Як прийняти та обгрунтувати рішення: аналітично-ієрархічний процес (АІП). Частина I. Приклади та застосування
We describe and discuss a logical approach, the Analytic Hierarchy Process (AHP), that can be used to make decisions. It involves breaking the problem down into finer and finer parts so that one is called upon to give a judgment comparing only a pair of issues in each judgment.
Збережено в:
| Дата: | 2019 |
|---|---|
| Автор: | |
| Формат: | Стаття |
| Мова: | Англійська |
| Опубліковано: |
The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute"
2019
|
| Онлайн доступ: | https://journal.iasa.kpi.ua/article/view/176738 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
| Назва журналу: | System research and information technologies |
| Завантажити файл: |  |
Репозитарії
System research and information technologies| _version_ | 1866302602135732224 |
|---|---|
| author | Saaty, Thomas L. |
| author_facet | Saaty, Thomas L. |
| author_sort | Saaty, Thomas L. |
| baseUrl_str | http://journal.iasa.kpi.ua/oai |
| collection | OJS |
| datestamp_date | 2019-08-23T03:22:41Z |
| description | We describe and discuss a logical approach, the Analytic Hierarchy Process (AHP), that can be used to make decisions. It involves breaking the problem down into finer and finer parts so that one is called upon to give a judgment comparing only a pair of issues in each judgment. |
| first_indexed | 2025-07-17T10:26:22Z |
| format | Article |
| fulltext |
© Tomas L. Saaty
Системні дослідження та інформаційні технології, 2002, 1 95
TIДC
ПРОБЛЕМИ ПРИЙНЯТТЯ РІШЕНЬ І
УПРАВЛІННЯ В ЕКОНОМІЧНИХ, ТЕХНІЧНИХ,
ЕКОЛОГІЧНИХ І СОЦІАЛЬНИХ СИСТЕМАХ
UDC 519.5
HOW TO MAKE AND JUSTIFY A DECISION: THE ANALYTIC
HIERARCHY PROCESS (AHP)
PART 1. EXAMPLES AND APPLICATIONS
THOMAS L. SAATY
We describe and discuss a logical approach, the Analytic Hierarchy Process (AHP),
that can be used to make decisions. It involves breaking the problem down into finer
and finer parts so that one is called upon to give a judgment comparing only a pair
of issues in each judgment.
1. INTRODUCTION
We describe and discuss a logical approach, the Analytic Hierarchy Process
(AHP), that can be used to make decisions. It involves breaking the problem down
into finer and finer parts so that one is called upon to give a judgment comparing
only a pair of issues in each judgment. This avoids mixing too many aspects of
the problem and not knowing what goes with what to obtain the final answer.
However, it does call for one to structure the problem hierarchically with broad
understanding of the people and their interests and of the issues involved. Once
one has the structure it becomes easier to convey to others the influences driving
that decision.
The AHP has had appeal to managers and decision-makers at all levels of
decision-making. It enables one to include both the strength of feelings needed to
express judgment and the logic and understanding relating to the issues involved
in the decision. It combines the multiplicity of judgments in a systematic way to
obtain the best outcome, or mix of actions to be taken. Finally, and more
significantly, these outcomes are derived in an agreeable way that are in harmony
with our intuition and understanding and not forced on us by technical
manipulations. There are easy to use software packages that implement this
approach and make decision-making a pleasure. Expert Choice© is the name of
one such software. For a free trial version go to www.expertchoice.com. We
should be able to say that, given the information, we agree on the method of
making the decision (if not the outcome of any particular decision). The matter
becomes a common concern, not a mystical phenomenon.
The process contributes to solving complex problems by structuring a hierarchy
of criteria, stakeholders, and outcomes and by eliciting judgments to develop
priorities. It also leads to prediction of likely outcomes according to these judgments.
The outcome can be used to rank alternatives, allocate resources, conduct
Thomas L. Saaty
ISSN 1681–6048 System Reasearch & Information Technologies, 2002, 1 96
benefit/cost comparisons, exercise control in the system. by evaluating the sensitivity
of the outcome to changes in judgment, and carry out planning of projected and
desired futures. A useful by-product is the measurement of how well the leader
understands the relations among factors. Although people generally are not
consistent, the main concern here is the degree of their inconsistency. Is their
understanding close to capturing the interactions observed? Or is it a random
understanding that only hits the target now and then?
Good decisions must survive the difficulties and hazards of people and
environment. We need to make decisions that are both desirable and survivable
rather than simply ones that we best like without regard to how lasting they may be.
Predicting outcomes plays an important role in making such choices. To do this well,
we can decompose a decision into separate structures involving scenarios of benefits,
costs, opportunities and risks and then carefully combine their separate outcomes
for the best decision.
