1

CH.2.
DECISION
THEORY AND
DECISION
ANALYSIS
INTRODUCTION

 The most of decision involves the making of choices under

uncertainty, that is choosing from some set of alternative course of
action in situations where one is uncertain about the actual
consequences of each course of action.
 Decision analysis provides a set of concepts and techniques to help
the decision maker in dealing with complex decision problems under
uncertainty and, also enables the decision maker to analysis a
complex situation with many different alternatives' stages, and
sequences.
 In other words, decision analysis concerns itself with choosing the
best act from a set of possible acts, given uncertainty as the event
which may exist.
 In simple words decision analysis concerns with choosing of best
things from the set which may exist.

2
WHAT IS DECISION THEORY?

 Simply put, decision theory is an analysis of decision-making

process.
 It seeks to evaluate how choices are made.

 Decision theory is a logical study of how decisions are made in a

structure or system where the decision environment is uncertain and
the decision variables unknown.
 Decision theory draws tools from mathematics, philosophy, statistics
and psychology in analyzing how decisions are made.
 This theory also has to do with how choices are logically made based
on probabilities and uncertain consequences.

3
HOW DOES DECISION THEORY WORK?

 Decision theory relates with how activities leading to decision

making are understood. This theory is sometimes studies in relation
to the game theory. There are diverse types of decision-making
processes, hence, decision theory also covers diverse areas.
 There are three main areas of the decision theory, they are:

1. Prescriptive decision theory: this theory aims to provide

recommendations on how to make the best decisions even in an
uncertain decision-making environment.
2. Descriptive decision theory: this theory seeks to describe how
irrational agents make certain decisions.
3. Normative decision theory: based on a set of values, this theory
provides direction for how decisions can be made.

4
STEPS IN DECISION THEORY:

 Decision making is the process of making choices by identifying a

decision, gathering information, and assessing alternative
resolutions.
 Using a step-by-step decision-making process can help you make
more deliberate, thoughtful decisions by organizing relevant
information and defining alternatives.
 This approach increases the chances that you will choose the most
satisfying alternative possible.

5
 STEPS IN DECISION THEORY:
 Step 1: Identify the decision: You realize that you need to make a decision. Try to clearly define the nature of the decision you must make.
This first step is very important.
 Step 2: Gather relevant information: Collect some pertinent information before you make your decision: what information is needed, the
best sources of information, and how to get it. This step involves both internal and external “work.” Some information is internal: you’ll
seek it through a process of self-assessment. Other information is external: you’ll find it online, in books, from other people, and from other
sources.
 Step 3: Identify the alternatives: As you collect information, you will probably identify several possible paths of action, or alternatives.
You can also use your imagination and additional information to construct new alternatives. In this step, you will list all possible and
desirable alternatives.
 Step 4: Weigh the evidence: Draw on your information and emotions to imagine what it would be like if you carried out each of the
alternatives to the end. Evaluate whether the need identified in Step 1 would be met or resolved through the use of each alternative. As you
go through this difficult internal process, you’ll begin to favor certain alternatives: those that seem to have a higher potential for reaching
your goal. Finally, place the alternatives in a priority order, based upon your own value system.
 Step 5: Choose among alternatives: Once you have weighed all the evidence, you are ready to select the alternative that seems to be
best one for you. You may even choose a combination of alternatives. Your choice in Step 5 may very likely be the same or similar to the
alternative you placed at the top of your list at the end of Step 4.
 Step 6: Take action: You’re now ready to take some positive action by beginning to implement the alternative you chose in Step 5.
 Step 7: Review your decision & its consequences: In this final step, consider the results of your decision and evaluate whether or not it
has resolved the need you identified in Step 1. If the decision has not met the identified need, you may want to repeat certain steps of the
process to make a new decision. For example, you might want to gather more detailed or somewhat different information or explore
additional alternatives. 6
TYPES OF DECISION-MAKING
ENVIRONMENT:
 We can experience several times in decision-making where we don't
have the necessary information to decide and keep hesitating. It's
often a decision that we have a lot of data with the circumstances
and are very specific.
 There are three types of settings for decision making that we
can define.
1. Decision Making in Certain Conditions.
2. Decision Making in Risky Conditions.
3. Decision Making in Uncertain Conditions.

7
1. DECISION MAKING IN CERTAIN
CONDITIONS
 Decision-making under such circumstances ensures that the person
who makes a decision has all the complete and appropriate
knowledge for the decision to be made. With all the data available,
the individual can predict the outcome of the decision.
 We can easily create a particular decision with confidence by being
able to predict the result. Typically, the product that gives the best
outcome will be used and carried out.

Example: Profit Alternatives

90 Government Treasury Bond
100 Saving Account
80 Insurance

Solution:
The decision: Saving Account
8
2. DECISION MAKING IN RISKY
CONDITIONS
 When a manager lacks perfect information or whenever an
information asymmetry exists, risk arises.
 Under a state of risk, the decision maker has incomplete information
about available alternatives but has a good idea of the probability
of outcomes for each alternative.
 While making decisions under a state of risk, managers must
determine the probability associated with each alternative on the
basis of the available information and his experience.
 In this case two criteria are used:

1. Determine the expected monetary value (EMV)

2. Determine the expected opportunity loss (EOL)

9
DECISION MAKING UNDER RISK

EXAMPLE:
 Khaldun, hospital administrator for Amman General hospital, is trying to build a (1)large
wing onto the existing hospital, (2)a small wing, or (3)no wing at all. If the population of
Amman continues to grow a large wing could return 300,000 J.D to the hospital each year.
If the small wing were built, it would return 120,000 J.D to the hospital each year. If the
population continues to grow. If the population of Amman remain the same, the hospital,
would encounter a loss of 170,000 J.D, if the large wing were built. Farther more, a loss of
90,000 J.D Would be realized if the small wing were constructed and the population
remains the same. The probability that the population continues to grow is 0,7 , and the
probability that the population remains the same is 0,3 .
1. What type of decision problem is this ?
2. Construct a decision table.
3. Determine the expected monetary value (EMV). 10

4. Determine the expected opportunity loss (EOL).

 ‫‪DECISION MAKING UNDER RISK‬‬
‫‪EXAMPLE:‬‬
‫‪ ‬يحاول خلدون‪ ،‬مدير مستشفى عمان العام‪ ،‬بناء جناح كبير وجناح صغير في المستشفى او عدم عمل أي شئ‪ ،‬إذا استمر‬
‫عدد س كان عمان ف ي النم و فق د يكون عائ د الجناح الك بير ‪ 300000‬دينار أردن ي إل ى المس تشفى ك ل عام‪ ،‬وإذا م ا ت م بناء‬
‫الجناح الصغير ‪ ،‬فسيكون العائد ‪ 120‬ألف دينار أردني إلى المستشفى كل عام‪ .‬أما إذا بقي عدد سكان عمان كما هو فإن‬
‫المستشفى سيواجه خسارة قدرها ‪ 170‬ألف دينار أردني إذا تم بناء الجناح الكبير‪ ،‬عالوة على ذلك يمكن تحقيق خسارة‬
‫قدرها ‪ 90‬ألف دينار أردني إذا تم بناء الجناح الصغير وظل عدد السكان كما هو‪.‬‬
‫‪ ‬علما بأن احتمال استمرار النمو السكاني هو ‪ ،7.0‬واحتمال بقاء السكان كما هو ‪ ،3.0‬المطلوب ما يلي‪:‬‬
‫‪ .1‬تحديد نوع مشكلة القرار؟‬
‫‪ .2‬بناء جدول القرار؟‬
‫‪ .3‬تحديد القيمة النقدية المتوقعة(‪ )EMV‬؟‬
‫‪ ‬الحظ أن حاالت الطبيعة ‪ State Of Nature‬مرتبطة باالحتماالت‬ ‫‪ .4‬تحديد خسارة الفرصة المتوقعة(‪ )EOL‬؟‬
‫المعطاة في التمرين‪.‬‬
‫‪ ‬وبالتالي فإن باقي األرقام تتعلق بالبدائل‪.‬‬
‫‪11‬‬
‫‪DECISION MAKING UNDER RISK‬‬

‫‪ Solution‬‬ ‫‪ ‬الحل‬

‫? ‪ 1/ What type of decision problem is this‬‬ ‫‪ ‬تحديد نوع مشكلة القرار؟‬
‫‪ Decision Making under risk.‬‬ ‫‪ ‬نـجد أ ن خلدون مدي ر المس تشفى عل ى عل م باحتمال حدوث ك ل‬
‫حال ة م ن الحاالت المتوقع ة‪ ،‬وبهذه الطريق ة تكون المشكل ة‬
‫مخاطرة‬

‫‪12‬‬
DECISION MAKING UNDER RISK

 Solution ‫ الحل‬
 2/ Construct a decision table. ‫ بناء جدول القرار‬

State of nature
Alternatives
Expected Monetary Value (EMV) Population remain Population
the same grow

= (300,000 × 0,7) + ( -170,000 × 0,3) = 159,000 -170,000 300,000 Build large wing
= (120,000 × 0,7) + ( -90,000 × 0,3) = 57,000 -90,000 120,000 Build small wing
0 0 Do nothing
0,3 0,7 Probability

Solution: ‫ لذلك سيكون البديل األول هو األفضل وهو بناء‬

The decision: Large wing ‫ ) ألنه يحقق أعلي‬Large wing( ‫جناح كبير‬
.‫ربح ممكن‬ 13
 DECISION MAKING UNDER RISK
 Construct a decision table. ‫ بناء جدول القرار (يتطلب عمل جدول لخسارة الفرصة لمستشفى عمان وذلك عن طريق طرح القيم‬
.) ‫الموجودة للحالة من أكبر قيمة في كل عمود من نفسها ومن القيم األخرى‬
State of nature
Population remain the same Population grow Alternatives

-170,000 300,000 Build large wing

-90,000 120,000 Build small wing
0
‫ يص بح النات ج بع د عملي ة الطرح‬
0 Do nothing
0,7
:‫على النحو التالي‬
0,3 Probability (P)
State of nature
.Expected opportunity loss (EOL) Population remain the Population grow Alternatives
same (P= 0.3) (P=0.7)
=(0 × 0,7) + (170,000 × 0,3) = 51,000 170,000 (0--170) 0 (300-300) Build large wing
=( 180,000 × 0,7) + (90,000 × 0,3) = 153,000 90,000 (0--90) 180,000 (300-120) Build small wing
= (300,000 × 0,7) + ( 0× 0,3) = 210,000 0 (0-0) 300,000 (300-0) Do nothing

The alternative with the least expected loss is chosen: ‫ لذل ك س يكون‬، ‫ يت م اختيار البدي ل الذي يحق ق أق ل خس ارة متوقع ة‬
Large wing .) Large wing( ‫البديل األول هو األفضل وهو بناء جناح كبير‬
14
 DECISION MAKING UNDER RISK
 EXAMPLE: A manager of a company wants to evaluate three alternatives to expand his production activities. These
alternatives are opening a mobile shop, opening a stationery, or opening a restaurant. This decision faces the expectation
of high, moderate or low demand (State of Nature), noting that the probability of high demand is 40% and its moderate is
35. % and a low of 25%. The director estimated the results of the alternatives, coupled with the cases of nature, as in the
following table:

Alternative State Of Nature

State State State
(HIGH) (1) (MODERATE)(2) (LOW) (3)
MOBILE SHOP 100 190 70
STATIONERY 200 100 90
RESTUARANT 300 80 100
Probability % 40 % 35 % 25

1. Determine the expected monetary value (EMV).

2. Determine the expected opportunity loss (EOL). 15

‫ نظرية وتحليل القرار‬:‫الموضوع الثاني‬.

DECISION MAKING UNDER RISK
1. Expected Monetary Value (EMV)

State Of Nature
Alternative
State (1) State (2) State (3)
EMV
MOBILE SHOP 100 190 70 = (100× 0.4)+ (190× 0.35)+(70× 0.25)= 124
STATIONERY 200 100 90 = (200× 0.4)+ (100× 0.35)+(90× 0.25)= 137.5
RESTUARANT 300 80 100 = (300× 0.4)+ (80× 0.35)+(100× 0.25)= 173
Probability (100%) % 40 % 35 % 25

Solution:
The decision: RESTUARANT ‫ إذن القرار األمثل هو اختيار المطعم ألنه يحقق أعلى قيمة‬
. ‫ ألن المصفوفة أرباح‬،‫) دوالر‬173( ‫وهي‬
‫ أما إذا كانت المصفوفة تخفيض تكاليف فلذلك نختار أقل قيمة‬
16
 ‫‪DECISION MAKING UNDER RISK‬‬ ‫ثانياً‪ :‬القرار في حالة المخاطرة‬
‫‪ -2 ‬حساب خسارة الفرصة المتوقعة ((‪:EOL‬‬
‫‪.Expected opportunity loss (EOL) .2‬‬ ‫‪ ‬يتطل ب عم ل جدول لخس ارة الفرص ة‪ ،‬وذل ك ع ن‬
‫طري ق طرح ك ل القي م الموجودة للحال ة م ن أك بر‬
‫قيمة موجودة في نفس الحالة (طرح القيم الموجودة‬
‫‪State Of Nature‬‬ ‫ف ي العمود م ن أك بر قيم ة بالعمود)‪ ،‬وبالتال ي‬
‫‪Alternative‬‬
‫)‪State (1‬‬ ‫)‪State (2‬‬ ‫)‪State (3‬‬ ‫سنحصل على جدول خسارة الفرصة كالتالي‪:‬‬
‫‪MOBILE SHOP‬‬ ‫‪100‬‬ ‫‪190‬‬ ‫‪70‬‬
‫‪STATIONERY‬‬ ‫‪200‬‬ ‫‪100‬‬ ‫‪90‬‬
‫‪RESTUARANT‬‬ ‫‪300‬‬ ‫‪80‬‬ ‫‪100‬‬
‫)‪Probability (100%‬‬ ‫‪% 40‬‬ ‫‪% 35‬‬ ‫‪% 25‬‬

‫‪State Of Nature‬‬
‫‪Alternative‬‬ ‫‪EOL‬‬
‫)‪State (1‬‬ ‫)‪State (2‬‬ ‫)‪State (3‬‬
‫‪MOBILE SHOP‬‬ ‫)‪200 (300-100‬‬ ‫‪0‬‬ ‫)‪(190-190‬‬ ‫)‪30 (100-70‬‬ ‫‪= (200× 0.4)+ (0× 0.35) +(30× 0.25)= 87.5‬‬
‫‪STATIONERY‬‬ ‫)‪100 (300-200‬‬ ‫)‪90 (190-100‬‬ ‫)‪10 (100-90‬‬ ‫‪= (100× 0.4)+ (90× 0.35) +(10× 0.25)= 74‬‬
‫‪RESTUARANT‬‬ ‫)‪0 (300-300‬‬ ‫)‪110 (190-80‬‬ ‫)‪0 (100-100‬‬ ‫‪= (0× 0.4) + (110× 0.35)+(0× 0.25) = 38.5‬‬
‫‪Probability‬‬ ‫‪%40‬‬ ‫‪%35‬‬ ‫‪%25‬‬ ‫‪17‬‬

‫‪Solution:‬‬
‫‪The decision: RESTUARANT‬‬ ‫‪ ‬اذاً القرار األمثل هو اقل قيمة واقل ندم يكون عندنا هو فتح المطعم = ‪5,38‬‬
3. DECISION MAKING IN UNCERTAIN
CONDITIONS
 When you are unaware of the situation, making a decision is similar
to the absence of information to help us decide. The decision-maker
doesn't know the future because of inadequate knowledge and can't
predict the outcome of any choice he has.
 So before making an important decision, all the knowledge and
alternatives available must be studied. The decision-making process
will help a great deal. The atmosphere in which they are made is
another aspect that impacts these decisions. In which these choices
are made, there are a few different types of environments.
 In this case criteria are used:

1. Maximax (Optimistic Decision)

2. Maximin (Pessimistic Decision)
3. Equally Likely (Laplace)
4. Criterion of Realism (Hurwicz)
5. Minimax
18
3. DECISION MAKING IN UNCERTAIN
CONDITIONS
 Returning to the previous example, add the following:
Unfortunately, khaldoun does not have any information about
future population.
State of nature Alternative
Population remain the Population
same grow

-170,000 300,000 Build large wing

-90,000 120,000 Build small wing
0 0 Do nothing
0,3 0,7 probability

Find the following:

1. Maximax (Optimistic Decision)
2. Maximin (Pessimistic Decision)
3. Equally Likely (Laplace)
4. Criterion of Realism (Hurwicz) 19
5. Minimax
 ‫ في هذه الطريقة نختار أعلى النتائج لكل بديل‬:)Maximax( ‫ى‬Z‫ى األعل‬Z‫ معيار أعل‬
Maximax: Used to find alternative that
maximizes the maximum payoff: ، ‫ وم ن ث م نختار البدي ل الذي يقابل ه أعل ى نات ج‬Maximax ‫ونضعه ا ف ي عمود‬
)‫وهو جناح كبير (ويسمى بالمتفائل‬
 Locate the maximum payoff for each
alternative. ‫ ث م يت م‬Maximax ‫ أ ي يت م اختيار اك بر رق م ف ي ك ل ص ف ووضع ه ف ي عمود‬
. Maximax ‫اختيار اكبر رقم في عمود‬
 Select the alternative with maximum number.
2 1

State of nature
Alternative
Maximax Row Population remain Population grow
(Optimistic) the same

300,000 -170,000 300,000 Build large wing

120,000 -90,000 120,000 Build small wing
0 0 0 Do nothing

Decision: Build large wing 20

Maximin: Used to find alternative that maximizes the ‫ ف ي هذه الطريق ة نختار أق ل النتائ ج لك ل‬:)Maximin( ‫ل‬ZZ‫ى األق‬ZZ‫ معيار أعل‬
minimum payoff: ‫ وم ن ث م نختار البدي ل الذي بقابل ه أعل ى‬Maximin ‫بدي ل ونضعه ا ف ي عمود‬
 Locate the minimum payoff for each alternative. .)‫ وهو عمل ال شيء (ويسمى بالمتشائم‬، ‫ناتج‬
 Select the alternative with maximum number. ‫ ث م يت م‬Maximin ‫ أ ي يت م اختيار أق ل رق م ف ي ك ل ص ف ووضع ه ف ي عمود‬
. Maximin ‫اختيار اكبر رقم في عمود‬

2 1

State of nature
Maximin Row Alternative
Population remain Population grow
(Pessimistic) the same

-170,000 -170,000 300,000 Build large wing

-90,000 -90,000 120,000 Build small wing
0 0 0 Do nothing

Decision: Do nothing
21
Equally Likely (Laplace): Considers all the ‫ يتم حساب هذه الطريقة عن‬:)Laplace( ‫و االحتماالت المتساوية‬Z‫ معيار البالس أ‬
payoff for each alternative: ‫ وه و بناء‬، ‫ وم ن ث م نختار البدي ل الذي يقابل ه أعل ى نات ج‬،‫طري ق إيجاد المتوس ط لك ل بدي ل‬
.‫جناح كبير‬
 Find the average payoff for each alternative.
‫ ثم‬Laplace ‫ عدده م) ووضعه في عمود‬/ ‫ أي يت م احتساب متوسط كل ص ف (مجموعهم‬
 Select the alternative with the highest average. . Laplace ‫يتم اختيار اكبر رقم في عمود‬

State of nature
Alternative
Laplace (Equally Likely)Row Population remain Population
the same grow

= (300.000 + (-170.000)) / 2 = 65,000 -170,000 300,000 Build large wing

= (120.000+ (-90.000)) /2=15,000 -90,000 120,000 Build small wing
= (0+0)/2=0 0 0 Do nothing

Decision: Build large wing

22
1. Criterion of Realism (Hurwicz):
A weighted average compromise between optimistic and pessimistic: .)Criterion of Realism/ Hurwicz(‫ هورويز‬/ ‫ معيار الواقعية‬
 Select a coefficient of realism (ᵅ).
 Coefficient is between (0 and 1). )ᵅ( ‫أي يتم احتساب نتيجة كل صف (من خالل المعادلة‬ •
 Compute the weighted average for each alternative. Minimum in( )1-ᵅ( +)Maximum in Row(
 Select the alternative with the highest value. ‫ ثم يتم اختيار‬Realism ‫) ) ووضعه في عمود‬Row
 Criterion of Realism= (ᵅ) (Maximum in Row)+(1-ᵅ)(Minimum in Row( . Realism ‫اكبر رقم في عمود‬
 Using Hurwicz (α = 0,9)
‫ اذا قيمة‬،0.9 ‫بافتراض أن مدير المستشفى حدد معيار الواقعية بـ‬ •
.0.10 ‫ تساوي‬α -1 ‫ وقيمة‬0.90 ‫ تساوي‬α

State of nature
Criterion of Realism Alternative
(Hurwicz) Population remain Population
the same grow
= (ᵅ) (Maximum in Row)+ (1-ᵅ) (Minimum in Row)

= 0.90× 300.000 + 0.10 × -170.000= 253,000 -170,000 300,000 Build large wing
= 0.90× 120.000 + 0.10 × -90.000= 99,000 -90,000 120,000 Build small wing
= 0.90× 0 + 0.10 × 0 = 0 0 0 Do nothing

Decision: Build large wing ‫ في اعلي رقم في‬α ‫ اذا كانت هناك ثالث بدائل يتم ضرب قيمة‬ 23
.‫ في اقل رقم بالصف لكل بديل‬α -1 ‫ ويتم ضرب قيمة‬،‫الصف‬
 1. Minimax: Based on opportunity loos regret, the difference between the :)Minimax regret( ‫ معيار أقل األعلى‬
optimal profit and actual payoff for decision: ‫) لذلك يتطلب عمل جدول لخسارة‬EOL( ‫ يعتمد هذا المعيار على خسارة الفرصة المتوقعة‬
 Create an opportunity loss table by determining the opportunity loos for not
.)‫الفرصة (وذلك عن طريق طرح القيم الموجودة لكل حالة من أكبر قيمة في كل عمود‬
choosing the best alternative.
 Opportunity loss is calculated by subtracting each payoff in the column. ‫ ثم اختيار البديل الذي يقابله أقل‬، ‫ ومن ثم إيجاد اكبر خسارة فرصة لكل بديل من البدائل‬
 Faid the minimax opportunity loss for each alternative and pick the .‫ وهو بناء جناح كبير‬، ‫ناتج‬
alternative with the minimax number.
‫ ث م يت م‬Minimax regret ‫ أ ي يت م اختيار أك بر رق م ف ي ك ل ص ف ووضع ه ف ي عمود‬
. Minimax regret ‫اختيار أقل رقم في عمود‬

State of nature
Alternative
Population remain the same Population grow

-170,000 300,000 Build large wing

-90,000 120,000 Build small wing
2 1
0 0 Do nothing

State of nature
Alternative
Minimax regret Population remain the Population grow
same

170,000 170,000 (0--170) 0 (300-300) Build large wing

180,000 90,000 (0--90) 180,000 (300-120) Build small wing
24
300,000 0 (0-0) 300,000 (300-0) Do nothing
Decision: Build large wing
  SUMMARY
State of nature
Alternative
Criterion of Realism Laplace Maximin Maximax Population Populatio
Row Row Row remain the n grow
same
= 0.90× 300.000 + 0.10 × - = (300.000 + (-170.000)) / 2 -170,000 300,000 -170,000 300,000 Build large wing
170.000= 253,000 = 65,000

= 0.90× 120.000 + 0.10 × - = (120.000+ (-90.000)) -90,000 120,000 -90,000 120,000 Build small wing
90.000= 99,000 /2=15,000

= 0.90× 0 + 0.10 × 0 = 0 = (0+0)/2=0 0 0 0 0 Do nothing

State of nature
Alternative
Minimax regret Population remain Population grow
the same

170,000 170,000 (0--170) 0 (300-300) Build large wing

180,000 90,000 (0--90) 180,000 (300-120) Build small wing
300,000 0 (0-0) 300,000 (300-0) Do nothing
25

‫ نظرية وتحليل القرار‬:‫الموضوع الثاني‬.

 EXAMPLE

Find the following:

1. Maximax (Optimistic Decision)
2. Maximin (Pessimistic Decision)
3. Equally Likely (Laplace)
4. Criterion of Realism (Hurwicz) Using Hurwicz (α = 0.8)
5. Minimax

Alternative State Of Nature

State (1) State (2)
A 100 190
B 200 100
C 300 80
Probability 0,7 0.3
26
 EXAMPLE

Find the following:

1. Maximax (Optimistic Decision)
2. Maximin (Pessimistic Decision)
3. Equally Likely (Laplace)
4. Criterion of Realism (Hurwicz) Using Hurwicz (α = 0.7)
5. Minimax

Alternative State Of Nature

State (1) State State
(2) (3)
A 100 190 70
B 200 100 90
C 300 80 100
Probability 0.5 0.25 0.25 27
THANKS 28