╨╧рб▒с>■  ЦШ■   ФХ                                                                                                                                                                                                                                                                                                                                                                                                                                            ье┴Y ┐"abjbjєWєW ЎС=С="]      ]ТТТТТТТжжжжж ▓╠ж╔╢ККККККККОРРРРРР$Їsd┤ТККККК┤╨ТТККК╨╨╨КТКТКОжжТТТТКО╨╛╨ОТТОК~ `иqQ║E┐жжи(О Decisions Under Uncertainty  
1. Purpose
2. Introduction
3. Decision Rules
4. References
5. Related Links
6. Bibliography
PPT presentation
Decisions Under Uncertainty
ENME 808s Product & System Cost Analysis 
End of Semester Class Project
Authors: Brian Reynolds, Brian Schaeffer
Professor: Dr. Peter Sandborn
1. Purpose: Explains the purpose and use of this tutorial.
2. Introduction: An introduction to "decisions under uncertainty".
3. Decision Rules:  Defines and explains the different decision rules commonly used for decisions under uncertainty and illustrates their use in a hypothetical decision problem.
4. References: References to information used in this tutorial.
5. Related Links: WWW links relevant to this topic.
6. Bibliography: Extended bibliography of sources relevant to this topic.

1. Purpose:  The purpose of this project is to give a tutorial level summary of "decisions under uncertainty" and its application to product and system cost analysis.  Included will be the summary of the topic, a list of relevant sources with links, and an extended bibliography.  The depth and breadth of coverage should be equivalent to one class lecture.

2. Introduction:

Typically, personal and professional decisions can be made with little difficulty.  Either the best course of action is clear or the ramifications of the decision are not significant enough to require a great amount of attention.  On occasion, decisions arise where the path is not clear and it is necessary to take substantial time and effort in devising a systematic method of analyzing the various courses of action. [2,3]

When a decision maker must choose one among a number of possible actions, the ultimate consequences of some if not all of these actions will generally depend on uncertain events and future actions extending indefinitely far into the future.  With decisions under uncertainty, the decision maker must:

1. Take an inventory of all viable options available for gathering information, for experimentation, and for action;
2. List all events that may occur;
3. Arrange all pertinent information and choices/assumptions made;
4. Rank the consequences resulting from the various courses of action;
5. Determine the probability of an uncertain event occurring.
[2,3]
Upon systematically describing the problem and recording all necessary data, judgments, and preferences, the decision maker must synthesize the information set before him/her using the most appropriate decision rules.  Decision rules prescribe how an individual faced with a decision under uncertainty should go about choosing a course of action consistent with the individualТs basic judgments and preferences.  This website will describe five such decision rules commonly used in industry [2,3]: 
  • Hurwicz criterion; 
  • Laplace insufficient reason criterion; 
  • Maximax criterion; 
  • Maximin criterion; 
  • Savage minimax regret criterion.

3. Decision Rules:

A tool commonly used to display information needed for the decision process is a payoff matrix or decision table.  The table shown below is an example of a payoff matrix.  The A's stand for the alternative actions available to the decision maker.  These actions represent the controllable variables in the system.  The uncertain events or states of nature are represented by the S's.  Each S has an associated probability of its occurance, denoted P.  (However, the only decsion rule that makes use of the probabilities is the Laplace criterion.)  The payoff is the numerical value associated with an action and a particular state of nature.  This numerical value can represent monetary value, utility, or both.  This type of table will be used to illustrate each type of decision rule. 
 

Actions\States S1 (P=.25) S2 (P=.25) S3 (P=.25) S4 (P=.25)
A1 20 60 -60 20
A2 0 20 -20 20
A3 50 -20 -80 20
Table 1: General Payoff Matrix style from Chankong [4].  This generic/hypothetical example illustrates 3 different actions that can be taken, and 4 different possible, uncertain states of nature with their respective payoffs.

i.  Hurwicz criterion.

This approach attempts to strike a balance between the maximax and maximin criteria.  It suggests that the minimum and maximum of each strategy should be averaged using a and 1 - a as weights. a represents the index of pessimism and the alternative with the highest average is selected.  The index a reflects the decision makerТs attitude towards risk taking.  A cautious decision maker will set a = 1 which reduces the Hurwicz criterion to the maximin criterion.  An adventurous decision maker will set a = 0 which reduces the Hurwicz criterion to the maximax criterion. [1]  A decision table illustrating the application of this criterion (with a = .5) to a decision situation is shown below.
 

Actions\States S1 S2 S3 S4 a = .5
A1 20 60 -60 20 0
A2 0 20 -20 20 0
A3 50 -20 -80 20 -15
Table 2: Hurwicz criterion illustration (a = .5); Here the probability of each state is not considered; results in a tie between the first two alternatives.

ii. Laplace insufficient reason criterion.

The Laplace insufficient reason criterion postulates that if no information is available about the probabilities of the various outcomes, it is reasonable to assume that they are equally likely.  Therefore, if there are n outcomes, the probability of each is 1/n.  This approach also suggests that the decision maker calculate the expected payoff for each alternative and select the alternative with the largest value.  The use of expected values distinguishes this approach from the criteria that use only extreme payoffs.  This characteristic makes the approach similar to decision making under risk. A table illustrates this criterion below. [1] 
 

Actions\States S1 (P=.25) S2 (P=.25) S3 (P=.25) S4 (P=.25) Expected Payoff:
A1 20 60 -60 20 0
A2 0 20 -20 20 5
A3 50 -20 -80 20 -7.5
Table 3: Laplace insufficiency illustration; Second alternative wins when expected payoff is calculated between equiprobable states.

iii. Maximax criterion.

The maximax criterion is an optimistic approach.  It suggests that the decision maker examine the maximum payoffs of alternatives and choose the alternative whose outcome is the best.  This criterion appeals to the adventurous decision maker who is attracted by high payoffs.  This approach may also appeal to a decision maker who likes to gamble and who is in the position to withstand any losses without substantial inconvenience. See the table below for an illustration of this criterion. [1] 
 

Actions\States S1 S2 S3 S4 Max Payoff
A1 20 60 -60 20 60
A2 0 20 -20 20 20
A3 50 -20 -80 20 50
Table 4: Maximax illustration; First alternative wins.

iv. Maximin criterion.

The maximin criterion is a pessimistic approach.  It suggests that the decision maker examine only the minimum payoffs of alternatives and choose the alternative whose outcome is the least bad.  This criterion appeals to the cautious decision maker who seeks to ensure that in the event of an unfavorable outcome, there is at least a known minimum payoff.  This approach may be justified because the minimum payoffs may have a higher probability of occurrence or the lowest payoff may lead to an extremely unfavorable outcome. This criterion is illustrated in the table below. [1] 
 

Actions\States S1 S2 S3 S4 Min payoff
A1 20 60 -60 20 -60
A2 0 20 -20 20 -20
A3 50 -20 -80 20 -80
Table 5: Maximin illustration. Second alternative wins.

v. Savage minimax regret criterion.

The Savage minimax regret criterion examines the regret, opportunity cost or loss resulting when a particular situation occurs and the payoff of the selected alternative is smaller than the payoff that could have been attained with that particular situation.  The regret corresponding to a particular payoff Xij is defined as Rij = Xj(max) Ц Xij where Xj(max) is the maximum payoff attainable under the situation Sj.  This definition of regret allows the decision maker to transform the payoff matrix into a regret matrix.  The minimax criterion suggests that the decision maker look at the maximum regret of each strategy and select the one with the smallest value.  This approach appeals to cautious decision makers who want to ensure that the selected alternative does well when compared to other alternatives regardless of what situation arises.  It is particularly attractive to a decision maker who knows that several competitors face identical or similar circumstances and who is aware that the decision makerТs performance will be evaluated in relation to the competitors. This criterion is applied to the same decision situation and transforms the payoff matrix into a regret matrix.  This is shown below. [1]  
 

Actions\States R1 R2 R3 R4 Max Regret
A1 30 0 40 0 40
A2 50 40 0 0 50
A3 0 80 60 0 80
Table 5: Minimax illustration. First alternative wins.

4. Related Links:
5. References:

[1]  Z.W. Kmietowicz, and A.D. Pearman. Decision Theory and Incomplete Knowledge. p. 7-9. Gower Publishing Company Limited:  Aldershot, Hampshire, England. 1981.

[2] Raiffa, Howard. Decision Analysis: Introductory Lectures on Choices Under Uncertainty. p. ix. Addison-Wesley Publishing Company: Reading, Massachusetts. 1970.

[3] Schlaifer, Robert. Analysis of Decisions Under Uncertainty. p. 65-68. Robert E. Krieger Publishing Company: Huntington, New York. 1978.

[4] Vira Chankong, Yacov Y. Haimes, Multiobjective Decision Making: Theory and methodology, (North Holland series in system science and engineering; 8). p. 32-38. Elsevier Science Publishing Co., Inc. New York, NY, 1983.

6. Bibliography:
  • Vira Chankong, Yacov Y. Haimes, Multiobjective Decision Making: Theory and methodology, (North Holland series in system science and engineering; 8).  Elsevier Science Publishing Co., Inc. New York, NY, 1983.
  • Sheldon M. Ross, A First Course in Probability -- 5th edition.  Prentice-Hall, Inc. Simon & Schuster, Upper Saddle River, NJ, 1998
  • Donovan Young, Modern Engineering Economy. John Wiley & Sons, Inc. New York, NY, 1993.
  • Jati K. Sengupta, Optimal Decisions Under Uncertainty.  Springer-Verlag. New York, NY, 1985.
  • Paul J. Ossenbruggen, Fundamental Principles of Systems Analysis and Decision-making.  John Wiley & Sons, Inc. New York, NY, 1994.
"a?FMЩє7д╥┌√%*ВаMi╬1Ъ√# $ } ─ ц : ¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤"a■: Ш ь E K T U q v ─  7 И ┴ ■ I Ш ф <РрLЧ╜TЧЭ椤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤жз├╚═фъєЇ\гя7Рщя°∙6<EFbg╨¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤╨cмў@Мды4В╔ЎCФ╪/<Кды7Ж╬^еЇB¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤BV[`ЫЬбЁёЎ127rsx╗┴╟═╓╫є°¤#$@¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤@EН│ HР┘)q╕ H Р ┌ !$!O!T!Й!К!╗!╝!э!ю!" "Q"W"X"¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤X"]"Ж"З"░"▒"┌"█"##/#5#6#;#d#e#Н#О#╖#╕#т#у# $$$$A$B$k$l$¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤l$Ц$Ч$┴$┬$ы$ё$%%K%У%р%5&Z&Э&щ&2'u'~'╔'(^(е(х(ю(:)Й)╛)╔)Ї)¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤Ї)∙).*/*X*Y*В*Г*м*н*╓*╫*ю*Ї*ї*·*#+$+M+N+w+x+в+г+╠+═+╪+▐+▀+ф+¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤ф+ ,,6,7,`,a,Л,М,╡,╢,┴,╟,╚,═,Ў,ў, -!-K-L-v-w-а-б-о-┤-╞-╟-.¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤.^.о.п./O/Ъ/у/;0Ж0╬01c1л1╥1▌12 2B2C2t2u2ж2з2╪2┘2 3 3%3+3¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤+3,313Z3[3Д3Е3о3п3┘3┌3444444D4E4m4n4Ч4Ш4┬4├4ь4э4°4■4 4¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤ 45-5.5W5X5В5Г5н5о5╫5╪5ц5ь5■5 5M6Ц6╬6╧67^7о7∙7K8Х8ч839>9i9¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤i9n9г9д9═9╬9ў9°9!:":K:L:`:f:g:l:Х:Ц:┐:└:щ:ъ:;;>;?;K;Q;R;W;¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤W;А;Б;й;к;╙;╘;■; ;(<)<5<;<<<A<j<k<Ф<Х<┐<└<ъ<ы<==!='=9=:=И=¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤И=╗=╝=·=J>Т>р>*?q?║?@O@t@@к@п@ф@х@AA8A9AbAcAМAНAбAзAиAнA¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤нA╓A╫ABB*B+BUBVBBАBНBУBФBЩB┬B├BыBьBCC@CACjCkCxC~CCДCнC¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤нCоC╫C╪CDD-D.DWDXDeDkD}D~D╠DEELEЦE▌E*FrF╛FGRGЯGчG.HxH├H¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤├H IVIгIЁI;JYJdJПJФJ╔J╩JєJЇJKKGKHKqKrKЖKМKНKТK╗K╝KхKцKLL¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤L8L9LaLbLnLtLuLzLгLдL═L╬LўL°L M!MIMJMVM\M]MbMЛMМM┤M╡M▐M▀MN¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤N N1N2N>NDNVNWNеN╪N▐NфNэNюN OOO+O1O:O;OWO\OйO╞O╦O╨OTPЭP┼P¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤┼P╞P╦P=QvQwQ|QюQXCX║X└X╞X╠X╒X╓XЄXўXAYcY╡Y¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤╡Y Z6ZЖZ╥Z[I[Р[│[\O\Ч\╝\┬\╦\╠\ш\э\9]]]b]g]п]№]E^e^f^k^╖^_¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤_ _!_&_n_к_л_░_ `;`<`A`К`╤`Ў`№`aaaaa"a¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤¤░╨/ ░р=!░"░#Ра$Ра%░ [$@ё $NormalmH <A@Є б<Default Paragraph Font"]Ў    "a`: ж╨B@X"l$Ї)ф+.+3 4i9W;И=нAнC├HLN┼PШU╡Y_"aacdefghijklmnopqrstuvwxyz"ab,<[ip}╧┘№ ,4@NPeЭбр°№o}╬┌0?Чи№_r╚╥╔╓╨┘д▓]$] ╠╬┌]$]  Peter SandbornL\\CALCESERVER\SANDBORN\SHARE\SANDBORN\Cost Course\Projects\reynolds.html.doc @А!]!]°w╒!]!]"]p@GРЗЯTimes New Roman5РАSymbol3&Р ЗЯArial"0И╨hbl<&cl<&x ╦L'гй"е└┤┤А0N^  <Peter SandbornPeter Sandborn■ рЕЯЄ∙OhлС+'│┘0`ИРЬи└╠▄ Ї  ( 4@HPXф<ssPeter SandborneteNormalaPeter Sandborn1teMicrosoft Word 8.0@F├#@ _PID_GUIDфAN{B31F6BEA-B12A-11D3-96EE-00104B608FFF}  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{■   }~АБВГ■   ЕЖЗИЙКЛ■   НОПРСТУ■   ¤   ¤   Ч■   ■   ■                                                                                                                                                                                                                                                                                                                                                                                                                           Root Entry         └FА╢╖U║E┐А&чQ║E┐ЩА1Table            |WordDocument        ЎSummaryInformation(    ДDocumentSummaryInformation8            МCompObj    jObjectPool            А&чQ║E┐А&чQ║E┐            ■                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                           ■       └FMicrosoft Word Document MSWordDocWord.Document.8Ї9▓q