The Analytic Hierarchy Process: Structuring,
Measurement, and Synthesis
John Wang
Montclair State University, USA
Chandana Chakraborty
Montclair State University, USA
Huanyu Ouyang
The People’s Hospital of Jiangxi Provence, China
INTRODUCTION
The challenges of evaluation and decision making are
encountered in every sphere of life and on a regular
basis. The nature of the required decisions, however,
may vary between themselves. While some decisions
may reflect individual solutions on simple problems,
others may indicate collaborative solutions on complex
issues. Regardless of their distinctive nature, all deci-
sions are outcomes of a mental process. The process
involves careful evaluation of merits of all the available
options leading ultimately to the choice of a single
solution. Numerous efforts have been made in the lit-
erature to develop decision models ideal for choosing
the best solution for a given problem. The dilemma in
using these decision models, however, can hardly be
avoided. With differences in underlying methodology,
each model serves a specific decision-making need of the
decision maker. In the absence of a universal framework
suitable for handling a variety of problems, decision
makers are often required to identify the model best
suited for their particular need. Furthermore, they need
to take account of the advantages and disadvantages
associated with the chosen model.
Recognizing the difficulty of model selection,
Thomas L. Saaty, the mathematician, developed a
decision-making approach known commonly as the
analytic hierarchy process (AHP), which relies mainly
on the innate human ability to make sound judgments
about small problems (Saaty, 1994, 1996; Saaty &
Alexander, 1981). The AHP is a popular method for
assessing multiple criteria and deriving priorities for
decision-making purposes. This model is different from
all other models in that it is able to handle both tangible
and intangible attributes of the decision maker. In addi-
Copyright © 2008, IGI Global, distributing in print or electronic forms without written permission of IGI Global is prohibited.
A
tion, it has the ability to monitor the consistency with
which a decision maker makes his or her judgments
(Roper-Lowe & Sharp, 1990). Unlike other available
models, AHP can be universally adapted to a wide
range of problems and, hence, is an excellent choice
for decision making in diverse problems in the fields
of quality control, finance, balanced scorecard, and
forecasting. It is for these reasons that AHP is now one
of the most highly regarded and used models in a wide
range of organizations including major corporations,
government agencies, and academia (Liedtka, 2005).
BACKGROUND
Dr. Thomas L. Saaty worked for the U.S. Department
of State in the 1960s. It was during this time that he
realized that many of the available models were too
general and abstract for application in a wide range of
decision-making needs (Forman & Gass, 2001). In his
attempt to create a universal framework as opposed
to a specialized framework for modeling real-world
problems, Saaty developed the AHP model in the 1970s
while working as a professor at the Wharton School of
Business (Saaty & Vargas, 1991). As narrated by Saaty,
he utilized the methodology taught by his grandmother
in developing the model. The methodology consisted
of breaking down a complex problem and weighing the
decision options against each other (Palmer, 1999).
AHP has three primary functions: structuring
complexity, measurement, and synthesis. The first
function, structuring, involves configuration of the
problem into a hierarchy that describes the problem.
With the overall goal placed at the top, the main at-
tributes are placed at a level below the top one. These
 The Analytic Hierarchy Process: Structuring, Measurement, and Synthesis
attributes can further be subdivided in consecutive
lower levels thereby simplifying the decisions at hand.
The second function, measurement, involves deriving
weights for the lowest level of attributes. This is done
by a series of pair-wise comparisons in which every
attribute on each level is compared with its siblings in
terms of its importance to the parent. Following this,
the options available to the decision maker are scored
with respect to the attributes. Finally, matrix algebra
is used to calculate the final score for each available
option (Roper-Lowe & Sharp, 1990).
MAIN fOCUS
Benefits of AHP
The AHP has the ability to elicit decision maker’s
responses on the relative importance of the problem
in three different ways: numerically, verbally, and
graphically. These elicited responses are inspired by a
pair-wise comparison process. With an option to sub-
mit responses in alternative formats, decision makers
using the AHP model provide meaningful responses
and, thus, produce better results.
Structuring the problems into a hierarchy with the
AHP allows the decision makers to deal with the as-
sociated complexity in a simple way. The methodology
reflects Saaty’s observation that human beings deal with
complexity simply by structuring it into homogeneous
clusters of factors (Forman & Gass, 2001). With prob-
lems broken up into clusters, individuals find it easier
to evaluate the importance of each alternative available
for solution. Other scientists have shared Saaty’s views
in this respect.
The effectiveness of structuring incorporated in the
AHP model is backed up by evidence. Most of today’s
organizations use a hierarchy structure in order to ease
the decision-making process.
Hierarchy building is a powerful instrument at the
initial stages of setting up problems and considering
alternatives. By allowing information to be organized,
the structure allows the decision maker to better un-
derstand the interaction of the elements of a problem
(Gass, 1985). Additionally, the structure minimizes the
possibility of overlooking elements of the problem; is-
0
sues and ideas ignored at previous levels can become
apparent at advanced levels.
The ability to measure consistency is also a major
strength of the AHP. AHP uses the eigenvalue technique
that allows for computation of a consistency measure,
an estimated arithmetical indicator of the inconsisten-
cies or intransitivity in a set of pair-wise ratings (War-
ren, 2004). This measure is popularly referred as the
