VOL. 11, NO.

11, JUNE 2016 ISSN 1819-6608

ARPN Journal of Engineering and Applied
Sciences

www.arpnjournals.com

MCDM-AHP METHOD IN DECISION MAKINGS

Nor Filianie Aziz, Shahryar Sorooshian and Fatimah Mahmud
Faculty of Industrial Management, University Malaysia Pahang Kuantan, Pahang,
Malaysia E-Mail: [email protected]

ABSTRACT
Analytic Hierarchy Process (AHP) is among the most popular methods of decision-makings. This article contains
the tutorial document composed of a short discussion on AHP presented at the International conference on “Proceedings
of Engineering Technology International Conference”. The conference would be held at Bali, Indonesia during August
10-11, 2015. The paper starts with the short review of Multi-criteria Decision Making and AHP. Then it discusses about
the techniques and formulae that are used in the AHP decision-making method. Lastly, this paper recommends AHP to
future researchers and professionals with highlights of the reasons to use the methodology process when engaged in
complex decision-making problems.

Keywords: multi-criteria decision-making (MCDM), multi-alternative decision making (MADM), analytic hierarchy process (AHP),
and decision techniques.

1. INTRODUCTION Technique (SMART); and others, for example: [4, 5, 6].

Multi-Criteria Decision Making (MCDM) Notwithstanding all these methods AHP method is
models are suitable for evaluating and making decision regarded as the most famous MCDM tool for decision
for the best alternatives options in order to choose the making problems based on literature studies.
perfect criteria [1]. This involves a general class of
operations research models, which considers problems in 2. AHP METHOD
decision making in the presences of many decision
According to Triantaphyllou and Mann [7], the
criteria. There are two types of the MCDM, which are;
nice mathematical properties of AHP have attracted many
Multi-Objective Decision Methods (MODM) and Multi-
researchers’ interest and AHP input data are easy to
Attribute Decision Methods (MADM) [1]. The Multi-
obtain. Analytic Hierarchy Process (AHP) that is created
Objective Decision Methods (MODM) is an approach,
under Multi- Criteria Decision Making (MCDM) is
which uses mathematical optimization technique and
composed of techniques are suitable for ranking of critical
mostly involves analysis procedure related to calculation
management problems [8]. The Analytic Hierarchy
design process. MODM usually involves maximization of
Process (AHP) was introduced by Saaty [9] since 1970’s.
mathematical problems involving function that is more
The AHP method is ranking process that is used in
objective which need to be simultaneously optimized.
making group decision and is widely used around the
Especially MODM was created to solve problems in
world in a variety of fields such as business, government,
mathematical programming and design via the best
industry, education, health, and others. The method also
alternatives. However, Multi-Attribute Decision Methods
allows for consistency test in judgment making room to
(MADM) is an approach developed for selecting the best
check and reduce inconsistencies in opinions or
criteria or alternative(s). This is used in decision-making
judgments. The scales of ratio and consistency index are
problems involving a number of decision-making
derived from the principal Eigen vectors and Eigen value
alternatives. This model is based on the list of criteria
respectively. The method focuses on prioritizing the
chosen, its parameters, variables that one wishes to
selection criteria, and distinguishing the more important
monitor in decision-making process [2]. The category of
criteria from the less important ones. Although some
MCDM has been used for selecting a minimum number
researchers argue some disadvantage of AHP [6, 10],
of alternatives. According to Ermatita, et al., [3] two
AHP is simple method with focus placed on peer to peer
levels are relevant with MADM and these are: (a)
comparisons that are suitable to evaluate both qualitative
Aggregation implementation: The decision that reflects
and quantitative design [11]. More also, AHP method uses
the result equivalent for all areas on each alternative is
judgment to analyze the data.
developed.
(b) Alternatives implementation: The alternatives ranking
3. AHP CONCEPTUAL
for the aggregation of the result makers. There are many
Although AHP is a very popular decision making
other methods which are used under MADM such as;
method, authors only find few articles to algorithmically
Analytical Hierarchy Process (AHP); Simple Additive
review AHP applications (i.e. [12]). This article tries to
Weighting (SAW); Ordered Weighted Averages (OWA);
fill the need of application tutorials. This paper discusses
Technique for Order Preference by Similarity to the Ideal
four main steps of AHP that could be used in decision-
Solution (TOPSIS); Elimination et Choice Translating
making problems and these include; problem modeling,
Reality (ELECTRE); Decision trial and Evaluation
weight valuation, weight aggregation, and sensitivity
Laboratory; Decision Trial and Evaluation Laboratory
analysis. The steps start with hierarchy construction
(DEMATEL); The Simple Multi Attribute Rating
where objective

