Hierarchy and Network Processes_免费下载

admin管理员组

文章数量:1662041

2024年7月7日发(作者：)

DECISION MAKING – THE ANALYTIC HIERARCHY AND

NETWORK PROCESSES (AHP/ANP)

Thomas L. SAATY

University of Pittsburgh

saaty@

Abstract

This is the first part of an introduction to multicriteria decision making using the Analytic

Hierarchy Process (AHP) and its generalization, the Analytic Network Process (ANP). The discussion

involves individual and group decisions both with the independence of the criteria from the

alternatives as in the AHP and also with dependence and feedback in the entire decision structure as in

the ANP. This part explains the Analytic Hierarchy Process, with examples, and presents in some

detail the mathematical foundations. An exposition of the Analytic Network Process and its

applications will appear in later issues of this journal.

Keywords: Decision making, Analytic Hierarchy Process (AHP), Analytic Network Process (ANP)

1. Introduction

(Saaty 1977, 1994,

2000a, 2000b and 2001)

Decision making involves criteria and

alternatives to choose from. The criteria

usually have different importance and the

alternatives in turn differ in our preference for

them on each criterion. To make such tradeoffs

and choices we need a way to measure.

Measuring needs a good understanding of

methods of measurement and different scales

of measurement.

Many people think that measurement needs

a physical scale with a zero and a unit to apply

to objects or phenomena. That is not true.

Surprisingly enough, we can also derive

accurate and reliable relative scales that do not

have a zero or a unit by using our

understanding and judgments that are, after all,

the most fundamental determinants of why we

want to measure something. In reality we do

that all the time and we do it subconsciously

without thinking about it. Physical scales help

our understanding and use of the things that we

know how to measure. After we obtain

readings from a physical scale, they still need

to be interpreted according to what they mean

and how adequate or inadequate they are to

satisfy some need we have. But the number of

things we don’t know how to measure is

infinitely larger than the things we know how

to measure, and it is highly unlikely that we

will ever find ways to measure everything on a

physical scale with a unit. Scales of

measurement are inventions of a technological

mind. Our minds and ways of understanding

we have had with us and will always have. The

brain is an electrical device of neurons whose

ISSN 1004-3756/04/1301/1

J

OURNAL OF

S

YSTEMS

S

CIENCE AND

S

YSTEMS

E

NGINEERING

CN11-2983/N ©JSSSE 2004

Vol. 13, No. 1, pp1-35, March, 2004

Decision Making – The Analytic Hierarchy and Network Processes (AHP/ANP)

firings and synthesis must perform

measurement with great accuracy to give us all

the meaning and understanding that we have to

enable us to survive and reach out to control a

complex world. Can we rely on our minds to

be accurate guides with their judgments? The

answer depends on how well we know the

phenomena to which we apply measurement

and how good our judgments are to represent

our understanding. In our own personal affairs

we are the best judges of what may be good for

us. In situations involving many people, we

need the judgments from all the participants. In

general we think that there are people who are

more expert than others in some areas and their

judgments should have precedence over the

judgments of those who know less as in fact is

often the case in practice.

Judgments expressed in the form of

comparisons are fundamental in our biological

makeup. They are intrinsic in the operations of

our brains and that of animals and one might

even say of plants since, for example, they

control how much sunlight to admit. We all

make decisions every moment, consciously or

unconsciously, today and tomorrow, now and

forever, it seems. Decision-making is a

fundamental process that is integral in

everything we do. How do we do it? The

Harvard psychologist Arthur Blumenthal tells

us in his book The Process of Cognition,

Prentice-Hall, Inc., Englewood Cliffs, New

Jersey, 1977, that there are two types of

judgment: “Comparative judgment which is

the identification of some relation between two

stimuli both present to the observer, and

absolute judgment which involves the relation

between a single stimulus and some

information held in short term memory about

some former comparison stimuli or about some

previously experienced measurement scale

using which the observer rates the single

stimulus.”

When we think about it, both these

processes involve making comparisons.

Comparisons imply that all things we know are

understood in relative terms to other things. It

does not seem possible to know an absolute in

itself independently of something else that

influences it or that it influences. The question

then is how do we make comparisons in a

scientific way and derive from these

comparisons scales of relative measurement?

When we have many scales with respect to a

diversity of criteria and subcriteria, how do we

synthesize these scales to obtain an overall

relative scale? Can we validate this process so

that we can trust its reliability? What can we

say about other ways people have proposed to

deal with judgment and measurement, how do

they relate to this fundamental idea of

comparisons, and can they be relied on for

validity? These are all questions we need to

consider in making a decision. It is useful to

remember that there are many people in the

world who only know their feelings and may

know nothing about numbers and never heard

of them but can still make good decisions, how

do they do it? It is unlikely that by guessing at

numbers and assigning them directly to the

alternatives to indicate order under a criterion

will yield meaningful priorities because the

numbers are arbitrary. Even if they are taken

from a scale for a particular criterion, how

would we combine them across the criteria as

they would likely be from different scales? Our

answer to this conundrum is to derive a relative

2

J

OURNAL OF

S

YSTEMS

S

CIENCE AND

S

YSTEMS

E

NGINEERING

/ Vol. 13, No. 1, March, 2004

本文标签： 免费下载