Decision–making groups need to formalize their agenda and structure the
interaction. A process that can unfold the complexity of the issues or problems is
much needed and would be particularly useful in group decision making.
Recognizing that perceptions and stakes can vary among group members, such a
process should also specify how individuals can bargain on specific differences.
Finally, it would be desirable to have a measure of the consistency of judgments
which individuals give and which the group settles on. The Analytic Hierarchy
Process described in this paper is such a process.
2. EXAMPLE 1 — CHOOSING A HUSBAND
1. Oksana is an attractive twenty seven year old MBA graduate with three
years business experience from a middle class family wants to choose a husband
from one of three suitors. Her criteria are age, looks, intelligence and economic
status. Her suitors are.
2. Taras, a rugged looking 30 year old engineer with a well paying job,
interested in his career and in rainsing a family. He is hard nosed, no nonsense kind
of person, gentle and loving.
3. Ivan, a 37 year old promising artist who is very romantic but whose career is
still developing. His income from day to day is uncertain, but he is a sensitive
imaginative genius with a lot of promise, but who is more interested in beauty and
spirit than he is in accumulating wealth.
4. Grigoriy, a 25 year old handsome, virile and fast moving young man with a
brilliant future in a famous and successful family business. He is extremely generous
and thoughtful, but he is also temperamental and assertive in his ways.
The matrix of pairwise comparison judgments of the criteria given by the
young lady in this case is shown in Table 1. The judgments are entered using the
fundamental scale of the AHP: a criterion compared with itself is always assigned
the value 1 so the main diagonal entries of the pairwise comparison matrix are all
1. The numbers 3, 5, 7, and 9 correspond to the verbal judgments «moderately
more dominant», «strongly more dominant», «very strongly more dominant», and
«extremely more dominant» (with 2, 4, 6, and 8 for compromise between the
previous values). Reciprocal values are automatically entered in the transpose
position, so she must make a total of (4 × 3)/2 = 6 pairwise judgments. There
How to make and justify a decision: the analytic hierarchy process. Part 1.
Системні дослідження та інформаційні технології, 2002, 1 97
are 16 positions in all with four ones on the diagonal, half of the remaining 12
are the reciprocals of the others, so six judgments need to be entered. The
priorities are obtained by raising the matrix to a large power to capture all the
interactions, adding the entries in each row and dividing by the total sum of the
rows. We are permitted to use decimal values between the integers, such as
2.6, if desired. It is mathematically demonstrated that it is necessary to use this
scale to get meaningful results in practice. It represents the normal range of
human sensitivity to phenomena that are homogeneous. When things are widely
scattered, they can be grouped into separate clusters with a common element in
adjacent clusters and the scale 1 to 9 is applied to compare the elements in each
cluster with the common element serving as a link. When there are actual
measurements for pairs being compared, such as money, we can use the ratio of
their measurements.
It is generally preferable to use the verbal judgment and then enter the
corresponding numerical value, although the software Expert Choice allows one
to do either and in addition to use geometric representations of the relative
strength of the paired comparison. In Table 1 below, when comparing Age on the
left with Looks on Top, she thought that Looks are very strongly more important,
and the value 1/7 is entered in the (Age, Looks) position, and automatically a 7 is
entered in the (Looks, Age) position. Similarly, in comparing Intelligence with
Looks, it is thought that Intelligence is slightly more important than Looks, and a
2 is entered in the (Intelligence,Looks) position, and a 1/2 is entered in the
(Looks, Intelligence) position and so on. We always compare the criterion on the
left as to how much more dominant it is than the criterion at the top. If it is not,
the reciprocal value is used.
Intelligence
(0.429)
Riches
(0.182)
Looks
(0.327)
Age
(0.062)
Taras
(0.194)
Ivan
(0.573)
Grigoriy
(0.232)
Choosing the Best
Husband for Oksana
GOAL
Fig. 1. The Best Husband Hierarchy
Thomas L. Saaty
ISSN 1681–6048 System Reasearch & Information Technologies, 2002, 1 98
T a b l e 1 . Comparing the Criteria for Importance with Respect to the Goal
Age Looks Intelligence Riches Priorities
Age 1 1/7 1/5 1/3 0.062
Looks 7 1 1/2 2 0.327
Intelligence 5 2 1 2 0.429
Riches 3 1/2 1/2 1 0.182
Inconsistency Value 4%
T a b l e 2 . Comparing the Suitors with Respect to the Criteria
Age Taras Ivan Gritsko Priorities
Taras 1 3 1/3 0.258
Ivan 1/3 1 5 0.105
Gritsko 3 1/5 1 0.637
Inconsistency Value 4%
Looks Taras Ivan Grigoriy Priorities
Taras 1 1/5 2 0.166
Ivan 5 1 7 0.740
Gritsko 1/2 1/7 1 0.094
Inconsistency Value 1%
Intelligence Taras Ivan Grigoriy Priorities
Taras 1 1/5 3 0.188
Ivan 5 1 7 0.731
Grigoriy 1/3 1/7 1 0.081
Inconsistency Value 6%
Riches Taras Ivan Grigoriy Priorities
Taras 1 5 1/4 0.237
Ivan 1/5 1 1/8 0.064
Grigoriy 4 8 1 0.699
Inconsistency Value 9%
In Table 3 we multiply the weights of the suitors by the weights of the criteria and
add to obtain the final ranking.
