You are on page 1of 32

The Analytic Hierarchy Process (AHP) for Decision Making

Decision Making involves setting priorities and the AHP is the methodology for doing that. Our own personal priorities are the most important influence on our decisions. The AHP is the way to derive priorities in a meaningful scientific way to make decisions.
1

We have the belief in mathematics and science that measurement demands that we always have an instrument with a scale marked on it that has a zero and an arbitrarily chosen unit to enable us to measure things one by one on that scale independently of other things. Our use of Cartesian axes makes us believe that everything in the world can be studied with functions defined in such a space with coordinates. That is not true. Our biology teaches us that we need to always compare things to decide which is bigger or better or more important more preferred or more likely to happen and so on, and that things are better understood, or in fact can only be understood relative to each other; that is they are dependent in some way of measurement on one another. To measure all things one by one and not compare them is simplistic and loses a very important property that cannot be captured with ordinary measurement. That is why the world of economics has some serious problems. I also believe that one day we will learn that our understanding of physics may lack this powerful and necessary way of looking at things. Let us look at the real world of decision making
2

Real Life Problems Exhibit:

Strong Pressures and Weakened Resources Complex Issues - Sometimes There are No Right Answers Vested Interests Conflicting Values

Decision Making
We need to prioritize both tangible and intangible criteria: In most decisions, intangibles such as political factors and social factors take precedence over tangibles such as economic factors and technical factors It is not the precision of measurement on a particular factor that determines the validity of a decision, but the importance we attach to the factors involved. How do we assign importance to all the factors and synthesize this diverse information to make the best decision?
4

OBJECTIVITY?
Bias in upbringing: objectivity is agreed upon subjectivity. We interpret and shape the world in our own image. We pass it along as fact. In the end much of it is obsoleted by the next generation.
5

The Meaning of Numbers

The AHP allows one to use numbers or statistics as they naturally arise from measurement. But one needs to exercise caution about what these numbers mean. In the Wall Street Journal of April 11, 2001, under Review and Outlook, an article about the meaning of scales appeared. It emphasized that units of measurement are fundamentally arbitrary and may make no sense to an individual and the individual must translate the meaning of a measurement to something that is more familiar and that is used more often. Scales have both an objective and a subjective aspect. They rely, not only on scientifically calibrated scales, but also on feel. It also argued that the Metric system, neat as it may be, has no organic roots. The meter is taken as one ten-millionth of the (inaccurately) measured distance between the poles. But a foot derives from the length of a foot, an inch from the first segment of the thumb and a cup from the amount of water a man can hold in both hands. What truly matters about units of measurement is that we understand the amounts and distances to which they correspond. If one has a good feel for the size of an ounce without doing any calculations, it does not matter that dividing a pint by 16 is inconvenient. The conclusion is that the impact of measurement on the mind and the corresponding judgment that counts and not the judgment itself. One can force new measurements onto packages, but one cannot force them into the way people think.
6

Knowledge is Not in the Numbers but in the Meaning of the Numbers

Isabel Garuti is an environmental researcher whose father-in-law is a master chef in Santiago, Chile. He owns a well known Italian restaurant called Valerio. He is recognized as the best cook in Santiago. Isabel had eaten a favorite dish risotto ai funghi, rice with mushrooms, many times and loved it so much that she wanted to learn to cook it herself for her husband, Valerios son, Claudio. So she armed herself with a pencil and paper, went to the restaurant and begged Valerio to spell out the details of the recipe in an easy way for her. He said it was very easy. When he revealed how much was needed for each ingredient, he said you use a little of this and a handful of that. When it is O.K. it is O.K. and it smells good. No matter how hard she tried to translate his comments to numbers, neither she nor he could do it. She could not replicate his dish. Valerio knew what he knew well. It was registered in his mind, this could not be written down and communicated to someone else. An unintelligent observer would claim that he did not know how to cook, because if he did, he should be able to communicate it to others. But he could and is one of the best.
7

Comparison Matrix
Given: Three apples of different sizes.

