You are on page 1of 49

HIERARCHICAL DECISION-MAKING DECISIONPROCESSES

Concepts and Importance


Parts One and Two

B. Srdjevic, 2006.

Contents

PART ONE: Concepts, Importance, Difficulties, And Necessities Embedded In Hierarchical Decision-Making Processes PART TWO: Some Important Aspects Related To Hierarchical Decision-Making

HIERARCHICAL

DECISION-MAKING

PROCESS

HIERARCHICAL
Means
 

 

Structuring the decision-making process decisionIdentifying decision elements (DEs) and key players decision makers (DMs) Defining decision lines: vertical and horizontal The hierarchical arrangement has been found to be the best way for human beings to cope with complexity.

HIERARCHICAL

DECISION-MAKING

PROCESS

DECISIONDECISION-MAKING
Means
 

Understanding the problem in hand Following selected consistent and coherent methodology eligible for evaluating DEs DMs being responsible in validating objectives, criteria and alternatives (willingness, good attitude, best knowledge etc.) Understanding that there will be consequences of decisions made

HIERARCHICAL

DECISION-MAKING

PROCESS

PROCESS
Means


Considering the spatial and temporal component of the problem Recognizing the process in selected methodology of evaluating DEs. Necessity of feed-back loops envisioned. feed-

Part I

Concepts, Importance, Difficulties, And Necessities Embedded In Hierarchical DecisionDecision-Making Processes

Typical Stages In The Decision-Making Process


Not necessarily that order

Information Collection and Management Modelling and Rational Decision Support Visualization and the Human Interface Group Decision-Making Knowledge Capture and Representation DSS Integration
Overlaps and feedback are usual !

Procedural Steps Within The Decision Making Framework


typical in water resources
No consensus jet which one is the best or most suitable for all cases plan/mnmt/cntrl

1. Selection of an approach and creation of adoptable framework for multicriteria analysis (MCA) and evaluation of decision elements. 2. Defining sustainability criteria and setting objectives, such as: (a) water availability/suitability criteria; (b) energy output/input ratios; (c) water requirements; (d) environmental costs; (e) economic viability with considerations of governmental and societal viewpoints; (f) net present values from governmental and social viewpoints; and(g) local entities' acceptability. 3. Formulation of planning or/and management alternatives. 4. Integrating output from spatial analysis using GIS, resource accounting and economic valuation for multi-criteria evaluation.

Procedural Steps Within The Decision Making Framework

cont.

e.g. sub-criteria, sub-sub-criteria, alternatives )

5. Decomposition of criteria set and other decision elements on a hierarchical basis. 6. Development of pay-off (decision) matrix. 7. Development of questionnaires for decision makers. 8. Administering the questionnaires including interviews with entities leaders and project officials. 9. Assigning weights to criteria. 10. Applying MCA (multicriteria analysis) and deriving decisions by using selected decision-making tool(s).

Decision-Making Concepts & Methodologies

Perhaps the most common formal approach to making a choice among alternatives is to list the pros and cons of each alternative. A list of pros and cons can often be embedded in a memorandum such as:
Issue: (i.g. Evaluation of alternative scenarios) Alternative 1 Pros: Cons: Alternative 2 Pros: Cons:

  

Define pros and cons of alternatives Discuss the alternatives and their pros and cons Avoid misuse of numbers (cardinal, ordinal, ratios, intervals cardinal,
Typical misuses

1. Not weighting the factors is commonly an error of omission and is often fairly obvious. 2. A more serious error is the inappropriate addition of ranks. The more. scores given as ordinal numbers represent ranks, but nothing more It is wrong to add these numbers because any results are meaningless! Question: Answer: How one can identify such mistakes? By thinking about the meaning of the numbers.

An example:  A scenario scoring a 1 on robustness is ranked better than a scenario scoring a 2, but how much better? Is the interval between scenarios ranked 1 and 2 the same as the interval between scenarios ranked 6 and 7? Not necessarily. necessarily The scenario ranked 1 might be 2 million R$ higher than the scenario ranked 2, while the interval between the 6 and 7 scenarios might be only 50 thousand R$ (40x less). The numbers ordinal in nature cannot be added !

The misuse of numbers is one reason that numerical analyses are sometimes flawed. But the misuse of numbers is not a reason to forego using them. We just must be careful that a sound theoretical foundation exists for whatever we do.

Levels of Measurement

Ratio Interval Ordinal Nominal

(Each level has all of the meaning of the levels below plus additional meaning.)

Nominal numbers

These numbers are just numerical representations for names. Nominal numbers are used for identification purposes only and imply nothing about the ordering. Example: Telephone numbers and personal ID numbers are nominal. Is one is older or better than someone else because his telephone number is higher? Obviously not. Remark: People rarely make mistakes with nominal numbers. However, errors arising from the misuse of ordinal numbers are not so rare.