T a b l e 3 . Synthesis of the Priorities
Criteria Age Looks Intelligence Riches Priorities
Suitors (0.062) (0.327) (0.429) (0.182) Synthesis
Taras 0.258 0.166 0.188 0.237 0.195
Ivan 0.105 0.740 0.731 0.064 0.573
Gritsko 0.0637 0.094 0.081 0.699 0.232
Ivan wins because he has better looks and is more intelligent. Both criteria have
high priorities of 0.327 and 0.429 or a total of 0.756 of the assessment. Thus, she
should marry Ivan the artist.
How to make and justify a decision: the analytic hierarchy process. Part 1.
Системні дослідження та інформаційні технології, 2002, 1 99
3. EXAMPLE 2 ― SELECTING A PIPELINE ROUTE IN SOUTH AMERICA
Executives of a South American oil company must choose among three locations for
construction of an oil pipeline. The alternatives are a northern route to Covenas, a
southern route to Orito, and a western route to Bahia. The southern and western
routes terminate on the west side of the Panama canal and the northern route
terminates on the east side; selecting the western route would result in a $1 per barrel
increase in profit due to transportation savings. While this is a factor affecting
NPV(Net Present Value), management has to balance its profit motives with
concerns for the environment, managerial control, and the riskiness of the venture.
So, in the end the cost savings due to transportation are not sufficient to make the
western route the best.
During the decision session discussion among vice presidents, engineers, and
operating managers of the company was instructive as each group member learned
from the others’ information and insights.
The problem of choosing the best route for the pipeline is shown in Figure 2.
Notice that some criteria have subcriteria – for example, risk is broken down into
vulnerability, reliability, and design considerations. The alternatives are evaluated
directly for the criterion Net Present Value, but for the subcriteria under Risk. ECP
refers to equal capital partners. The reader will have no difficulty in understanding
the factors represented here.
The next step is to make comparative judgments. The decision-making
group from the oil company assessed the relative importance of all possible
pairs of criteria with respect to the overall goal Select Best Pipeline Route,
coming to a consensus judgment on each pair. Their judgments are arranged
into a matrix. The process has a scientific way of combining individual
judgments that we will not go into here. The question to ask when comparing
two criteria is: which is more important and how much more important is it with
Select Best Pipeline Route
GOUL
Net Present
Value (0.588)
Risk
(0.200)
Management
Control (0.084)
Partners
(0.079)
Environment
(0.049)
Vulnerability
(0.498)
Reliability
(0.285)
Design Complex
(0..217)
Flexibility
(0.500)
Leverage
(0.235)
Agreement
(0.265)
ECP
(0.105)
Equity Ptmrs
(0.309)
3rd Parties
(0.582)
Social
(0.049)
Physical
(0.640)
NORTH
(0.389)
SOUTH
(0.277)
WEST
(0.334)
Fig. 2. The Pioeline Decision Model
Thomas L. Saaty
ISSN 1681–6048 System Reasearch & Information Technologies, 2002, 1 100
respect to selecting the best pipeline route? First the criteria are compared for
their importance with respect to the goal as shown in below:
T a b l e 4 . Comparing the Criteria for Importance with Respect to the Goal
Net Present
Value
(NPV)
Risk Management
Control
Partners Environ-
ment
Prio-
rities
NPV 1 6 7 7 6 0.588
Risk 1/6 1 3 5 3 0.200
Mgt. Control 1/7 1/3 1 1 3 0.084
Partners 1/7 1/5 1 1 3 0.079
Environment 1/6 1/3 1/3 1/3 1 0.049
Inconsistency Value 11%
Then the subcriteria are compared with respect to their importance to the criterion
they fall beneath. For an example of comparing the subcriteria beneath the
criterion Partners, see Table 5. Finally the alternatives are compared with respect
to each of the criteria or subcriteria above them as in Table 6.
T a b l e 5 . Judgments under Criterion Partners.
Partners ECP Equity 3rd Party Priorities
ECP 1 1/3 1/2 0.163
Equity 3 1 2 0.540
3rd Parties 2 1/2 1 0.297
Inconsistency Value 0%
T a b l e 6 . Judgments for Alternatives under Subcriterion Physical.