Apple A

Apple B

Apple C

We Assess Their Relative Sizes By Forming Ratios

Size Comparison Apple A Apple A S1/S1 Apple B S1/S2 Apple C S1/S3

Apple B
Apple C

S2 / S1
S3 / S1

S2 / S2
S3 / S2

S2 / S3
S3 / S3
8

Pairwise Comparisons
Size
Apple A
Size Comparison Apple A

Apple B
Apple B

Apple C
Apple C
Resulting Priority Eigenvector Relative Size of Apple

Apple A

6/10

Apple B

1/2

3/10

Apple C

1/6

1/3

1/10

When the judgments are consistent, as they are here, any normalized column gives the priorities.
9

Consistency
In this example Apple B is 3 times larger than Apple C. We can obtain this value directly from the comparisons of Apple A with Apples B & C as 6/2 = 3. But if we were to use judgment we may have guessed it as 4. In that case we would have been inconsistent.
Now guessing it as 4 is not as bad as guessing it as 5 or more. The farther we are from the true value the more inconsistent we are. The AHP provides a theory for checking the inconsistency throughout the matrix and allowing a certain level of overall inconsistency but not more.

10

Verbal Expressions for Making Pairwise Comparison Judgments

Equal importance Moderate importance of one over another Strong or essential importance

Very strong or demonstrated importance

Extreme importance

11

Fundamental Scale of Absolute Numbers Corresponding to Verbal Comparisons

1 Equal importance

3
5

Moderate importance of one over another

Strong or essential importance

7
9

Very strong or demonstrated importance

Extreme importance

12

Which Drink is Consumed More in the U.S.?

An Example of Estimation Using Judgments
Drink Consumption in the U.S. Coffee Wine Tea Beer Sodas Milk Coffee 1 Wine 9 1 2 9 9 9 Tea 5 1/3 1 3 4 3 Beer 2 1/9 1/3 1 2 1 Sodas 1 1/9 1/4 1/2 1 1/2 Milk 1 1/9 1/3 1 2 1 Water 1/2 1/9 1/9 1/3 1/2 1/3

1/9
1/5 1/2 1 1 2

Water

The derived scale based on the judgments in the matrix is:

Coffee Wine Tea Beer Sodas Milk Water .177 .019 .042 .116 .190 .129 .327 with a consistency ratio of .022. The actual consumption (from statistical sources) is: .180 .010 .040 .120 .180 .140 .330

13

Estimating which Food has more Protein

Food Consumption in the U.S. A: Steak B: Potatoes C: Apples D: Soybean E: Whole Wheat Bread

A
1

B
9

C
9

D
6

E
4

F
5

G
1

1/9
1/9 1/6 4 1/5 1

1
1 2 4 3 4

1
1 3 3 5 9

1/2
1/3 1 2 1 6

1/4
1/3 1/2 1 1/3 3

1/3
1/5 1 3 1 5

1/4
1/9 1/6 1/3 1/5 1

F: Tasty Cake
G: Fish

The resulting derived scale and the actual values are shown below:
Steak Potatoes Apples Soybean W. Bread T. Cake Derived .345 .031 .030 .065 .124 .078 Actual .370 .040 .000 .070 .110 .090 (Derived scale has a consistency ratio of .028.) Fish .328 .320
14

Extending the 1-9 Scale to 1-

The 1-9 AHP scale does not limit us if we know how to use clustering of similar objects in each group and use the largest element in a group as the smallest one in the next one. It serves as a pivot to connect the two.

We then compare the elements in each group on the 19 scale get the priorities, then divide by the weight of the pivot in that group and multiply by its weight from the previous group. We can then combine all the groups measurements as in the following example comparing a very small cherry tomato with a very large watermelon. 15

.28 Lime

.65

.22 Honeydew

.70

.10 Honeydew

.60

.60 =6 .10 6 5.69 34.14

This means that 34.14/.07 = 487.7 unripe cherry tomatoes are equal to the oblong watermelon

16

Basic Decision Problem

Criteria: Car: Alternatives: Low Cost > Operating Cost > Style A V B B V A B V A

Suppose the criteria are preferred in the order shown and the cars are preferred as shown for each criterion. Which car should be chosen? It is desirable to know the strengths of preferences for tradeoffs.

