You are on page 1of 14

Dr-Muhammad Tayyab

Criteria = 4
Education level
Soft Skills
Job Experience
Commercial Awareness AHP
1 Pairwise Comparison of the candidates

a. Degree level comparison

Pairwise comparison matrix Normalized Matrix (cell value / column


Sajid Sabieh Nawaz Sajid
Sajid Normalization Sajid #DIV/0!
Sabieh Sabieh #DIV/0!
Nawaz Nawaz #DIV/0!
Summat: 0 0 0 Summat: #DIV/0!

b. Soft skills comparison

Pairwise comparison matrix Normalized Matrix (cell value / column


Sajid Sabieh Nawaz Sajid
Sajid Normalization Sajid #DIV/0!
Sabieh Sabieh #DIV/0!
Nawaz Nawaz #DIV/0!
Summat: 0 0 0 Summat: #DIV/0!

c. Professional experience comparison

Pairwise comparison matrix Normalized Matrix (cell value / column


Sajid Sabieh Nawaz Sajid
Sajid Normalization Sajid #DIV/0!
Sabieh Sabieh #DIV/0!
Nawaz Nawaz #DIV/0!
Summat: 0 0 0 Summat: #DIV/0!

d. Commercial awareness comparison

Pairwise comparison matrix Normalized Matrix (cell value / column


Sajid Sabieh Nawaz Sajid
Sajid Normalization Sajid #DIV/0!
Sabieh Sabieh #DIV/0!
Nawaz Nawaz #DIV/0!
Summat: 0 0 0 Summat: #DIV/0!

Combined Preference Vector of the candidates


Degree Soft Skills Expernc
Sajid #DIV/0! #DIV/0! #DIV/0!
Sabieh #DIV/0! #DIV/0! #DIV/0!
Nawaz #DIV/0! #DIV/0! #DIV/0!
Summat: #DIV/0! #DIV/0! #DIV/0!

2 Pairwise Comparison of the criteria

Pairwise comparison matrix Normalized Matrix (cell value / co


Degree Soft Skils Exprnc Com awar Degree
Degree Degree #DIV/0!
Soft Skils Normalization Soft Skils #DIV/0!
Exprnc Exprnc #DIV/0!
Com awar Com awar #DIV/0!
Summat: 0 0 0 0 Summat: #DIV/0!

3 Overall Score of the candidates (Combined preference vectors of the ca

Combined Preference Vector of the candidates Preference Vector of


the Criteria
Degree Soft Skills Expernc Awareness
Sajid #DIV/0! #DIV/0! #DIV/0! #DIV/0! Degree
Sabieh #DIV/0! #DIV/0! #DIV/0! #DIV/0! Soft Skils
Nawaz #DIV/0! #DIV/0! #DIV/0! #DIV/0! Exprnc
Summat: #DIV/0! #DIV/0! #DIV/0! #DIV/0! Com awar
Summat:

4 Consistency Test

a. Find out the consistency index (CI)

aa. Pairwise comparison matrix of criteria X preference vector of criteria)


Preference Vector of
Pairwise comparison matrix the Criteria
Degree Soft Skils Exprnc Com awar
Degree 0 0 0 0 Degree
Soft Skils 0 0 0 0 Soft Skils
Exprnc 0 0 0 0 Exprnc
Com awar 0 0 0 0 Com awar
Summat: 0 0 0 0 Summat:
ab. Divide this new matrix with preference vector of the criteria

Preference Vector of
A new vector is formed the Criteria

Degree #DIV/0! Degree #DIV/0!


Soft Skils #DIV/0! Soft Skils #DIV/0!
Exprnc #DIV/0! Exprnc #DIV/0!
Com awar #DIV/0! Com awar #DIV/0!
Summat: #DIV/0! Summat: #DIV/0! Summat:

ac. Determine the CI using a formula [CI = (Average - n) / (n-1)]

CI = ###
T
If CI = 0, then
If CI ≠ 0, then yo
degree of cons

b. Find out the degree of consistency

ba. Observe the Random Index (RI) from the random experiment table against your "n"

bb. Calculate the degree of consistency using the formula (degree of consistency = con

n= 4
CI/RI = ###

In general, the degree of consistency is satisfactory if CI/RI =< 0.10. If CI/RI > 0, then there are probably serious in
meaningful.