Physical North South West Priorities
North 1 1/3 1/2 0.163
South 3 1 2 0.540
West 2 1/2 1 0.297
Inconsistency Value 1%
The priorities for each set of judgments, as shown in Table 7 are combined
by multiplying the values for North, South and West under each criterion or
subcriterion directly linked to the routes, by the weighted value shown for the
criterion or subcriterion and adding. The priorities of the criteria and subcriteria
(called covering criteria of the alternatives) sum to 1.000. The best overall choice
is seen to be the Northern route. Fig. 3 is a bar representation of the outcome.
NORTH 0.389
SOUTH 0.277
WEST 0.334
Fig. 3. The final results
How to make and justify a decision: the analytic hierarchy process. Part 1.
Системні дослідження та інформаційні технології, 2002, 1 101
T a b l e 7 . Synthesis for Final Answer
NPV
0.588
Vulnr
0.099
Reliab
0.057
Design
0.043
Flexibl
0.042
Lever
0.020
Agree
0.022
North 0.312 0.605 0.630 0.659 0.726 0.182 0.182 …
South 0.304 0.210 0.151 0.540 0.172 0.273 0.273
West 0.384 0.184 0.218 0.297 0.102 0.545 0.545
ECP
0.009
Equity
0..024
3rd Pty
0.046
Social
0.018
Phys
0.031
Overall
Priority
… 0.122 0.094 0.669 0.190 0.163 0.389
0.230 0.167 0.243 0.547 0.540 0.277
0.648 0.740 0.088 0.263 0.297 0.334
The sensitivity graph shown in 4 allows analysis of the result. The graph
shows how each route performs on each criterion. The importance of the criteria
themselves is represented by the vertical bars. For example, Risk is about 20% of
the concern, while NPV is almost 60%. The line for the North alternative
intersects the Risk axis the highest up, so it is best on Risk. But it is worst on the
environment, and so on, but overall, the Northern route just edges out the Western
route and is the best.
Fig. 4. Sensitivity Graph showing Performance of Alternatives on Criteria
4. ABSOLUTE MEASUREMENT
Another method of ranking alternatives to evaluate them one at a time rather
than compare them in pairs. This is particularly useful when their number is large
and there is expert knowledge about standards they must satisfy. This process
involves introducing and prioritizing intensities for each of the criteria or
subcriteria above the alternatives. For example, the criterion Net Present Value
Thomas L. Saaty
ISSN 1681–6048 System Reasearch & Information Technologies, 2002, 1 102
may take on the intensities, Very High, High, Medium, Low and Very Low. One
pairwise compares these intensities as to how preferred they are to each other just
as we did in the comparison matrices above, and thus obtain priorities for them.
After doing this, one assigns one intensity from each criterion for each alternative,
weights it by the importance of its criterion and adds to obtain the overall priority
for that alternative. This process also lends itself well to resource allocation as
shown in the following example of allocating raise money to employees.
Evaluating Employees for Raises. Employees are evaluated for raises.
The criteria are Dependability, Education, Experience, and Quality. Each criterion
is subdivided into intensities, standards, or discrimination categories as shown in
Figure 5. Priorities are set for the criteria by comparing them in pairs. The
intensities are then pairwise compared according to importance with respect to
their parent criterion (example as in Table 8). Their priorities are often divided by
the largest intensity for each criterion (second column of priorities in Fig. 5)
particularly useful in preserving the ranks of the alternatives from the addition or
deletion of other alternatives. Finally, each individual is rated in Table 9 by
assigning the intensity rating that applies to him or her under each criterion and
adding. The score of each assigned intensity is weighted by the priority of its
criterion and summed over the criteria to derive a total ratio scale score for each
individual. This approach can be used whenever it is possible to set priorities for
intensities of the criteria, which is usually possible when sufficient experience
with a given operation has been accumulated.
Fig. 5. Employee evaluation hierarchy
GOAL
Dependability
0.4347
Educatiion
0.2774
Experience
0.1755
Quality
0.1123
Outstanding
(0.182) 1.000
Above Average
(0.114) 0.627
Average
(0.070) 0.382
Below Average
(0.042) 0.232
Unsatisfactory
(0.027) 0.149
College
(0.203) 1.000
High School
(0.052) 0.258
Exceptional
(0.098) 1.000
None
(0.022) 0.111
A Lot
(0.050) 0.511
Average
(0.020) 0.204
None
(0.008) 0.081
Good
(0.100) 1.000
Poor
(0.012) 0.125
How to make and justify a decision: the analytic hierarchy process. Part 1.