consistency index. Pair-wise comparison ratings are
considered consistent and acceptable as long as the
consistency ratio (CR) is lower than 0.10. A ratio higher
than 0.10 warrants additional review and evaluation
of the results. The management team in charge of the
evaluation process can take precautionary measures to
avoid costly mistakes of repeated trials. In particular,
the team should take precaution when dealing with
a large number of alternatives capable of producing
inconsistencies.
The AHP is also well known for its ability to
compare intangible and tangible factors. This is easily
accomplished through pair-wise comparisons with a
nine-point scale. Even though it is feasible to do these
comparisons, it has been suggested not to mix both fac-
tors in the same hierarchy. For cases that fail to satisfy
this condition, an alternative approach of using a link
attribute can prove useful. This link attribute helps to
make a meaningful comparison of both tangible and in-
tangible factors (Roper-Lowe & Sharp, 1990). The link
attribute is weighed against the tangible and intangible
factors in pair-wise comparisons. In the following stage,
the intangible factors can be weighed against tangible
ones by scaling the weight of the link attribute.
Finally, AHP is also useful in providing records about
circumstances surrounding each decision. Records can
be reviewed at a later point in time to determine how
and why a particular decision was arrived at. This can
become very useful when evaluating decisions that
were previously made but need to be considered again
for changes in circumstances. For example, additional
information about the alternatives used in pair-wise
comparisons can become available in the future. Deci-
sion makers can quickly process this new information
in order to reevaluate original scores and measure the
impact of new data on their initial decision. In cases
where organizations are required to make similar de-
cisions on an ongoing basis, the same hierarchy can
be reused as a background for constructing a new
hierarchy.
The Analytic Hierarchy Process: Structuring, Measurement, and Synthesis
Shortcomings of AhP
Rank Reversal
One of the most highly commented about weaknesses
of the AHP has been rank reversal. Rank reversal basi-
cally refers to a change in the rankings or the ordering
of the alternatives if an additional alternative is added
or an existing one is removed (Warren, 2004). It usu-
ally happens because of the implicit nature of the AHP.
Under rank reversal, the linkage between alternatives
is caused by normalization of the eigenvectors. Hence,
the addition of an alternative or the removal of an al-
ternative leads to a change in the weights of the other
alternatives. This is opposed to the MAUT (multiat-
tribute utility theory) model, which has an explicit
nature and for which the addition of new alternatives
does not affect the rank of the other alternatives.
Granted its existence, rank reversal is considered a
secondary problem and its effects are seen only when
the alternatives are closely related. There are additional
questions of closed and open systems (Forman & Gass,
2001). A closed system is one in which there is a limited
number of resources and, hence, one needs to allocate
them evenly between the alternatives. Let us say that
a company has limited construction supply and wants
to allocate it efficiently between three different houses
they are building. Given resource limitation, an addi-
tional housing project would require resources to be
reallocated between the houses. Thus, in the real world
where managers and CEOs are looking for increased
efficiency with a given amount of resources, AHP ap-
pears to be a common choice. For cases representing
open resources or unlimited resources, AHP can cause
rank reversal.
Subjectivity and Time Consumption
An additional weakness of the AHP method is that it
is subjective. In performing pair-wise comparisons,
it often becomes apparent that different people have
different ways of evaluating a particular component.
Consequently quantifying components accurately can
become extremely difficult. Recently, a solution for this
problem was designed by integrating methods that deal
with differences in evaluating techniques. An example
of such a method is the fuzzy logic theory.
Another disadvantage of the AHP is the redundancy
of entering rankings for different alternatives in pair-
wise comparisons (Ishizaka, 2004). As the number of
alternatives, criteria, and hierarchical levels increase, A
so does the number of comparisons needed to evaluate
them. For example, a hierarchy with five criteria and
nine alternatives would require the decision maker to
perform 190 comparisons. With more alternatives, the
total number of comparisons could increase infinitely.
Clearly, therefore, AHP can become a time-consuming
tool for evaluating solutions, especially in group set-
tings, where some decisions can potentially turn into
lengthy debates. Methods have been developed to ad-
dress this issue by reducing the number of comparisons;
however, making fewer comparisons can result in loss
of accuracy. Many professionals suggest subdividing
the alternatives into smaller clusters that reduce the
number of comparisons needed. Many others have
suggested that the time-consumption aspect of the AHP
has very little relevance in today’s technologically
advanced societies; numerous software technologies
are available to compute AHP models.
A Specific Example of Application of
AhP
Let us now see an example of how AHP actually works
and what kind of problems it can help solve in real-life
business.
A hospital has a vacancy in its marketing depart-
ment and is looking for a potential new employee to
fill the position. The hospital is specifically looking
for four different characteristics in its candidates: (a)
experience, (b) salary requirement, (c) education, and
(d) personality. The company has narrowed down
its choices to three final candidates (John, Leon, and
Leigh) and will use the AHP model to select the best
candidate based on the identified criteria.
• Our Goal is: Select the best candidate for the
marketing position
• Our Criteria are: Experience, Salary Requirement,
Education, and Personality
• Our Alternatives are: John, Leon, and Leigh
There are two types of AHP models that we need
to consider:

 The Analytic Hierarchy Process: Structuring, Measurement, and Synthesis

1. Consistent: In this model, the following condi-
tions are true:
a. Transitive property: If A is better than B
and B is better than C, then A is better
than C.
b. Numerically consistent ratings: If
A = 3B and A = 2C, then B= 2/3C
c. The value of the CR is 0.
2. Inconsistent: This model pays no attention to
transitive property or numerical consistency. However,
to be in acceptable range, the value of CR should be
less than 0.10.
Step 1: Create rankings for criteria and alternatives.
The initial step is to compare the four criteria and decide
which skills are more important for this marketing job.
The comparisons are done via the use of a rating scale
as proposed by Saaty himself.
So, if experience is more important than personality,
we can give it a rating of anything from 2 to 9 based upon
how much more important experience is than personal-
ity. A rating of 3 means that experience is moderately
more important than personality. All of these decisions
are made by the higher level personnel who have the
ability to evaluate the importance of each skill in the
context of their company’s needs. Let us suppose the
hospital develops the following pair-wise comparison
matrix for the set of criteria referred to above.
Step 2: Normalize and figure out weights.
To normalize:
• Divide each column value by the total sum of the
column and
• Take the average of each row
The sums of the columns are 2.25, 9, 3.375, and 6.75.
After dividing each column value by their respective
sums, we get the matrix in Table 3.
As can be seen from Table 3, since all the rows in a
consistent AHP are multiples of each other, we really
only need to normalize one column to get the weights.
Hence, taking the average for each of the rows above,
we get the following weights for the criteria:
Experience: 0.44; Salary Requirement: 0.11; Educa-
tion: 0.298; Personality: 0.147
It is therefore apparent that the hospital is assigning
major emphasis on the experience of the candidates;
the weights are approximately 44% for experience,
29.8% for education, 14.7% for personality, and 11%
for salary requirement.
Step 3: Do the same for each of the alternatives and
figure out the weights. Following the same methodol-
ogy used in Step 2, we need to figure out the weights
for the other alternatives. So if Employee A has more

Table 1.
Intensity of Importance Definition
1 Equal Importance
3 Moderate Importance
5 Strong Importance
7 Very Strong Importance
9 Extreme Importance
2, 4, 6, 8 For compromise between the above values

Table 2.
Criteria Experience Salary Req. Education Personality
Experience 1 4 1.5 3
Salary Req. 0.25 1 0.375 0.75
Education 0.67 2.67 1 2
Personality 0.33 1.33 0.5 1
CR = 0.0000
Since this is a consistent model, note that the value of CR as computed by the software is 0.0000.


