of T2 :
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.

Define n (in this case n=10)


Define the number of variable/ quality characteristics
Define the values of - (display in matrices form)
Define the variances and covariances and its value
Define S in matrix form
Make a syntax for transposing Make a syntax for inversing S
Make a syntax for matrix multiplication of( - ) by S-1
Make a syntax for matrix multiplication of the result number 8 by
Get the value of T2

and multiply it by n

Step to find the value of T(i)2 :


In example 11.1 there are 2 kind of i ( i.e I =1 and 2). As we now that T(i)2 is the value of the
statistic for all process variables except the ith one, so if we want to find T(i=1)2 it means the
value of the statistic for all process variables except the 1th one or in other words except x1 , and
thus for i=2.
First we find the value of T(i=1)2
1. Define n (in this case n=10)
2. Define the number of variable/ quality characteristics, that is x2
3. Define the values of - (display in matrices form)
4. Define the variances and covariances and its value
5. Define S in matrix form
6. Make a syntax for transposing 7. Make a syntax for inversing S
8. Make a syntax for matrix multiplication of( - ) by S-1
9. Make a syntax for matrix multiplication of the result number 8 by
and multiply it by
n
10. Get the value of T2

T2

T(1)2

T(2)2

d1

d2

2.150761

0.003904

0.926067

2.146858

1.224694

2.149975

4.82E-05

0.845579

2.149927

1.304396

6.759771

0.012337

2.348831

6.747434

4.41094

8.30588

0.000771

3.150457

8.305109

5.155422

1.897853

0.001735

0.793953

1.896118

1.1039

0.032525

4.82E-05

0.011433

0.032477

0.021092

7.52814

0.000771

3.000457

7.527369

4.527683

3.001724

0.021253

0.926067

2.980471

2.075657

5.933385

0.012337

2.572409

5.921047

3.360976

2.420444

0.000771

0.898831

2.419673

1.521612

1.135678

0.055711

0.214685

1.079967

0.920993

9.941585

4.82E-05

3.842734

9.941537

6.098852

3.870446

0.000771

1.45127

3.869675

2.419176

1.118662

0.001735

0.4781

1.116927

0.640562

2.551127

0.032578

0.719563

2.518548

1.831564

0.078286

0.003904

0.014685

0.074383

0.063602

0.182089

0.009446

0.112246

0.172644

0.069843

0.000771

0.000771

0.000457

2.74E-07

0.000314

0.356802

4.82E-05

0.142734

0.356754

0.214068

0.611031

0.003904

0.285823

0.607127

0.325208

We can make in one program to calculate di from this syntax:

You might also like