Системні дослідження та інформаційні технології, 2002, 1 103
T a b l e 8 . Ranking Intensities
Outstan-
ding
Above
Average Average Below
Average
Unsatis-
factory Priorities
Outstanding 1.0 2.0 3.0 4.0 5.0 0.419
Above
Average 1/2 1.0 2.0 3.0 4.0 0.263
Average 1/3 1/2 1.0 2.0 3.0 0.160
Below
Average 1/4 1/3 1/2 1.0 2.0 0.097
Unsatisfactory 1/5 1/4 1/3 1.2 1.0 0.062
Inconsistency Ratio = 0.015
T a b l e 9 . Ranking Alternatives
Dependa-
bility Education Experience Quality Total Norma-
lized
1. Adams, V Outstanding College Exceptional Good 1.000 0.245
2. Becker, L Average College Average Good 0.592 0.145
3. Hayat, F Average College A lot Good 0.645 0.158
4. Kesselman,S Above
Average
High
School None Poor 0.373 0.091
5. O’Shea, K Average College Average Poor 0.493 0.121
6. Peters, T Average College None Good 0.570 0.140
7. Tobias, K Above
Average None A Lot Poor 0.470 0.100
The raises can be made in proportion to the normalized values on the right.
5. OTHER APPLICATIONS
Organ Transplantation. The City of Pittsburgh was for a while a leader
in the world in organ transplantations. Because there are more patients who need
livers, hearts and kidneys than there are available organs, it had become essential
to assign priorities to the patients. The goal is divided into: emotionally dependent
and financially dependent patients: Both are divided into single, married, and
divorced with and without dependent and financially dependent patients. Then
each of them is further divided into: medical history (time on donor list, degree of
disability), physical history (degree of ability to endure rehabilitation, willingness
to cooperate, etc.), and social status (criminal record, volunteer work). A patient is
ranked according to the intensities under each criterion. The higher the total score
the better the opportunity to receive a transplant. The covering criteria and
subcriteria of the alternatives are each assigned intensities for which priorities are
developed as in the salary raise example. In general, one would use the intensities
to score a patient. When there is no intensity, either the full value of the criterion
is assigned, or a zero value otherwise. For example, a person with no criminal
record was assigned the full priority of the Criminal Record criterion which was
in this case 0.033. A patient with a criminal record would receive a zero for that
criterion.
Thomas L. Saaty
ISSN 1681–6048 System Reasearch & Information Technologies, 2002, 1 104
Here are details of other examples of complex decisions that have been made
using the AHP.
Vehicle Fleet Management. Vehicle fleet management policies and
programs are tools by which the fleet manager meets the business/transportation
requirements of organization personnel. These policies may be formulated by
committees, management, or by the fleet manager. The model below provides
fleet managers with a framework to select and establish the best policies
according to their organizational objectives. Important objectives of fleet
managers, who operate on a fixed budget, include minimizing cost, improving
relationships with internal constituencies (purchasing, sales, marketing,
accounting, vehicle maintenance, and finance departments), improving
relationships with external constituencies (leasing companies, auto manufacturers,
auto dealers, contract mechanics, parts vendors, and insurance companies),
improving quality of the fleet, and improving safety, maintenance, and scheduling
aspects of fleet management. This particular decision examines two alternative
policies: fleet downsizing (smaller vehicle size), and decreasing the number of
vehicles in the fleet. Not implementing a new policy, nor maintaining the status
quo, could also be considered as an alternative.
Product Marketing Strategy. The traditional focus of product
management has been on the introduction of new products. However, in recent
years, there has been increasing concern on the part of organizations about
product lines that are fast approaching obsolescence or are slow–moving. While
companies have formalized evaluation programs to monitor and assess the
performance of their product offerings, the decision-making process for product
modification and/or elimination strategies remains unstructured. This decision
determines a future marketing strategy for a product identified as a problem or for
late–to–mature products. The decision to delete the product, modify the product
marketing strategy, or maintain the status quo is examined in light of the desired
corporate image, market share, competition, synergies of manufacturing,
marketing, distribution, end uses, investment requirements for R&D and
equipment, human resource requirements, manufacturing resource requirements,
financial considerations, preferences of customers, suppliers and distributors, and
the corporate legal, political, and economic environment. Slight modifications to
the decision add the capability to assess whether or not proposed additions should
be made to the product line.
Bank Acquisition. The New National Bank (NNB) has conducted an
analysis to uncover and identify its basic motive for expansion. One motive for
expansion is sales enhancement through product extension; a second motive for
expansion is cost reduction through economies of scale or improved operating
efficiency which eliminates excess capacity and overhead expenses. NNB realizes
that acquisitions can be a powerful tool in adapting to deregulation, excess
capacity, and thinning margins. However, achieving value–added growth through
acquisition can be difficult for bank acquisition because prices frequently exceed
twice the book value. After determining that expansion was desirable, and
acquisition was preferable to internal development, NNB narrowed down its
choice to two possible candidates which would best fit its acquisition strategy:
Bank A and Bank B.