17

Decision Making is a process that requires us to:

Understand and define the problem as completely as possible. Structure a problem as a hierarchy or as a system with dependence loops. Elicit judgments that reflect ideas, feelings and emotions.

Represent those judgments with meaningful numbers.

Synthesize Results

18

Goal Satisfaction with School

Learning

Friends

School Life

Vocational Training

College Prep.

Music Classes

School A

School B

School C

19

School Selection
L 1 1/4 F 4 1 SL 3 7 1 5 5 6 VT 1 3 1/5 1 1 3 CP 3 1/5 1/5 1 1 1/3 MC Weights 4 .32 1 1/6 1/3 3 1 .14 .03 .13 .24 .14
20

Learning Friends School Life Vocational Trng. College Prep. Music Classes

Comparison of Schools with Respect to the Six Characteristics

Learning A B C A B C 1 3 2 1/3 1/2 1 1/3 3 1
Priorities

Friends A B C A B C 1 1 1 1 1 1 1 1 1

Priorities

Priorities

Priorities

Priorities

Vocational Trng. Priorities

A A B C 1 1/9 1/7

B 9 1 5

C 7 1/5 1 .77 .05 .17 A B C

College Prep. A B C
1 2 1 1/2 1 1/2 1 2 1

Music Classes

B 6 1 3

21

Composition and Synthesis

Impacts of School on Criteria

.32 L

A
B

22

The School Example Revisited Composition & Synthesis:

Impacts of Schools on Criteria
Distributive Mode (Normalization: Dividing each entry by the total in its column)
.32 L A B C .16 .59 .25 .14 F .33 .33 .33 .03 SL .45 .09 .46 .13 VT .77 .05 .17 .24 CP .25 .50 .25 .14 MC .69 .09 .22 Composite Impact of Schools .37 .38 .25 .32 L A B C .27 1 .42

Ideal Mode (Dividing each entry by the maximum value in its column)
.14 F 1 1 1 .03 SL .98 .20 1 .13 VT 1 .07 .22 .24 CP .50 .50 .50 .14 Composite NormalMC Impact of ized Schools 1 .65 .34 .13 .32 .73 .50 .39 .27

The Distributive mode is useful when the The Ideal mode is useful in choosing a best uniqueness of an alternative affects its rank. alternative regardless of how many other The number of copies of each alternative similar alternatives there are. also affects the share each receives in allocating a resource. In planning, the scenarios considered must be comprehensive and hence their priorities depend on how many there are. This mode is essential for ranking criteria and sub-criteria, and when there is dependence.
23

Evaluating Employees for Raises

GOAL
Dependability (0.075) Outstanding (0.48) .48/.48 = 1 Education (0.200) Doctorate
(0.59) .59/.59 =1

Very Good (0.28) .28/.48 = .58

Good (0.16)

Masters
(0.25).25/.59 =.43

6-15 years (0.25)

3-5 years (0.10) 1-2 years (0.04)

Very Good (0.21)

Good (0.11) Poor (0.04)

Above Avg. (0.23)

Average (0.10) Negative (0.04)

Above Avg. (0.23)

