0% found this document useful (0 votes)
8 views4 pages

Entropy Example Tutorial

The document outlines a step-by-step process for calculating weights using entropy from a decision-making matrix. It includes normalization of the matrix, calculation of entropy values, degree of diversity, and final weight determination. The results confirm that the sum of the weights equals one, validating the calculations.

Uploaded by

Eng Fareida
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
8 views4 pages

Entropy Example Tutorial

The document outlines a step-by-step process for calculating weights using entropy from a decision-making matrix. It includes normalization of the matrix, calculation of entropy values, degree of diversity, and final weight determination. The results confirm that the sum of the weights equals one, validating the calculations.

Uploaded by

Eng Fareida
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

Entropy

The target result is to obtain to weight. As the following steps

Start from decision making matrix below

c1 c2 c3 c4 c5
A1 120 40 82.5 182 25
A2 130 30 85.6 239 25
A3 95 35 80.9 147 21
A4 110 35 81.7 153 22
A5 120 45 82.3 177 28
A6 100 25 80.4 193 23
A7 98 30 81.2 162 20

Step One:

Calculate the Normalization matrix (P) according the Eq. below

Pij=Xij/∑(Xij)

In order that find summation of each column (c1, c2, c3, c4 and c5) for the matrix above

sum 773 240 574.6 1253 164

For example to summation:


c1
A1 120
A2 130
A3 95
A4 110
A5 120
A6 100
A7 98
sum 773

Now, calculate the normalization of the first value in first criteria c1 and in first
alternative A1=120 as the following:

P11=120/773=0.155239
C1 and A2 =130

P12=130/773=0.168176 and so on

Normalization
0.15523 0.16666 0.14357 0.14525 0.15243
9 7 8 1 9
0.16817 0.14897 0.19074 0.15243
6 0.125 3 2 9
0.12289 0.14583 0.14079 0.11731 0.12804
8 3 4 8 9
0.14230 0.14583 0.14218 0.12210 0.13414
3 3 6 7 6
0.15523 0.14126 0.17073
9 0.1875 0.14323 1 2
0.12936 0.10416 0.13992 0.14024
6 7 3 0.15403 4
0.12677 0.14131 0.12195
9 0.125 6 0.12929 1

Step 2:

Calculate the Entropy value ( ej ) by the Eq. blow, in the following steps:

First find

Pij * ln(Pij)

= 0.155239* ln(0.155239)= -0.28918

-0.28918 -0.29863 -0.27867 -0.28023 -0.28674


-0.29981 -0.25993 -0.28364 -0.31603 -0.28674
-0.25764 -0.28077 -0.27602 -0.2514 -0.26318
-0.27746 -0.28077 -0.27735 -0.25677 -0.26948
-0.28918 -0.31387 -0.27834 -0.27647 -0.3018
-0.26457 -0.2356 -0.27518 -0.28813 -0.27549
-0.26184 -0.25993 -0.27652 -0.26449 -0.2566
Then the summation of each column in the matrix above

sum -1.93968 -1.9295 -1.94572 -1.93352 -1.94002


Then calculate –ln(m) m is the number of the alternative (A1-A7) so there are 7
alternatives.

-ln(7)= -1.94591  1/-1.94591= -0.5139

each value of the vector of summation above multiplied by -0.5139

0.99679 0.99156 0.99990 0.99363 0.99697


9 7 4 1 4

Step 3:

Find the degree of diversity (dj) by subtraction the entropy value of the vector above from
1.

dj=1-ej

0.00843 0.00636
0.003201 3 9.59E-05 9 0.003026

Step 4:
The final step is find first the summation of the dj vector above as the following:

sum dj 0.021124315

Then finally division the dj by the summation value as the following , in order to find the
weights as below:
The weights (w1, w2, w3, w4 and w5):

wj=dj/ sum(dj)

0.15154 0.3991 0.0045 0.3014 0.14323


4 9 4 9 5
For prove the results are correct, the condition is the summation weights should be equal
one.

sum(wj)=1

0.151544+0.39919+0.00454+0.30149+0.143235=1 PROVE IT

You might also like