BANK A is headquartered in Pittsburgh, Pennsylvania, and is the nation's
thirteenth largest bank holding company with assets of $3 billion. The building it
How to make and justify a decision: the analytic hierarchy process. Part 1.
Системні дослідження та інформаційні технології, 2002, 1 105
resides in and owns is considered to be a historical landmark. BANK B is
headquartered in Wilmington, Delaware and has assets of $1.5 billion. Its credit
card operation is one of the largest in the nation. In addition, Delaware state law
has no ceiling on credit card interest rates and has lower income tax rates. NNB
has developed a comprehensive framework for evaluation that combines strategic,
financial, operating, and integrating factors. The final determination of the ideal
candidate depends on a comparison of the benefit/cost advantages of each. The
costs used in this comparison are compatibility, personnel, interest rate ceiling on
credit cards, and taxes. The benefits include financial value, geographic
expansion, product expansion, and reputation.
Mergers and Acquisitions. When studying motivations for mergers and
acquisitions, analysts run up against a lack of reliable evidence with which to
conclusively support motivation theory; the majority of mergers do not actually
seem to be profitable for the acquiring firm's stockholders. To better determine
the future success of proposed mergers, judgments may be entered for a specific
merger prospect for an acquisition firm into the model below by two different
decision makers, The first decision maker holds the neoclassical view that
mergers occur because managers want to maximize shareholders wealth.
Managers are more interested in retaining their jobs in the merged firm than they
are in personal profit or post–merger stock price. The second decision maker
believes managers wish to maximize their own wealth and satisfy a psychological
need for growth and power. While managers are concerned about the merged
firm's eventual profit and stock price, this concern stems from a desire to
maximize their personal profits. It is interesting to note that although the decision-
makers assume different motives for the mangers, there were no significant
differences in their predictions of successful mergers. The criteria the decision-
makers use include target company's stockholders and managers, the Federal
Trade Commission, and the Justice Department. The alternatives are no attempt at
a merger, a failed merger, and a successful merger.
Choosing a Telecommunications System. The success or failure of
many businesses today depends a great deal on the ability of a business to
communicate with its customers, with other offices, and with suppliers. This
model is built to assist a small to medium size health care company, Megadata,
choose a telecommunications system. Megadata provides doctors, hospitals, and
other health care providers with easy and efficient transfers of medical insurance
information and diagnoses. Rapid transfer of this information allows doctors and
hospitals to minimize their operating expenses and make better decisions by
sharing information between experts across the country. Megadata depends on its
ability to transfer large amounts of data via the national telephone network. The
company is considering buying a new Private Branch Exchange (PBX) to
minimize long–term costs, increase flexibility to expand as the company expands,
minimize down–time due to maintenance, provide ease of use for end-users and
network administrators, and include several key options to improve operations.
Megadata's alternatives are: to ignore the increasing demand for telephonic
transfer of data and continue operating with the current system (it is estimated that
the system will operate for another two to three years without serious impact on
income), to buy a «top of the line» system that will provide the company with the
advantages of many new technological breakthroughs, or to purchase a
Thomas L. Saaty
ISSN 1681–6048 System Reasearch & Information Technologies, 2002, 1 106
refurbished older, but fairly reliable system, that can provide adequate service for
another six to ten years.
Government/Public Strategy: Should Dam Construction occur on
Alaska's Susitna River? No other single man–made endeavor has ever
changed the face of the earth to the same extent as have dams. In addition to the
alteration of the land's geography, far-reaching implications of dams encompass
political, social, ecological, economic, and psychological aspects. While dam
construction benefits man through flood control, provision of electrical power and
water resources, irrigation, and recreational facilities, the detriments are often
more opaque and not so easily quantified. Dam construction causes displacement
of citizens, and there has been greater incidence of disease accompanying dam
construction historically. While some marine species may be cultivated by
existing dam reservoirs, other species face potential extermination. Dams also
tend to affect other ecosystems in the surrounding area, and cause erosion of
surrounding terrain. Additionally, there exists a safety hazard from the dam as it
advances in age. This model determines whether to build a proposed dam on
Alaska's Susitna River. The Susitna, with a drainage area of more than
CAPut!’,000 square miles is the sixth largest river in Alaska and is an important,
productive, fishery resource which contributes to the support of significant
wildlife populations in the virtually untouched and relatively inaccessible lands
across which it flows. Various development schemes to harness the extensive
hydro-electric potential of the Susitna have been advanced by public agencies and
private organizations. This model was used to evaluate whether or not to build a
proposed dam given short and long term environmental effects, safety factors,
power generated, employment created, and other financial, economic, and
political considerations.