Ordinal numbers

They imply an order or ranking among elements. The order may be either increasing or decreasing depending on the application. A ranking implies an ordering among elements but nothing more. It does not imply anything about the differences (or intervals) between items. Example: If we know only that a professional football team finished in second place at the end of the season, we do not know if the team was one game behind the first place team or 15 games behind. Remark: Care must be taken not to add or multiply ordinal data. Errors arising from the addition of ordinal data are very common.

Interval numbers

Interval scale data possesses the meaning of Nominal and Ordinal data, as well as having meaning about the intervals between objects. Corresponding intervals on different parts of an interval scale have the same meaning. Example: If we have interval level data then we can infer that the interval between two objects with values of 20 and 5 (an interval of 15) is equivalent to the interval between two objects with values of 80 and 65. Remark: Interval level data can be used in arithmetic operations such as addition and multiplication. However, after adding interval level data, one can not infer that a total of 100 is twice as good as a total of 50. If one were to allocate resources based on this inference, then the allocation would be incorrect.

Ratio numbers

Ratio level data (sometimes called ratio scale) have Nominal, Ordinal, and Interval properties, as well as the property of ratios. Corresponding ratios on different parts of a ratio scale have the same meaning. Example: If we have ratio scale data, then the ratio between two objects with values of 100 and 50 is equivalent to the ratio of two objects with values of 6 and 3. Remark: A ratio scale is often defined as one having a true zero point. However, it is easier to think of a ratio scale as one for which equivalent ratios are considered equal. Temperature measured on the Celsius scale is not a ratio measure, since it would be wrong to infer that there is twice as much heat when the temperature is 40 degrees as when the temperature is 20 degrees. (If the Kelvin scale is used instead, which has the ratio property, then such an inference would be correct.)

Mathematical operations allowed

Nominal, Ordinal, Interval, Ratio succeeding scales have additional meaning and can be used in more arithmetic operations as summarized below:  Addition/subtraction and multiplication/division require at least interval level meaning.  An interval level number can be multiplied by a constant or a ratio level number but cannot be multiplied by another interval level number.   There are no restrictions when using ratio level numbers. A decision method that produces ratio scale numbers is considered the most flexible and accurate. accurate.

Weights and Score Matrix


Typical Weights and Scores Matrix

Usual Procedure
Typical Weights and Scores Matrix

 Criteria are assigned weights using a scale such as 0 to 10 or 0 to 100.  Then each alternative is scored against each criterion.  The alternative scores for each criterion are then multiplied by the weight for the criterion and summed to give a total score for each alternative  This score represents the overall preference for or performance of the alternative.  If used carefully, weights and scores can be an effective methodology.

Difficulties

There are, however, several practical difficulties:    When assigning weights, what do the numbers really mean? On a scale of 0 to 100, what is an 80, or what is a 40? When one give an 80 to one criterion and give a 40 to another, does he really mean that the 80 is twice as important as the 40? If the answer is yes, then the weights possess the ratio scale property. But how can one be consistent enough to insure this if he is dealing with 20, 30 or 100 criteria? If he assigned an 80 to the first criterion and later assign a weight of 10 to the 95th criterion, can he remember what he did earlier and does he really mean that the 95th criterion is only one eighth as important as the first criterion?

Channel capacity and short term memory

Experiments have proven that the human brain is limited in both its short-term memory capacity and its discrimination ability (channel capacity) to about 7 things. If a person has to choose from a range of 20 alternatives, he will give inaccurate answers because the range exceeds the bandwidth of his channel for perception. In many cases, 7 alternatives are the approximate limit of his channel capacity. -- James Martin

Martin s conclusion is based on the results of numerous psychological experiments, including the well known study The Magical Number Seven, Plus or Minus Two: Some Limits on Our Capacity for Information Processing, Psychological Review, (Vol. 63, No. 2, March 1956), pp. 81-97, by G. A. Miller. The distribution is a bell shaped curve with an average of 7. Some people can recall more than 7, some fewer. But only about 11% of the population can recall 9 things from their short term memory, still fewer 10 things, and so on.

Need for Hierarchical Structure

 If we try to assign weights to 20, 30 or 100 columns, we are bound to be many mistakes! We can cope with this complexity as we do with other complex situations - by arranging the criteria into groups, each group having sub-groups, and so on.  If one try to recall a sequence of 9 or 11 digits as someone reads them, he will probably find himself grouping (psychologists call this chunking) the digits into groups in an effort to overcome the limitations of his own short-term memory.  Again, hierarchical arrangement has been found to be the best way for human beings to cope with complexity.

Orders of Magnitude

Another problem arises when a weights and scores approach involves more than a handful of criteria (columns). Some of the criteria might be orders of magnitude more important than others. If one criterion is assigned a 0.02 on a scale of 0-10, and another is assigned a 9.0, do we really mean that one is 450 times more (less) important than another? Our ability to accurately compare things that differ by orders of magnitude is not nearly as good as our ability to compare things that differ by less than an order of magnitude. A hierarchical grouping avoids this problem as well.