The Analytic Hierarchy Process: Structuring, Measurement, and Synthesis

Table 3.
Criteria Experience Salary Req. Education Personality A
Experience 0.444 0.444 0.444 0.444
Salary Req. 0.111 0.111 0.111 0.111
Education 0.298 0.297 0.296 0.296
Personality 0.147 0.148 0.148 0.148

Table 4. Weights based on employee experience

Experience John Leon Leigh
John 1 0.5 1.5
Leon 2 1 3
Leigh 0.67 0.33 1
CR = 0.0000
After normalization, the weights are the following.
John: 0.273; Leon: 0.545; Leigh: 0.182

Table 5. Weights based on employee salary requirement

Salary Req. John Leon Leigh
John 1 5 3
Leon 0.2 1 0.6
Leigh 0.33 1.67 1
CR = 0.0000
After normalization, the weights are as follows.
John: 0.652; Leon: 0.130; Leigh: 0.218

Table 6. Weights based on employee education

Education John Leon Leigh

John 1 0.75 0.25
Leon 1.33 1 0.33
Leigh 4 3 1

CR = 0.0000
After normalization, the weights are the following.
John: 0.158; Leon: 0.210; Leigh: 0.632

Table 7. Weights based on employee personality

Personality John Leon Leigh

John 1 0.5 0.75
Leon 2 1 1.5
Leigh 1.33 0.67 1

CR = 0.0000
After normalization, the weights are as follows.
John: 0.231; Leon: 0.462; Leigh: 0.308


 The Analytic Hierarchy Process: Structuring, Measurement, and Synthesis
Table 8.
Weights for Each Candidate Based on Alternatives
Experience Salary Req. Education
John 0.273 0.652 0.158
Leon 0.545 0.13 0.21
Leigh 0.182 0.218 0.632
experience than Employee B, we will give Employee
A a higher rating for experience than Employee B.
Using this strategy, the following weights have been
calculated for each of the alternatives.
Step 4: Rank the candidates overall by priority.
Synthesis of the previous steps yields the following
weights:
To compute overall priority for each candidate, we
will multiply the weight for each alternative by the
weight for the corresponding criterion. So, the overall
priority for the candidates would be the following:
John = (0.273 * 0.44) + (0.652 * 0.11) + (0.158 * 0.298)
+ (0.231 * 0.147) = 0.273
Leon = (0.545 * 0.44) + (0.130 * 0.11) + (0.210 *
0.298) + (0.462 * 0.147) = 0.385
Leigh = (0.182 * 0.44) + (0.218 * 0.11) + (0.632 *
0.298) + (0.308 * 0.147) = 0.338
To summarize, the overall weights for the employees
are as follows:
John: 0.273; Leon: 0.385; Leigh: 0.338
Hospital executives have selected Leon to fill the
position for their marketing vacancy because he has
the highest ranking (0.385).
The above demonstration is based on a consistent
model since an inconsistent model has a similar pro-
cess.
Examples across Different Areas and
Organizations
Over the years, AHP has been successfully applied
to a broad range of difficult situations requiring more
of a formal approach. Some of the examples include