Here are some other recent decisions that have been made using the AHP:
• British Airways used it in the late 1990’s to decide on the vendor to use
for the entertainment system on its entire fleet of airplanes in the new century.
• An oil company used it in the late 1980’s to determine the best type of
platform to build for drilling for oil in the North Atlantic. A platform costs around
3 billion dollars to build, but the demolition cost was an even more significant
factor in the decision.
• The process was applied to the U.S. versus China conflict in the
intellectual property rights battle of 1995 over Chinese individuals copying music,
video and software tapes and CD’s. An AHP analysis involving three hierarchies
for benefits, costs and risks showed that it was much better for the U.S. to not
sanction China. Shortly after the study was complete, the U.S. awarded China
most-favored nation trading status and did not sanction it.
• Xerox Corporation has used the AHP to allocate close to a billion dollars
to its research projects.
• The Ford Motor Company, in 1999, used the process to establish
priorities for criteria that improve customer satisfaction. Ford gave Expert Choice
Inc, the software company which helped them with the study, an Award for
Excellence for helping them achieve greater success with their clients.
• In 1986 the Institute of Strategic Studies in Pretoria, a government-backed
organization, used the process to analyze the conflict in South Africa and
recommended actions ranging from the release of Nelson Mandela to the removal
How to make and justify a decision: the analytic hierarchy process. Part 1.
Системні дослідження та інформаційні технології, 2002, 1 107
of apartheid and the granting of full citizenship and equal rights to the black
majority. All of these recommended actions were implemented within a relatively
short time.
• The AHP has been used in student admissions, military personnel
promotions and hiring decisions.
• In sports it was used in 1995 to predict which football team would go to
the Superbowl and win (correct outcome, Dallas won over my hometown,
Pittsburgh), and the process was applied in baseball to analyze which Padres
players should be retained. Such applications take several days to make, to
capture sufficient knowledge and detail.
• IBM used the process in 1991 in designing its successful mid-range AS
400 computer. IBM won the prestigious Malcolm Baldrige award for excellence
for that effort. A book, the Silverlake Project (Bauer, 1992), was written about the
AS 400 project and has a chapter devoted to how AHP was used in
benchmarking.
Since the AHP helps one organize one’s thinking, it can be used tos deal
with many decisions that are often made intuitively. As a minimum the process
allows one to experiment with different criteria and different judgments. A trial
version of the AHP software can be obtained from www.expertchoice.com.
REFERENCES
1. Bauer R.A., Collar E., Tang V. The Silverlake Project. Transformation at IBM. —
Oxford: University Press, 1992.
2. Fortune Magazine, May 10, 1999, «Click Here for Decisions». P. 153–156.
3. Saaty T.L. Decision Making for Leaders, RWS Publications, 4922 Ellsworth Ave.
Pittsburgh, PA. U.S.A. 2000.
4. Saaty T.L. The Analytic Hierarchy Process, McGraw Hill International. —
New York, 1980. Translated to Russian, Portuguese, and Chinese. Revised
edition, paperback. — Pittsburgh: RWS Publications, 1990, 1996.
5. Saaty T.L., Vargas L.G. The Logic of Priorities; Applications in Business, Energy,
Health, and Transportation. — Boston: Kluwer-Nijhoff. 1982. Paperback edition.
— Pittsburgh: RWS Publications, 1991.
6. Saaty T.L., Kearns K.P. Analytical Planning. The Organization of Systems. —
Oxford: Pergamon Press, 1985. Translated to Russian. Paperback edition. —
Pittsburgh: RWS Publications, 1991.
7. Saaty T.L. Alexander J. Conflict Resolution: The Analytic Hierarchy Process. —
New York: Praeger, 1989.
8. Saaty T.L., Vargas L.G. Prediction, Projection and Forecasting. — Boston: Kluwer
Academic, 1991.
9. Saaty T.L., Forman E.H. The Hierarchon — A Dictionary of Hierarchies. RWS
Publications, 4922 Ellsworth Avenue, Pittsburgh, PA 15213–2807, 1993.
10. Saaty T.L. Fundamentals of Decision Making with the Analytic Hierarchy Process,
paperback, RWS Publications, 4922 Ellsworth Avenue, Pittsburgh, PA
15213-2807, original edition 1994, revised 2000.
11. Saaty T.L., Vargas L.G. Decision Making in Economic, Political Social and
Technological Environments: The Analytic Hierarchy Process. Paperback, RWS
Publications, 4922 Ellsworth Ave., Pittsburgh, PA 15213, 1994.