1
are highlighted and criteria and alternatives identified. AHP also need to listed some criteria and
Then comparisons of the criteria and alternatives and their alternative conceptual. It is because to easier AHP ranking
relative rankings are calculated [11]. Firstly, the criteria in and evaluate [13]. There are three steps to illustrate AHP
the hierarchy construction must be determined and this data hierarchy [14]: goal/objective, alternative, and
can be done through different stages from the top level, criteria/sub-criteria will been seeing Table-2 below.
middle level, and to the bottom level. Top level: This set
the goal or the objective and is on upper stage in the Table-2. AHP reference conceptual of criteria
hierarchy; Middle level: made up of both criteria and sub- for selecting alternative.
criteria in the second stage; and Bottom level: made up of
down stage including alternatives. In addition, the
weighting of criteria, sub-criteria and alternatives must be
evaluated. This process or function is vital to group
decision making as its presents important top elements for
target achievement. AHP is simple peer-to-peer
differences and to determine the best criteria and/or
alternatives there should be focus on weight of the
factors. More also, comparison matrix will have to be
expanded by calculating the weights of the criteria and
local weight of the alternatives to obtain the matrix
weight. Table-1; shows the important evaluation scales of
pair-wise comparison. It is related to AHP process for
making a decision such as ranking size to evaluate the
scales. In this table, we have five related important values
that will be transforming to matrix. In this regard, this
article recommends Expert Choice Software for the easier
AHP weighting and all of the criteria and sub-criteria that
are created should be analysis and evaluate using
computer software.
4. AHP PROCESS
Table-1. Evaluation scales of pair-wise comparison [8]. Although, the technique of AHP processes have
been explained. There are forth mathematical calculations
in Analytic Hierarchy Process that needs to be followed as
stated by Richard et al. [12] and Alam, et al. [14]: Firstly,
the process of hierarchy ranking were be constructed
starting from goal or objective to achieve and identify
and/or determine both of criteria and alternatives.
Secondly, the pair-wise comparison matrices of
alternatives and criteria must been constructed as
illustrated in the formulae below.

The above formula deals with the matrix of

dimension (n x n), where n = comparison number of
factors. The elements value that are related to the diagonal
of the matrix is equal to 1 such as Aij = 1. Based on AHP,
it is possible for preference that is reciprocal and this is
expressed by Aij=1/Aji for i ≠j. For example if factor of i-
th is, x times more important than the j-th factors (e.g: Aij
= x), meanwhile the automatically assume that j-th factors
is 1/ x as important between i-th element (aji= 1/x) and/or
Aji= 1/aji.
Thirdly, the calculations of criteria weight and
alternative local weight that are selected from existing Consistency Ratio.
matrices are calculated through the use normalization
procedure. Equations of criteria weight and alternatives
local weight are explained as below.

Calculation of total data of each row.

Where [15];
CR ≤ 10% = Inconsistent is acceptable
CR > 10% = Revise process of subjective judgment.
Normalization of local weight. The random consistency index is an important
value to calculate the CR. Random Index (RI) is value to
calculate the CR. Table-3 show the value index of RI
from n = 1 until n = 15 with RI value such as table
shown. Table 3 show the value index of RI from n = 1
Eigen Vector. until n = 15 with RI value such as table shown.

Table-3. The Random Consistency Index [15, 16].

Weight Vector.