Weights for Criteria
Personality Experience 0.44
0.231 Salary Requirement 0.11
0.462 Education 0.298
0.308 Personality 0.147
selection among competing alternatives, allocation of
resources, and forecasting. Especially in multicriteria
environments, AHP has widely been chosen to solve
complex problems and forecast outcomes.
Much of its continued success can be attributed to
Dr. Saaty’s software company Expert Choice, which
develops computer applications that utilize AHP and
has been successful in helping countless organizations
save millions of dollars. In addition, the software has
made implementation of the AHP process less exhaust-
ing. The software keeps track of thousands of pair-wise
comparisons while allowing the user to move and update
variables. Also, calculations that would otherwise be
too unwieldy and time consuming to do manually can
now be done with relative ease.
Dr. Saaty has also developed a new version of his
Expert Choice applications for businesses called Aliah.
It is specifically being targeted for the Fortune 500
companies as a soup-to-nuts tool for business develop-
ment. The aim is to make AHP the de facto standard
of strategic planning (Palmer, 1999).
NASA: Optimal Project Selection
As a result of increased public awareness with regard to
government spending on space exploration, Congres-
sional mandates were put in place to hold the National
Aeronautic and Space Administration (NASA) more
accountable in the process of evaluating and selecting
the best projects that maximize the return from the
taxpayers’ funds. NASA was forced to abandon its
unstructured and intuitive process and replace it with a
multicriteria group-decision model, known as CROSS
(consensus-ranking organizational support system),
which uses the AHP to help NASA scientists evaluate
advanced technological projects (Tavana, 2003). With
its successful implementation, NASA shortened the
evaluation time, improved the quality of decisions, and
reduced the number of costly decision reversals that
The Analytic Hierarchy Process: Structuring, Measurement, and Synthesis
were present under the old method. More importantly,
it provided members of the decision-making group with
a structured approach for systematic evaluation.
AHP in Health Care
The health care system consists of complex areas that
are unique in their characteristics and that are a part of
many dimensions fitted into singular unit; therefore,
it is very difficult to approach the measurement of the
performance of health care services. Most health care
system delivery is evaluated through three categories
of measurement—structure, process, and outcome—in
order to improve the performance of multispecialty
tertiary care. According to Hariharan, Dey, Moseley,
Kumar, and Gora (2004), the AHP model was applied
to two tertiary care teaching hospitals in Barbados and
India. The results of application identified specific areas
where neither hospital performed well; recommenda-
tions were provided for improvement in those areas.
This study has established AHP as a useful tool for a
process-based performance measurement of a tertiary
care hospital. In addition, the study identified many
advantages of the AHP in the performance measure-
ment of a hospital: It is both objective and subjective in
nature regarding the measurement of the performance,
it allows for a group decision-making process and of-
fers a sound mathematical basis, its application is user
friendly, it identifies deficiencies in specific areas of
the hospital, and it allows carrying out a sensitivity
analysis that assists managers in understanding the
effects of their decisions and prioritizing areas in need
for improvement.
Other Industries: General Motors and Xerox
Corporation
AHP has also been successfully implemented at some
of the largest and well-known global corporations. For
example, Xerox uses AHP for R&D (research and de-
velopment) decisions, technology implementation, and
design selection (Forman & Gass, 2001). In addition, the
process is used to make marketing decisions, customer
requirement structuring decisions, and market segment
prioritization decisions. As a result of its successful
implementation, Xerox successfully eliminated the
selection of decisions that would normally be overturned
under their old intuitive selection process.
In the automobile industry, General Motors (GM)
has been very successful in using AHP. Advanced A
engineering staff use the AHP to assist them with
automobile design evaluation and selection. The best
and most cost-effective automobiles are selected with
the model’s assistance. The model’s selection process
is also credited for huge cost savings realized by GM.
Moreover, the methodology enables GM to be efficient
in its risk management functions.
Farm-Level Decisions
AHP has been successfully implemented in farm-level
decisions with an objective to aid farmers in selecting
one of the following farm systems: conventional, or-
ganic, or biological. Because of AHP’s ability to deal
with multiple criteria in agricultural economics, the
methodology proved to be a valuable tool in helping
farmers make their selections. These system selections
are based on the following objectives, also known as
subgoals in the AHP process: profit maximization,
improvement of human health, and environmental
protection through improved quality of land, water,
and air (Mawapanga & Debertin, 1996).
fUTURe TReNDS
Dr. Saaty has created a foundation called Creative
Decisions, which sponsors education, research, and
software development in advanced methods of deci-
sion making involving the AHP. A software application
called Super Decisions has been developed through
this organization. This software implements a more
advanced version of the AHP known as the analytic
network process (ANP), which generalizes the pair-
wise comparisons into clusters that can influence one
another’s priorities (Saaty, 2005). The main concept of
the ANP is that influence does not only flow downward
as in the AHP, but can also flow upward and across fac-
tors causing nonlinear results of priorities of alternative
choices (Super Decisions, 2005).
Another emerging development is the use of fuzzy
logic with the AHP. In its traditional formulation, the
AHP measurements are offered in exact numbers. In
many practical situations, however, decision makers are
often reluctant to assign an exact number on the basis
of incomplete or uncertain information. Also, they are