Arbitrary assignment

Difficulty with the weights and scores methodology stems from the assignment of weights and scores in what often appears to be an arbitrary fashion. How can we justify that a criterion (such as customer perception) was given an 8? What does the 8 really mean? And why not a 7, or a 9? Similarly, when scoring a particular alternative with respect to a criterion such as customer perception, we may refer to customer interview studies and past experience and assign a 3. But what does the 3 really mean?

Words instead of numbers

Justification would be much easier if we use a less precise way of expressing judgments, such as words instead of numbers. Suppose we use words instead of numbers. By intuition - words are often easier to justify than numbers. For example, if you say that, with respect to corporate image, alternative A is 3 times more preferable than alternative B, can you justify why it is exactly 3? Why not 2.9, or 3.1? But if you said, instead, that A is moderately more preferable than B, this can be justified with a variety of arguments, including, perhaps, some hard data.

But what can be done with the words ?

 

Question: What to do with verbal judgment such as moderate ? Words have different meanings to different people. In fact, anyone can put arbitrary numbers behind words in a computer program. But will the numbers accurately reflect the meaning the words had to the individual or group making the judgments? Even if the decision-maker specifies numerical equivalencies for words, will he or a group of his colleagues consistently remember the assignments accurately enough to insure that the results reflect their actual judgments? Will errors due to the use of imprecise words be a problem? Again answer is YES !

THEREFORE

THEREFORE
 We have to introduce sophisticated approaches that recognize necessity of dealing with rather weak than rigorous optimization methods, tolls and supporting (computerized) systems. We have to accept the truth that there is no best solution methodology for all problems encountered. We have to fund our decisions within correct theoretical framework. We have to perform valid sensitivity analyses and verify decision made by monitoring and evaluating consequences incurred. Quite often, we have to simultaneously handle individual judgments (subjectivity) of decision makers along with objectivity issues such as technical (or economical) value of certain decision element within related functional schema in the decision tree.

 

Part II

Some Important Aspects Related To Hierarchical Decision-Making Decision-

Aspects

1. 2. 3. 4. 5. 6.

Individual and Group Decision-Making (D-M) Decision(DParticipative D-M models DD-M with complete and incomplete information D-M with certainty and uncertainty D-M classifications related to problem structure Mathematics, modeling, models, tools, systems (decision support)

Hints for implementation of a hierarchical decision-making process decision-

1. Carefully analyse the decision problem in hand Carefully and structure it. 2. Learn from the others. 3. Use existing software tools (develop additional or new only if necessary). 4. Follow-up the consequences of the decision Followmade, and be ready to repeat some calculations.

Various issues

1. Clustering 2. Groups, aggregations 3. Optimism and pessimism of the DM(s)

Clustering the decision elements


 4 levels  5 criteria/7 sub-criteria  5 alternatives  Total=17 decision elements

L1: Goal

L2: Criteria

L3: Sub-criteria

L4: Alternatives

First pass

 cluster 2 criteria into 1  cluster 3 sub-criteria into 1  to reduce total number of elements to 14

to obtain

 4 levels  4 criteria/5 sub-criteria  5 alternatives  Total=14 decision elements

Now perform additional clustering

To obtain the final hierarchy

 3 levels (one level less one less)


 4 criteria (no sub-criteria no sub-criteria)  5 alternatives  Total=9 decision elements

Compare

CLUSTERING

EXPANDING

Groups, aggregations

- Normalized weight of the k th member of a group G - Weight of the i th alternative for k th member of a group G - Aggregated weight of the i th alternative for a group G

In either aggregation, WAMM or GMM, the final additive normalization of the weights of all alternatives is required !

Optimism and Pessimism of the DM


Define the optimism level of alternative i across all criteria j=1, ,n be

oi ! max vij j !1,...,n


where vij is the performance of alternative i for criterion j.

Thus, oi is the value of the best consequence that can result if alternative i is taken. The max-max return criterion is: Choose alternative k such that:

ok ! max oi ! max max vij


i !1,...,m i j

Hurwicz (1951) argued that the DM should rank alternatives according to the weighted average of the security and optimism levels: si + (1 Where )oi

0 e E e 1 is the optimism-pessimism index of the DM.

Hurwicz recommends the decision rule: Choose alternative k such that:

Esk  (1  E )ok ! max{Esi  (1  E )oi }


i

This is what is known as Hurwicz s optimism-pessimism index

There are two available decisions: D1 and D2. Which one to use, or how to combine them into the one: D ? The was the Whoanswer is simple: That was the or pessimist, D1 D2 D2 ? Fine tuner of optimism v pessimism

1 Not important

5 Very important

9 Absolutely important

Where is D ?

1 Not important

5 Very important

9 Absolutely important

Geometric

Additive

1 Not important

3 Weak importance

5 Very important

9 Absolutely important

What would be your preference, 3 or 5 ? Theory says 3, because the judgments are fairly divergent. Theory suggests that there was an inconsistency, so better is to be little pessimistic . Therefore 3. Do you agree ? Me NOT, because I am an optimist. But I shall follow the theory rather.

To be continued with Parts III and IV

You might also like