Average (0.14) Below Avg. (0.06) Unsatisfactory (0.03) 24

.16/.48 = .33

Final Step in Absolute Measurement

Rate each employee for dependability, education, experience, quality of work, attitude toward job, and leadership abilities.
Dependability 0.0746 Peters, T. Hayat, F. Becker, L. Adams, V. Kesselman, S. Kelly, S. Joseph, M. Tobias, K. Washington, S. OShea, K. Williams, E. Golden, B. Outstand Outstand Outstand Outstand Good Good Blw Avg. Outstand V. Good Outstand Outstand V. Good Education 0.2004 Doctorate Masters Masters Bachelor Bachelor Bachelor Hi School Masters Masters Hi School Masters Bachelor Experience 0.0482 >15 years >15 years >15 years 6-15 years 1-2 years 3-5 years 3-5 years 3-5 years 3-5 years >15 years 1-2 years .15 years Quality 0.3604 Excellent Excellent V. Good Excellent Excellent Excellent Excellent V. Good V. Good V. Good V. Good V. Good Attitude 0.0816 Enthused Enthused Enthused Abv. Avg. Enthused Average Average Enthused Enthused Enthused Abv. Avg. Average Leadership 0.2348 Outstand Abv. Avg. Outstand Average Average Average Average Abv. Avg. Abv. Avg. Average Average Abv. Avg. Total Normalized

1.000 0.752 0.641 0.580 0.564 0.517 0.467 0.466 0.435 0.397 0.368 0.354

0.153 0.115 0.098 0.089 0.086 0.079 0.071 0.071 0.066 0.061 0.056 0.054

The total score is the sum of the weighted scores of the ratings. The money for raises is allocated according to the normalized total score. In practice different jobs need different hierarchies.
25

Focus:

At what level should the Dam be kept: Full or Half-Full

Decision Criteria:

Financial

Political

Envt Protection

Social Protection

Decision Makers:

Congress

Dept. of Interior

Courts

State

Lobbies

Factors:

Clout

Legal Position

Irreversibility of the Envt

Archeological Problems

Current Financial Resources

Groups Affected:

Farmers

Recreationists

Power Users

Environmentalists

Objectives:

Irrigation

Flood Control

Flat Dam

White Dam

Cheap Power

Protect Environment

Alternatives:

Half-Full Dam

Full Dam 26

Should U.S. Sanction China? (Feb. 26, 1995)

BENEFITS
Protect rights and maintain high Incentive to make and sell products in China (0.696) Rule of Law Bring China to responsible free-trading 0.206) Help trade deficit with China (0.098)

Yes .60 No .40

Yes 0.729 No 0.271

Yes .50 No .50

COSTS
\$ Billion Tariffs make Chinese products more expensive (0.094) Retaliation (0.280) Being locked out of big infrastructure buying: power stations, airports (0.626)

Yes .90 No .10

Yes 0.787 No 0.213

Yes .75 No .25

RISKS
Long Term negative competition (0.683) Effect on human rights and other issues (0.200) Harder to justify China joining WTO (0.117)

Yes .30 No .70

Yes 0.597 No 0.403

Result:

YES

= 1.55

NO

.271 .213 x .403

= 3.16 27

Benefits/Costs*Risks

7 6 5 4 3 2 1

. . . . .. .. . . . . .. ... . . . . ..... . .. . .
18 30 42 54 66 78 90

. . .. .

No Yes

0 6

102 114 126 138 150 162 174 186 198 210

Experiments

28

Best Word Processing Equipment

Focus

Benefits

Criteria

Time Saving

Filing

Quality of Document

Accuracy

Features

Training Required

Printer Speed

Alternatives

Focus

Costs

Criteria

Capital

Supplies

Service

Training

Alternatives

Lanier .54

Syntrex .28

Oyx .18 29

Best Alternative
30

Group Decision Making and the Geometric Mean

Suppose two people compare two apples and provide the judgments for the larger over the smaller, 4 and 3 respectively. So the judgments about the smaller relative to the larger are 1/4 and 1/3.

Arithmetic mean 4+3=7 1/7 1/4 + 1/3 = 7/12

Geometric mean 4 x 3 = 3.46 1/ 4 x 3 = 1/4 x 1/3 = 1/ 4 x 3 = 1/3.46
That the Geometric Mean is the unique way to combine group judgments is a theorem in mathematics.
31

TWO EXERCISES
Do a simple decision like choosing a vacation place or a job or a city to move to or a restaurant and do it by paired comparisons throughout.
Do the same decision using the same weights for the criteria, but then set up standards for the criteria and rate the alternatives one at a time.
32