Thomas L. Saaty
ISSN 1681–6048 System Reasearch & Information Technologies, 2002, 1 108
12. Saaty T.L. The Analytic Network Process: Decision Making with Dependence and
Feedback, RWS Publications, 4922 Ellsworth Ave., Pittsburgh, PA 15213, 1996,
completely revised and published 2001.
13. Saaty T.L. Decision Making with Dependence and Feedback: The Analytic Network
Process (ANP) and ECNet Software, Guide, Manual and Examples, Expert
Choice, Inc., Pittsburgh, 1997.
14. Saaty T.L. The Brain, Unraveling the Mystery of How it Works: The Neural
Network Process, RWS Publications, 4922 Ellsworth Avenue, Pittsburgh, PA
15213, 2000
15. Saaty, T.L., Vargas L.G. Models, Methods, Concepts and Applications of the
Analytic Hierarchy Process, Kluwer Academic Publishers, Boston, 2000.
Received 10.09.2001
|
| id | journaliasakpiua-article-176738 |
| institution | System research and information technologies |
| keywords_txt_mv | keywords |
| language | English |
| last_indexed | 2025-07-17T10:26:22Z |
| publishDate | 2019 |
| publisher | The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" |
| record_format | ojs |
| resource_txt_mv | journaliasakpiua/97/6b498e4c6d862ba49b036fbf880ae997.pdf |
| spelling | journaliasakpiua-article-1767382019-08-23T03:22:41Z How to make and justify a decision: the analytic hierarchy process. Part 1. Examples and Applications Как принять и обосновать решение: аналитическо-иерархический процесс (АИП). Часть I. Примеры и приложения Як прийняти та обгрунтувати рішення: аналітично-ієрархічний процес (АІП). Частина I. Приклади та застосування Saaty, Thomas L. We describe and discuss a logical approach, the Analytic Hierarchy Process (AHP), that can be used to make decisions. It involves breaking the problem down into finer and finer parts so that one is called upon to give a judgment comparing only a pair of issues in each judgment. Описывается и обсуждается логический подход — аналитическо-иерархический процесс (АИП), который может применяться для принятия решений. Он состоит в делении проблемы на все более и более мелкие части до тех пор, пока каждое решение не сведется к сравнительной оценке двух альтернатив. Розглядається та обговорюється логічний підхід — аналітично-ієрархічний процес (АІП), який можна застосовувати до прийняття рішень. Він полягає в поділенні проблеми на все дрібніші і дрібніші частки, поки кожне рішення не буде зведено до порівняння двох альтернатив. The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" 2019-08-23 Article Article application/pdf https://journal.iasa.kpi.ua/article/view/176738 System research and information technologies; No. 1 (2002); 95-108 Системные исследования и информационные технологии; № 1 (2002); 95-108 Системні дослідження та інформаційні технології; № 1 (2002); 95-108 2308-8893 1681-6048 en https://journal.iasa.kpi.ua/article/view/176738/176562 Copyright (c) 2021 System research and information technologies |
| spellingShingle | Saaty, Thomas L. Як прийняти та обгрунтувати рішення: аналітично-ієрархічний процес (АІП). Частина I. Приклади та застосування |
| title | Як прийняти та обгрунтувати рішення: аналітично-ієрархічний процес (АІП). Частина I. Приклади та застосування |
| title_alt | How to make and justify a decision: the analytic hierarchy process. Part 1. Examples and Applications Как принять и обосновать решение: аналитическо-иерархический процесс (АИП). Часть I. Примеры и приложения |
| title_full | Як прийняти та обгрунтувати рішення: аналітично-ієрархічний процес (АІП). Частина I. Приклади та застосування |
| title_fullStr | Як прийняти та обгрунтувати рішення: аналітично-ієрархічний процес (АІП). Частина I. Приклади та застосування |
| title_full_unstemmed | Як прийняти та обгрунтувати рішення: аналітично-ієрархічний процес (АІП). Частина I. Приклади та застосування |
| title_short | Як прийняти та обгрунтувати рішення: аналітично-ієрархічний процес (АІП). Частина I. Приклади та застосування |
| title_sort | як прийняти та обгрунтувати рішення: аналітично-ієрархічний процес (аіп). частина i. приклади та застосування |
| url | https://journal.iasa.kpi.ua/article/view/176738 |
| work_keys_str_mv | AT saatythomasl howtomakeandjustifyadecisiontheanalytichierarchyprocesspart1examplesandapplications AT saatythomasl kakprinâtʹiobosnovatʹrešenieanalitičeskoierarhičeskijprocessaipčastʹiprimeryipriloženiâ AT saatythomasl âkprijnâtitaobgruntuvatiríšennâanalítičnoíêrarhíčnijprocesaípčastinaiprikladitazastosuvannâ |