 The Analytic Hierarchy Process: Structuring, Measurement, and Synthesis
unable to provide an exact pair-wise comparison. To
overcome this problem, scholars have integrated the
AHP model with the fuzzy logic theory. The inclusion
of fuzzy logic has helped with the accurate assessment
of alternatives under incomplete or imprecise informa-
tion sets (Wu, Chang, & Lin, 2006).
The future of the AHP looks exciting. The method
continues to be a popular choice among governments,
scholars, educators, and businesspeople all over the
world. Further improvement in AHP is in progress
and these expected improvements would only add to
AHP’s future popularity.
CONCLUSION
AHP is a dream come true for Saaty, whose main idea
was to create a single unified model for all real-life
decision-making problems. Organizations can easily
implement AHP for their day-to-day decision-making
processes. The use of AHP has been profitable. It has
helped firms save millions by improving their overall
operations and resource allocation techniques. With the
development of Saaty’s new foundation called Creative
Decisions, we can expect a revision of the AHP model
and the introduction of a more advanced model, the
ANP. In the future, we can expect to see more and more
industries starting to use the AHP model.
RefeReNCeS
Forman, E. H., & Gass, S. I. (2001). The analytic hi-
erarchy process: An exposition. Operations Research,
49(4), 469-486.
Gass, S. I. (1985). Decision making, models and
algorithms: A first course. New York: John Wiley &
Sons, Inc.
Hariharan, S., Dey, P. K., Moseley, H. S. L., Kumar,
A. Y., & Gora, J. (2004). A new tool
of measurement of process based performance of mul-
tispecialty tertiary care hospitals. International Journal
of Health Care Quality Assurance, 17(6), 302-312.
Ishizaka, A. (2004, September 22-24). Advantages of
clusters and pivots in AHP. Paper presented at the 15th
Mini-Euro Conference MUDSM (Managing Uncer-
tainty in Decision Support Models).

Liedtka, S. L. (2005). Analytic hierarchy process and
multi-criteria performance management systems. Cost
Management, 1(6), 30-38.
Mawapanga, M. N., & Debertin, D. L. (1996). Choosing
between alternative farming systems: An application of
the analytic hierarchy process. Review of Agricultural
Economics, 18(3), 385-401.
Palmer, B. (1999). Click here for decisions. Fortune,
13(9), 153-156.
Roper-Lowe, G. C., & Sharp, J. A. (1990). The analytic
hierarchy process and its application to an information
technology decision. The Journal of the Operational
Research Society, 41(1), 49-59.
Saaty, T. L. (1994). How to make a decision: The analytic
hierarchy process. Interfaces, 24(6), 19-43.
Saaty, T. L. (1996). Decision making with dependence
and feedback: The analytic network process. Pittsburgh,
PA: RWS Publications.
Saaty, T. L. (2005). Theory and applications of the
analytic network process. Pittsburgh, PA: RWS Pub-
lications.
Saaty, T. L., & Alexander, J. M. (1981). Thinking with
models: Mathematical models in the physical, biologi-
cal and social sciences. New York: Pergamon Press.
Saaty, T. L., & Vargas, L. G. (1991). Prediction, pro-
jection and forecasting. Boston: Kluwer Academic
Publishers.
Super Decisions. (2005). The analytic network process
for decision-making. Retrieved July 29, 2007, from
http://www.superdecisions.com/anp_intro.php3
Tavana, M. (2003). CROSS: A multicriteria group-
decision-making model for evaluating and prioritizing
advanced-technology projects at NASA. Interfaces,
Linthicum, 33(3), 40-56.
Warren, L. (2004). Uncertainties in the analytic hier-
archy process. Australian Defense Science Magazine.
Retrieved July 29, 2007, from http://www.dsto.defence.
gov.au/publications/3476/DSTO-TN-0597.pdf
Wu, C., Chang, C., & Lin, H. (2006). Evaluating the
organizational performance of Taiwanese hospitals
using the fuzzy analytic hierarchy process. Journal of
American Academy of Business, 9(2), 201-210.
The Analytic Hierarchy Process: Structuring, Measurement, and Synthesis

Key TeRMS vector that when multiplied by the scalar is equal to the
vector obtained by letting the transformation operate A
Alternatives: They are multiple choices from which on the vector.
you have to choose one based upon their weights on
Eigenvector: An Eigenvector is a nonzero vector
the different criteria. The alternative with the highest
that is mapped by a given linear transformation of a
overall rating is selected as the most efficient choice
vector space onto a vector that is the product of a scalar
in an AHP.
multiplied by the original vector.
Consistency Measure (CM): Also known as con-
Hierarchy: It is a system of ranking and organizing
sistency ratio or consistency index, it is an estimated
in which each component is a subordinate to another
arithmetical indicator of the inconsistencies or intran-
component directly above or below depending on the
sitivity in a set of pair-wise ratings.
layout.
Criteria: These are one of the three main parts
Rank Reversal: This is one of the secondary prob-
of the AHP that need to be defined before solving a
lems related to AHP that occurs when the rankings for
problem. A criterion is a standard on which your judg-
the alternatives are changed with either the addition of
ments are based.
or removal of an alternative.
Eigenvalue: An eigenvalue is a scalar associated
Synthesis: Synthesis is the combining of separate
with a given linear transformation of a vector space
elements to form a coherent whole.
and having the property that there is some nonzero