Remember that in this instance, we have evaluated the degree of consistency only for the pairwise comparisons in
not mean that we have verified the consistency for the entire AHP. We would still have to evaluate the pairwise co
matrices before we could be sure the entire AHP for this problem was con
SDSB-LUMS

AHP Sajid
Candidates = 3
Sabieh Nawaz

d Matrix (cell value / column sum) Preference Vector


Sabieh Nawaz Degree
#DIV/0! #DIV/0! Row Average Sajid #DIV/0!
#DIV/0! #DIV/0! Sabieh #DIV/0!
#DIV/0! #DIV/0! Nawaz #DIV/0!
#DIV/0! #DIV/0! Summat: #DIV/0!

d Matrix (cell value / column sum) Preference Vector


Sabieh Nawaz Soft Skills
#DIV/0! #DIV/0! Row Average Sajid #DIV/0!
#DIV/0! #DIV/0! Sabieh #DIV/0!
#DIV/0! #DIV/0! Nawaz #DIV/0!
#DIV/0! #DIV/0! Summat: #DIV/0!

d Matrix (cell value / column sum) Preference Vector


Sabieh Nawaz Experience
#DIV/0! #DIV/0! Row Average Sajid #DIV/0!
#DIV/0! #DIV/0! Sabieh #DIV/0!
#DIV/0! #DIV/0! Nawaz #DIV/0!
#DIV/0! #DIV/0! Summat: #DIV/0!

d Matrix (cell value / column sum) Preference Vector


Sabieh Nawaz Awareness
#DIV/0! #DIV/0! Row Average Sajid #DIV/0!
#DIV/0! #DIV/0! Sabieh #DIV/0!
#DIV/0! #DIV/0! Nawaz #DIV/0!
#DIV/0! #DIV/0! Summat: #DIV/0!

of the candidates
Awareness
This matrix (combination of the preference
#DIV/0! vestors) provides us the detail about preference
#DIV/0! of each candidate according to the individual
#DIV/0! criterion.
This matrix (combination of the preference
vestors) provides us the detail about preference
of each candidate according to the individual
criterion.
#DIV/0!

Preference Vector of
malized Matrix (cell value / column sum) the Criteria
Soft Skils Exprnc Com awar
#DIV/0! #DIV/0! #DIV/0! Degree #DIV/0!
#DIV/0! #DIV/0! #DIV/0! Row Average Soft Skils #DIV/0!
#DIV/0! #DIV/0! #DIV/0! Exprnc #DIV/0!
#DIV/0! #DIV/0! #DIV/0! Com awar #DIV/0!
#DIV/0! #DIV/0! #DIV/0! Summat: #DIV/0!

This vector provide us the ranking of criteria


according to their importance

preference vectors of the candidates X Preference vector of the criteria)

Preference Vector of Overall score of the


the Criteria candidates

#DIV/0! Sajid #DIV/0!


#DIV/0! Sabieh #DIV/0!
#DIV/0! Nawaz #DIV/0!
#DIV/0! Summat: #DIV/0!
#DIV/0!

This vector is the final ranking of the candidates


as per importance of the criteria and their own
traits

Preference Vector of A new vector is


the Criteria formed

#DIV/0! Degree #DIV/0!


#DIV/0! Soft Skils #DIV/0!
#DIV/0! Exprnc #DIV/0!
#DIV/0! Com awar #DIV/0!
#DIV/0! Summat: #DIV/0!
#DIV/0! Average = SUM / n n=no. of criteria
#DIV/0! n= 4
#DIV/0!
Avg #DIV/0!
#DIV/0!
#DIV/0!

This should be very close to "n"

This should be very close to "zero".


If CI = 0, then you are a perfectly consistent decision maker.
If CI ≠ 0, then you need to know how much you are deviating, i.e.
degree of consistency that is acceptable must be determined.

e against your "n"

(degree of consistency = consistency ratio (CR) = CI / RI)

there are probably serious inconsistencies, and the AHP results may not be
ul.

the pairwise comparisons in the decision criteria preference matrix. This does
e to evaluate the pairwise comparisons for each of the four individual criterion
AHP for this problem was consistent.
SOLU

1 Pairwise Comparison of the sites

a. Market base

Pairwise comparison matrix Normalized Matrix (cell value / column s


Atlanta Birmingha Charlotte Atlanta
Atlanta 1 3 2 Normalization Atlanta 0.545455
Birmingham 0.333333 1 0.2 Birmingham 0.181818
Charlotte 0.5 5 1 Charlotte 0.272727
Summat: 1.833333 9 3.2 Summat: 1

b. Income level

Pairwise comparison matrix Normalized Matrix (cell value / column s


Atlanta Birmingha Charlotte Atlanta
Atlanta 1 6 0.333333 Normalization Atlanta 0.24
Birmingham 0.166667 1 0.111111 Birmingham 0.04
Charlotte 3 9 1 Charlotte 0.72
Summat: 4.166667 16 1.444444 Summat: 1

c. infrastructure

Pairwise comparison matrix Normalized Matrix (cell value / column s


Atlanta Birmingha Charlotte Atlanta
Atlanta 1 0.333333 1 Normalization Atlanta 0.2
Birmingham 3 1 7 Birmingham 0.6
Charlotte 1 0.142857 1 Charlotte 0.2
Summat: 5 1.47619 9 Summat: 1

d. transport

Pairwise comparison matrix Normalized Matrix (cell value / column s


Atlanta Birmingha Charlotte Atlanta
Atlanta 1 0.333333 0.5 Normalization Atlanta 0.166667
Birmingham 3 1 4 Birmingham 0.5
Charlotte 2 0.25 1 Charlotte 0.333333
Summat: 6 1.583333 5.5 Summat: 1
Combined Preference Vector of the candidates
M.base Income Infrast.
Atlanta 0.501263 0.28192307692 0.178973
Birmingham 0.118476 0.05980769231 0.685066
Charlotte 0.380261 0.65826923077 0.135962
Summat: 1 1 1

2 Comparison of criteria

Pairwise comparison matrix for criteria Normalized Matrix (cell value / co


M.base Income Infrast. Transp M.base
M.base 1 0.2 3 4 M.base 0.151899
Income 5 1 9 7 Income 0.759494
Infrast. 0.333333 0.111111 1 2 Infrast. 0.050633
Transp 0.25 0.142857 0.5 1 Transp 0.037975
6.583333 1.453968 13.5 14 1

3 Finding the actual site winner

Combined Preference Vector of the candidates Preference Vector


M.base Income Infrast. Transp
Atlanta 0.501263 0.281923 0.17897252 0.156034 M.base
Birmingha 0.118476 0.059808 0.68506571 0.619617 Income
Charlotte 0.380261 0.658269 0.13596177 0.224349 Infrast.
Summat: 1 1 1 1 Transp

4 Contistency

Pairwise comparison matrix for criteria Preference Vector


M.base Income Infrast. Transp
M.base 1 0.2 3 4 M.base
Income 5 1 9 7 Income
Infrast. 0.333333 0.111111 1 2 Infrast.
Transp 0.25 0.142857 0.5 1 Transp
6.583333 1.453968 13.5 14

a new vector Preference Vector


0.83272519382 Degree
2.85239686254 M.base 0.19934745682
0.34740102872 Income 0.65348331583
0.24736293318 Infrast. 0.08599583557
Transp 0.06117339178

Observe the Random Inde


SOLUTION OF AHP PRACTICE QS.

Matrix (cell value / column sum) Preference Vector


Birmingha Charlotte Degree
0.333333 0.625 Row Average Atlanta 0.501263
0.111111 0.0625 Birmingham 0.118476
0.555556 0.3125 Charlotte 0.380261
1 1 Summat: 1

Matrix (cell value / column sum) Preference Vector


Birmingha Charlotte Degree
0.375 0.230769 Row Average Atlanta 0.281923
0.0625 0.076923 Birmingham 0.059808
0.5625 0.692308 Charlotte 0.658269
1 1 Summat:

Matrix (cell value / column sum) Preference Vector


Birmingha Charlotte Degree
0.225806 0.111111 Row Average Atlanta 0.178973
0.677419 0.777778 Birmingham 0.685066
0.096774 0.111111 Charlotte 0.135962
1 1 Summat: 1

Matrix (cell value / column sum) Preference Vector


Birmingha Charlotte Degree
0.210526 0.090909 Row Average Atlanta 0.156034
0.631579 0.727273 Birmingham 0.619617
0.157895 0.181818 Charlotte 0.224349
1 1 Summat: 1
the candidates
Transp
0.156034
0.619617
0.224349
1

lized Matrix (cell value / column sum) Preference Vector


Income Infrast. Transp Degree
0.137555 0.222222 0.28571429 M.base 0.199347
0.687773 0.666667 0.5 Income 0.653483
0.076419 0.074074 0.14285714 Infrast. 0.085996
0.098253 0.037037 0.07142857 Transp 0.061173
1 1 1 1

Preference Vector
Degree
0.199347 Atlanta 0.309093 2
0.653483 Birmingham 0.159518 0.1595181898 3
0.085996 Charlotte 0.531388 6
0.061173 1

Preference Vector
a new vector
0.199347 0.83272519
0.653483 2.85239686
0.085996 0.34740103
0.061173 0.24736293

M.base 4.17725517 Average = SUM / n n=no. of criteria


Income 4.36491153 n= 4
Infrast. 4.03974246
Avg 4.156386
Transp 4.04363606
16.6255452

[CI = (Average - n) / (n-1)]

CI 0.052129

Observe the Random Index (RI) from the random experiment table against your "n"

CI/RI 0.057921

In general, the degree of consistency is satisfactory if CI/RI =< 0.10. If CI/RI > 0
me
0.052129

CI/RI =< 0.10. If CI/RI > 0, then there are probably serious inconsistencies, and the AHP results may not be
meaningful.

You might also like