Decision Making Using The Analytic Hierarchy Process (AHP) A Step by Step Approach
Decision Making Using The Analytic Hierarchy Process (AHP) A Step by Step Approach
Decision Making Using The Analytic Hierarchy Process (AHP) A Step by Step Approach
net/publication/322887394
Decision Making Using the Analytic Hierarchy Process (AHP); A Step by Step
Approach
CITATIONS READS
62 13,443
1 author:
Hamed Taherdoost
University Canada West
104 PUBLICATIONS 2,310 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Hamed Taherdoost on 02 February 2018.
Abstract: - Analytical Hierarchy Process is one of the most inclusive system which is considered to make
decisions with multiple criteria because this method gives to formulate the problem as a hierarchical and
believe a mixture of quantitative and qualitative criteria as well. This paper summarizes the process of
conducting Analytical Hierarchy Process (AHP).
Key-Words: - Analytical Hierarchy Process (AHP), Research Methodology, Hierarchical Tree, AHP Survey /
Questionnaire, and AHP Formulas.
Table 1: Scores for the importance of variable To reach a convergence among the set of answers in
to successive repetition of this process, calculation
Importance should be repeated several times in order to take a
Definition of Importance Scale
Scale decision when facing an incompatible matrix. Then,
1 Equally Important Preferred the following formula is applied to transform the
2 Equally to Moderately Important Preferred raw data into meaningful absolute values and
3 Moderately Important Preferred normalized weight w = (w1, w2, w3… wn):
4 Moderately to Strongly Important Preferred
5 Strongly Important Preferred
6 Strongly to Very Strongly Important Preferred Aw = λmax w, λmax ≥ n
7 Very Strongly Important Preferred
8 Very Strongly to Extremely Important Preferred
a jwj n
9 Extremely Important Preferred
λmax
w1
A={aij} with aij=1/ aij
Table 2: Sample AHP Questionnaire
A: pair wise comparison
How important are the following security criteria in w: normalized weight vector
comparison
λmax : maximum eigen value of matrix A
aij: numerical comparison between the values i and j
Factor 9 8 7 6 5 4 3 2 1 2 3 4 5 6 7 8 9 Factor
Privacy 9 8 7 6 5 4 3 2 1 2 3 4 5 6 7 8 9 Reliability
In the next step, in order to validate the results of the
Privacy 9 8 7 6 5 4 3 2 1 2 3 4 5 6 7 8 9 Validation
AHP, the consistency ratio (CR) is calculated using
the formula, CR = CI/RI in which the consistency
Privacy 9 8 7 6 5 4 3 2 1 2 3 4 5 6 7 8 9 Verification
index (CI) is, in turn, measured through the
Privacy 9 8 7 6 5 4 3 2 1 2 3 4 5 6 7 8 9 Integrity
following formula:
Privacy 9 8 7 6 5 4 3 2 1 2 3 4 5 6 7 8 9 Confidentiality
Privacy 9 8 7 6 5 4 3 2 1 2 3 4 5 6 7 8 9 Availability
max n
CI
The data analyze procedure involves the following n 1
steps. First the pairwise comparison matrix which is
called matrix A is extracted from the data collected
from the interviews. The principal right eigenvector The value of RI is related to the dimension of the
of the matrix A is computed as ‘w’. matrix and will be extracted from Table 3. It should
be noted that consistency ratio lower than 0.10
If aik . a kj = a ij is not confirmed for all k, j, verifies that the results of comparison are
and i the Eigenvector method is selected acceptable.
(Jalaliyoon, et al., 2012).
Table 3: The value of Random Consistency Index,
If the matrix is incompatible and in case of Source: Golden and Wang (1990)
incomplete consistency, pair comparisons matrix
cannot be used normalizing column to get Wi. Dimension RI
1 0
For a positive and reversed matrix, Eigenvector 2 0
technique can be used which in it: 3 0.5799
4 0.8921
5 1.1159
e T (1,1, ,1) 6 1.2358
7 1.3322
Ak . e A 8 1.3952
W lim k k 9 1.4537
eT . A k . e 10 1.4882
References: