TO CHOOSE A LEADER OF THE WORLD.pdf


Aperçu du fichier PDF to-choose-a-leader-of-the-world.pdf - page 10/12

Page 1 2 3 4 5 6 7 8 9 10 11 12



Aperçu texte


10

TO CHOOSE A LEADER OF THE WORLD

The number q21q22 q23 vector components are the priority criteria for K1, K2 and K3 respectively
q 2 = (0. 69; 0.243; 0.088).
This procedure should be done for all matrices of paired comparisons.
Index of consistency of IP in each matrix and for the entire hierarchy can also be expressed in the following way:
Determines the amount of each of the j – t h column of the matrix judgement
sj = а1j + а2j+ а3j + ……… + аn j, j=1,2,3, …. ,n
Then the result is multiplied by the j-t h component of a normalized vector priorities q2, i.e., the sum of the first column in the first
judgement component sum judgements of the second column in the second, etc.
рj= sj·q2j,
j=1,2,3, ……, n
The sum of the numbers r j reflects the proportionality of preferences, the closer this value to n (number of objects and actions in a
matrix of paired comparisons) and the more consistent judgments
λmax = р1+р2+р3+ ……+рn
Deviation from consistency is consistency index

ИС 

max  n
n 1

.

Attitude consistency (OC). In order to determine how exactly (ИС) consistency index reflects the consistency of judgements of
his compare with random index (СИ), which corresponds to the coherence matrix with random thoughts that are selected from the
timeline
1/9, 1/8, 1/7, 1/6, 1/5, 1/4, 1/3, 1/2, 1, 2, 3, 4, 5, 6, 7, 8, 9,
Assuming equal probability of selecting any of these numbers.
Table shows the mean values of random index consistency (СИ) for random matrices judgments of different order.
Attitude consistency index of (ИС) to the mean value of C is called a consistency index of random attitude consistency OС
ОС 

ИС
СИ

OS value less than or equal to 0.10 is considered acceptable.
Returning to the example we have:
S 1 = 1 + 1/3 + 1/7 = 31/21
S 2 = 3 +1 + 1/3 = 3/13
S 3 = 7 + 3 + 1 = 11
p1 = s1·q21 = 31/21· 0,669 = 0,988
p2 = s2·q22 = 13/3 · 0,243 = 1,051
p3 = s3·q23 = 11 · 0,088 = 0,967

The size of the
matrix
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

average value
random index
consistency
( СИ )
0.00
0.00
0.58
0.90
1.12
1.24
1.32
1.41
1.45
1.49
1.51
1.48
1.56
1.57
1.59

λmax = р1+р2+р3 = 0,988 + 1,051 + 0,967 = 3,007
ИС = (λmax - n)/( n - 1) = (3,09 - 3)/(3 -1) = 0,004
ОС = ИС/СИ = 0,045/0,58 = 0,006.
Vectors are priorities and attitudes the coherence matrices are defined for all judgments
from the second level.
To determine the priorities of the alternatives to local priorities by the priority of the
associated criteria and find the amounts for each item in accordance with criteria that are
affected by this element. Relabel
q 3 k -vector k matrix priorities based on priorities
q3ki -i- th element of the vector of the matrix k. priorities of
judgments, located on the third level;
q2 k-k-th element of the vector of the matrix priorities of judgments, located on the second level;
q j -priority of the j- th element of the third level.
Then the priority of the j-th element of the third level is defined as
q1 = q311* q21 321 + q * q22 331+ q * q23
q2 = q31221 * q + q322 22 * q + q332* q23
q 3 = q 313* q21 + q323* q22 q +333· q23
For example, assume that the matrix of paired comparisons are listed in table
Priorities of the alternatives to get to read as follows:
q1 = q311· q21 + q321· q22 + q 331· q23 = 0.30. 0.67 + 0.58 ·0.24 + 0.67. 0.09 = 0.40
q2 = q312· q21 + q322· q22 + q 332· q23 = 0.63 · 0.67 + 0.35. 0.24+ 023 · 0.09 = 0.53
q3 = q 313· q21 + q322· q22 + q 333· q23 = 0.06 · 0.67 + 0.07 · 0.24 + 0.10 .0.09 = 0.07