Lastly, obtained the alternatives global weight that are

synthesize with the local weight

From above equation, B explains alternatives of

local weight (row) and the criteria of local weight
(column). Also, V explained transpose of criteria local
weight and the Global weight is derived by multiplying
the matrices of B and V. For easier AHP weighting, all of
alternatives and criteria that are created should be
analyzed and evaluated using computer software. More 5. CONCLUSIONS
also, consistency is an important aspect of AHP and must In this paper, the authors presented a short
be checked. The AHP consistency test has been review of Multi-criteria Decision Making (MCDM),
represented in the formulae below. However, the Multi- Attribute Decision Methods (MADM), and
consistency of data can also be tested using Expert Choice Analytic Hierarchy Process (AHP). They then discussed
Software. about the techniques and formulae that are used in the
AHP decision-making method. It was observed that,
Maximal Eigenvalue. Analytic Hierarchy Process (AHP) method is suitable for
ranking and analyzing complex decision-making
problems. AHP is also regarded as one of perfect and
easiest method under MCDM because it is easy to use and
makes room for checking and reducing inconsistencies in
opinion(s). There is therefore sufficient evidence based on
the formulae, along with comprehensive literature review
by the authors to strongly recommend AHP to future
researchers and
professionals to use the methodology process when [10] Sorooshian, S. 2015. Alternative Method for
engaged in complex decision-making problems involving Evaluation of DaGang Deep Drilling Application.
many criteria, sub-criteria, and alternatives. Also, this Electronic Journal of Geotechnical Engineering, Vol.
paper recommends Expert Choice Software and other 20, Bund. 13, pp. 5209-5212.
software that may be useful and will make weighting, all
of alternatives and criteria easier. [11] Zamani, R. and Yousefi, P., 2013. Optimal decision
making approach for selecting effort estimation
REFERENCES model. International Journal of Machine Learning
and Computing, Vol. 3, No. 4, pp.1-4.
[1] Sorooshian, S. and Dodangeh, J. 2013. A Moderated
Practice for Strategy Implementation Analysis” [12] Ansah, R. H., Sorooshian, S., and Mustafa, S, 2015,
American Journal of Applied Sciences, 10 (9): 1039- Analytic Hierarchy Process Decision Making
1042. Algorithm. Global Journal of Pure and Applied
Mathematics. Vol. 11, No. 4, pp. 2393-2400.
[2] Bohanec, M. and Rajkovič, V. 1999. Multi-Attribute
Decision Modeling: Industrial Applications of DEX. [13] Anvari, A., Zulkifli N., Sorooshian S., and
Informatica, vol. 23, pp. 487–491. Boyerhassani O., 2014. An integrated design
methodology based on the use of group AHP-DEA
[3] Ermatita, Hartati, S., Wardoyo, R., and Harjoko, A. approach for measuring lean tools efficiency with
2011. ELECTRE methods in solving group decision undesirable output. The International Journal of
support system bioinformatics on gene mutation Advanced Manufacturing Technology, Vol. 70, No.9-
defection simulation. International Journal of 12, pp.2169-2186.
Computer Science and Information Technology
(IJCSIT), Vol. 3, No.1, pp.1-13. [14] Alam, M. N., Jebran, J. K., and Hossain, M. A., 2012.
Analytical Hierarchy Process (AHP) Approach on
[4] Sorooshian, S. and Dodangeh, J. 2013. Modeling on consumers preferences for selecting telecom
Performance Drivers of Project Management. operators in Bangladesh. Information and Knowledge
Advances in Environmental Biology, 7(13), Pages: Management, Vol 2, No.4, pp. 7-19.
3890-3894.
[15] Karami, A. 2011. Utilization and comparison of
[5] Sorooshian, S. and Suziyana, M. D. 2013. Analysis multi-attribute decision making and techniques to
on Factors of Non- Compliance of Halal Standard. rank Bayesian network options. Master Degree
Journal of Engineering and Applied Sciences, 8(9- Project in Informatics, One Year Level ECTS 30, 1-
12): 280-281. 51.
[6] Sorooshian, S., Wenping, L., and MD Yusof, I. 2015. [16] Dolalah, D., AL-Oqla, F., and Hayajneh, M. 2010.
Landslide Susceptibility Mapping: A Technical Note. Application of the Analytic Hierarchy Process (AHP)
Electronic Journal of Geotechnical Engineering, Vol. in Multi- criteria Analysis of the selection of cranes.
20, Bund. 22, pp. 12547 12550. Jordan Journal of Mechanical and Industrial
Engineering. Vol. 4, No. 5, 567-578.
[7] Triantaphyllou, E. and Mann, S. 1995. Using the
Analytic Hierarchy Process for Decision Making in
Engineering Applications: Some Challenges.
International Journal of Industrial Engineering:
Application and Practice, Vol. 2 (1), pp. 35 – 40.

[8] Cabala and Pawel, 2010. Using the Analytic

Hierarchy Process in Evaluating Decision
Alternatives. Operations Research and Decision,
No.1, pp.5-23.

[9] Saaty, T. L. 1980. The Analytic Hierarchy Process.

McGraw- Hill International, New York, NY, USA.