Page 1 of 22

SUBJECT: DECISION SCIENCE MODELS (DSM)

Unit-I Input-output Analysis

INPUT-OUT PUT ANALYSIS:

Meaning: The input-output analysis is a technique of studying the production structure of an economy,
considering the mutual interdependence of the various production sectors.

The input-output analysis, thus, seeks to analyze inter-industry relationship in order to understand the
interdependences and complexities of the system and to find the conditions for maintaining balance between
demand and supply of each industry. The input-output analysis is also known as inter-industry flow
analysis.

We know that output of each interacting industry is consumed in two ways:

i. Intermediate demand: where output of an industry is used as input a) by the producing

industry itself; and b) by other industries in the economy
ii. Final Demand: Consisting of consumption demand by households, government and export
sector.

The input-output technique provides valuable help in economic forecasting and planning both at the national
as well as firm levels. It helps the company executives in making necessary and timely adjustments in their
production schedules.

Objects of Input-Output Analysis:

1 To reveal the nature of inter-industry dependence

2 To ensure consistency among various industries so that there no bottlenecks or unused stocks.
3 To use it as a forecasting device
4 To have detailed quantitative account of underlying changes in the economy.

Main Features Input-Output Analysis:

1 The input-output analysis handles a problem of purely technological nature.

2 The basis thrust of input-output analysis is empirical investigation rather than theoretical finesse.
3 The input-output incorporates the nature of interdependence between various sectors/industries in
an economy.
 Page 2 of 22

Assumptions of Input-Output Analysis:

The Input-Output Analysis is based on a number of assumptions. These are

1. No two products are produced the jointly.

2. In a productive process, all inputs are employed in rigidly fixed proportion.
3. Factors and commodity prices are given.
4. Amount of factor services as well as the nature and extent of consumer demand is given.
5. Industries do not enjoy the external economies or diseconomies.
6. Firms enjoy constant returns to scale.
7. There is a pure competition in producing sector.
8. Both inputs and outputs are expressed in monetary unites.
9. Production relations are linear.
10. Labour is the only primary inputs.

Types of Input-Output Model:There are two types of Input-Output Models:

Type I: Open Input-Output Model [where, total output is consumed by Intermediate demand and Final
Demand]

Type I: Closed Input-Output Model[total output is consumed by Intermediate demand (Final Demand
equals zero)]Open Input-Output Model is generally preferred.

Components of Input-Output Model:There are three Components of an Input-Output Model:

1. Input-output transaction table ; consisting of information on inter-industry lows, final demand

and primary inputs
2. Technological Coefficient matrix; stating proportionally between the output and individual inputs
of a sector.
3. The set of balance equations ;
 Page 3 of 22

HAWKINS-SIMON CONDITION OF INPUT-OUTPUT ANALYSIS:

Hawkin - Simon condition have been developed to ensure that the solution of the input-output model does
not contain negative output. Negative output means that more quantity is used as input than what is got as
output.

For Example:If more than one quintal of seed is required to produce one quintal of wheat, then we get
negative output, clearly such a situation is economically non-viable. Hawkin-Simon conditions are
designed to ensure that the system is economically viable.Hawkin-Simon prescribes two conditions for non-
negative output. These are:

Condition 1. The value of |I – A| must always be positive.

Condition 2.The principle diagonal element of [I – A] matrix should all be positive. It implies
that one unit of output of any sector should not use more than 1 unit of its own output as
Where I = identity matrix ,A = technological Matrix , [I-A] = Leontief Matrix

Before solving an Input-Output problem it should be ensured that the system fulfills Hawkin-Simon
conditions.

INPUT – OUTPUT ANALYSIS IS DONE IN FOUR STEPS:

Step 1: construction of input – output table

Step 2: Calculating technical coefficients

Step 3: Building up the set of simultaneous equations

Step 4: Solving the set of simultaneous equations using relationship X=[I-A]^-1. D

USES OF INPUT – OUTPUT ANALYSIS

Input-Output technique has become very powerful tool of economic analysis. It is considered useful at the
level of a firm as well as an economy. Some of its main uses are:

1. The input–output table is helpful to a producer as it provides him the information about the nature
and the amount of goods which are translated between him and the other producers.

2. The interrelation between various sectors, as revealed in the input- output table, provides indication
regarding prospective trends in which they are likely to combine with each other.
 Page 4 of 22

3. Input–output analysis also proves quite helpful in forecasting.

4. Input- Output analysis is also quite helpful in providing necessary information for formulation of
economic policies.

5. It is useful for national income accounting as it provides detailed breakdown of the micro-
aggregates and the money flow.

6. It is also helpful in providing necessary information for formulation of economic policies.

LIMITATIONS OF INPUT-OUTPUT ANALYSIS

i. In the input-output analysis ‘Labour’ is the only input which is considered scarce. This is not true
in practice.

ii. The input-output analysis is based on the basic premise that the input coefficients are fixed.

iii. The assumption of fixed coefficients of production ignores any possibility of factor substitution.

iv. Another problem relates to the way of ‘final demand’ is taken in the input-output analysis. Final
demand is taken as given and treated as independent of the production sector.

v. Input – output analysis is based on linear equations relating outputs of one industry to inputs of the
others. This appears unrealistic.

vi. Input – output analysis also ignores price changes.

vii. The input- - output model is technical in nature. It tells us how much outputs are possible, given
inputs. It does not tell us why the inputs and outputs are of a particular pattern in economy.
 Page 5 of 22

Unit-II:
What Game of
is theory Theory
games (or Game Theory)?

What is Game Theory? Explain with suitable example and write its characteristics?

The theory of games (or Game Theory) is a mathematical theory that deals with the general features of
competitive situations. This theory is helpful when two or more individuals or organizations with conflicting
objectives try to make decisions. In such situations, a decision made by one decision-maker affects the
decision made by one or more remaining decision-makers and final outcome depends upon the decision of
all parties.It deals with human process in which an individual decision making unit who can be individual,
a group, a formal or informal organization, society, is not complete control of other decision-making unit,
the opponents, and is addressed to problem involving conflict, co-operation both at various level. Theory of
games was developed by Von Neumann (called the father of game theory) in 1921.

Example:

I. Firms struggling to get their market shares

II. Launching advertising campaigns by companies marketing competing products

III. Negotiations between organizations and Unions

A game refers to a situation in which two or more players are competing. It involves the player who have
different goals and objectives and whose fates are intertwined. In a game situation, each of the players has
a set of strategies available. A strategy refers to the action to be taken by a player in a various contingencies
in playing games. There is set of outcomes each of which is result of particular choice of strategies made
by players on a given play of game, and pay-offs are accorded to each player in each of possible outcomes.

Objective

The main objective in theory of games is to determine rules of rational behavior in the game situation, in
which the outcome are dependent on action on the action of interdependent players.
 Page 6 of 22

CHARECTERISTICS OF GAMES

A competitive game has the following characteristics:

i. There is finite number of participants or competitors. If the number of participants is 2, the game is
called two-person game.

ii. Each participant has available to him a list of finite number of action. The list may not be same for
each participant.

iii. Each participant knows all the possible choices available to other but does not know which of them
is going to be chosen by them.

iv. A play is said to occur when each of the participants choose one of the course of action available to
him.

v. Every combination of courses of action determines outcomes which result in gains to the
participants. The game may be positive, negative or zero. Negative gain is called a loss.

vi. The gains (payoffs) for each play are fixed and specified in advance and are known to each player.

Game Models:

There are various type of game models. They are based on various factors like the number of players
participating, the sum of gains or losses and the number of strategies available.

1. Number of persons: if a game involves only two players, it is call two person game; if there are
more then two players, it is name n-number game. An n-person game does not imply on exactly n
number of player are involved in it. Rather it means that the participants can be classified into n
mutually exclusive groups, with all members in a group having identical interest.

2. Sum of payoff: if the sum off pay offs (gain or losses) to the player is zero, the game is called zero-
sum or constant sum game, otherwisenon zero game.

3. Number of strategies: if the number of strategies (move or choice) is finite, the game is called a
finite game; if not, it is called infinite game.

4. TWO-PERSON ZERO-SUM GAME:A two person zero sum game is the one which involves
two peopleand where the gain made by one equals the loss incurred by the other.

For an example, suppose there are two firms A and B in an area which, for long period in past, have
been selling a competing product and are now engaged in struggle for a larger share of the market. Now
 Page 7 of 22

with the total market of given size any share of the market gained by one firm must be lost by the other
and therefore the sum of gain and loss equal to zero.

Payoff:

It is the outcome of the game. Payoff (gain or loss) matrix is the table showing the amount received by
the player named at the left-hand-side after all possible plays of the game.

5. Saddle Point:A saddle point is an element in the game matrix that is both the largest in its column
and the smallest in its row. Not all game matrices have saddle points, but if they do, they will clearly
be the equilibrium strategies, since they both maximize player 1's payo, and minimize player 2'sloss.
For example, consider the game in which Player I has 3 choices of strategy and Player II has 4 and
the payoff matrix is

Player B
Player A Strategy IIIIIIIV Row
minimum
I 3 4 -1 2 -2
II 2 5 2 4 2 MAXIMIN
III -5 2 1 0 -5

Column. 3 5 2 4
maxmum
MI
NI
MA
Here MAXIMIN=MINMAX, The amount that they settle on, here 2, is known as the value of the game and
X is 2.
the element (II, III) of the matrix is a saddle point which

6. Maximum and Minimum Principle

In a two-person zero-sum game, say, we have two players A and B. If we look from A’s perspective, we
may safely assume that A wishes to maximize his gain while B would like to minimize his losses. So,
whenever A adopts a strategy to maximize his gains, B would adopt a counter strategy with the objective
of minimizing A’s gain or in other words, minimizing his own losses. The problem is to determine the best
strategy for A and B, assuming that both are acquainted with the information contained in the pay-off matrix
and that each one is not aware of the move the other is likely to take. From the discussion above, it is clear
that we are in a situation where A would go for maximizing his minimum gains while B would strive for
minimizing his maximum losses. So, A is obtaining a value called the maximin value and the corresponding
strategy is called the Maximin strategy. Similarly, B is obtaining a value called the minimax value and the
corresponding strategy is called the Minimax strategy. The payoff at the saddle point is called the value of
the game.
 Page 8 of 22

7. Fair and Unfair Game:

i. A game is said to be fair if Maximin value = minimax value = 0;
ii. Otherwise, term it as unfair. Anyway in the real world, things are seldom fair!).
iii. Actually, a game is said to be strictly determinable if Maximin value = minimax value ≠ 0.

8. Pure Strategy and Mixed Strategy:

I. Pure Strategy: it is the decision rule to always select a particular course of action. It is usually
represented by a number with which the course of action is associated.

II. Mixed Strategies:It is decision, in advance of all plays, to choose a course of action for each play in
accordance with some probability distribution. Thus, a mixed strategy is a selection among the pure
strategies with some fixed proportions or Ratio(or probabilities).

9. Optimal Strategy:
The strategy that puts the player in most preferred position irrespective of the strategy of his opponents is
called anOptimal Strategy.

10. Nonzero-sum game:

Here, a third party (the house or kitty) receives or make some payment is called the Nonzero-sum game.

Dominance Rule:
In the game some time a strategy available to a player might be found to preferable to some other strategy.
Such a strategy called Dominate the other one. This concept of domination is very usefully employed in
simplifying the game thus helps in finding solution to the game.

Example:Let’s consider a pay off matrix have two firms A and B,

B’s strategy
___________________________________
b1 b2 b3
___________________________________
a1 12 -8 -2
A’s strategy a2 6 7 3
a3 -10 -6 2
__________________________________

Consider the open to firm A. we noticed that every element of the second exceeds the corresponding element
of third row. Clearly a shall never prefer to play a 3, because, in comparison to see strategy, it shall be better
 Page 9 of 22

off by adapting a2, regardless of what strategy adopted by B. thus a3 is dominated by a2 and hence is can
be deleted.
B’s strategy
___________________________________
b1 b2 b3
___________________________________
A’s strategy a1 12 -8 -2
a2 6 7 3
__________________________________
From reduce matrix, we observe that the first column value are larger than their counterparts in the third
column. Since B would like to minimize the payoff to A. it would always prefer to choose b3 to b1. Thus
strategy b1 is dominated by strategy b3.

i) Dominance Rule for Row:If all the elements of a row (k th), are less than or equal to the
corresponding elements of any other row (r th), then kth row is dominated by the r th row.
Hence kth row can be deleted / eliminated.

ii) Dominance Rule for Column: If all the elements of a column (k th), are greater than or equal
to the corresponding elements of any other column (r th), then kthcolumn is dominated by the
r thcolumn. Hence kthcolumn can be deleted / eliminated.

4. Limitations of GameTheory:

The discussion was restricted to the two-person zero sum games. There are practically no applications of
game theory to the real world situations as of now a partial list of limitations is given below:

[1] The environment in which management decisions are made is rarely a two-person.

[2] Some of gains and losses of the opponents may not be zero.

[3] In real life situations it is very rare that both the parties will have equal information and intelligence.

[4] It is not easy to find value of game accurately.

[5] No consideration about the risk and uncertainty involved

Few Questions and Answer:

 Page 10 of 22

1. Can there be “two-person zero-sum” game which has a value of zero? If so, write the pay-
off matrix of one such game and find out the optimal strategies of the players.

Ans: Yes, a two-person zero-sum’ game can have a value of Zero. One such game given below,
where player A has two strategies and player B has two strategies available.
B’s strategy

b1 b2 Row Minimum

a1 0 5 0 MAXIMIN
A’s strategy a2 -8 7 -8

Column Maximum 0 7

MINIMA
X

Maximin= Minimax=0
The Optimal Strategies are: Player A= a1, Player B =b1
Value of game = 0

2. There is a famous Italian game called Two-Finger-Morra. This game played by two people
each of whom shows one or two fingers and simultaneously guesses the number of fingers
his opponent will show. If just one player guesses correctly, he wins an amount equal to
the sum of the fingers shown by himself and by his opponent; otherwise the game is
considered a draw. Obtain the pay-off matrix corresponding to it.

Ans: If (1,1) represent a situation that a player holds one finger and guesses that his opponent will
also show one finger;

Similarly, (1,2) indicates that a player holds one finger and his opponent will also show two
fingers;

Player B
(1, 1) (1, 2) (2, 1) (2, 2)
Player A (1, 1) 0 2 -3 0
(1, 2) -2 0 0 3
(2,1) 3 0 0 -4
(2, 2) 0 -3 4 0

Unit-III: Markov Chain Analysis

 Page 11 of 22

What is Markov Chain Analysis?

In the Markovian analysis, the analysis of a given system is based on the following two set of input data-
the transition matrix (containing the transition probabilities) and the initial condition in which the system
is. Based on these inputs, the model provides for the following predictions:

➢ The probability of the system being in a given state at a given future time.
➢ The steady state (that is, long run or equilibrium) probabilities.

We shall consider first the inputs and then the analysis and the output

Brand Switching Example:

Suppose that there are three brands of detergents D1, D2, and D3selling in the market. The market share
has been observed month-to-month for changes in the brand loyalty- that is to say, whether and how
customers change their brand of detergent over time. Let us say that the rate of brand switching has settled
over time as follows. Every month, the customers are being drawn from brand D1 to brand D2 to extent of
30%, 10% drawn to D3, and remaining 60% staying with D1 itself. However 20% of those using brand
D2 in a given month are drawn to D1, 30% shift to D3, 50% remaining stay for D2. Similarly, the behavior
with regard brand D3, 80% customers strict with Brand D3, 15% shift to brand D1, while remaining 5%
shift to brand D2. The results are shown in following figure:

D1 D1

D2
D2

D2 D2

At present the market shares for bran D1, D2, and D3, are 30%, 40%, and25 % respectively.

There are two inputs; viz. “transition probabilities” and the “Initial Conditions”. And there are two outputs;
viz. “Specific–state Probabilities” and the “Steady State Probabilities”.
 Page 12 of 22

What is Markov Processes?

As indicated earlier, the detergent brand switching example is a stochastic process and since a customer
brand choice is dependent only on his choice of the immediately preceding month and not on his choice
two, three or four months ago, this stochastic process is termed as Markov process. Further if the transition
probabilities if the customers, switching from one brand (known as a state) to another remain constant from
time to time, then the markov process is termed as stationary or homogeneous, markov chain.

INPUTS OF MARKOV CHAIN ANANLYSIS:

There are two inputs; viz. “transition probabilities” and the “Initial Conditions”.

1. Transition Probabilities

The transition probabilities are required for obtaining both type of prediction mentioned above. It may be
recalled that the Markov process describes movement of the system from a certain state in the current stage
(may be current period) to one of the n possible states in the next stage. This probability is known as the
transition probability, and expressed as Pij, being the probability that the system moves from current state
i to another state j in the next time period. The transition probabilities may be presented in the tabular or
matrix form.Example: Refer to the detergent example and construct the transition probabilities matrix

To Brand (next month)

From Brand (this month) D1 D2 D3

D1 0.60 0.30 0.10

D2 0.20 0.50 0.30
D3 0.15 0.05 0.80

Alternately, P = 0.60 0.30 0.10

0.20 0.50 0.30

0.15 0.05 0.80

It may be noted that the rows of the matrix of transition probabilities indicate theretention and loss of market
to other brands, while the columns contain the retention and gain of market.

For example, Row-Wise:Row 1 indicates that detergent D1 retain 0.60 (or sixty per cent) of its customer,
loses 0.30 of its customers to D2 and loses 0.10 to brand D3. Row 2 indicates that detergent D2 retain one-
half of its customers and loses 0.20 and 0.30 of its customers to brand D1 and D3, respectively. Similarly,
row 3 indicate a loss of, respectively, 0.15 and 0.05 of customers to D1 and D2, and retention of 0.80 of the
customers of D3.
 Page 13 of 22

Now, proceeding column-wise, it may be seen that the column 1 of the matrix indicate that detergent D1
retains 0.60 of its customers while gains 0.20 of the customers of D2 and 0.15 of the D3’s customers. Other
column values can be similarly interpreted. Notice that for every row in the matrix, the sum of the
probabilities is equal to 1. This is because, there are only three detergent brands- one may stick to the brand
used or shift to either of the two other brands. They are collectively exhaustive and therefore, the sum of
the probabilities would equal one.

2. The Initial Conditions:

The initial condition describe the situation the system presently is in. for instance, as indicated earlier, as
the market is divided 30%, 45% and 25% between the brands D1, D2 and D3 respectively. It may be
expressed in term of a row vector [0.30 0.45 0.25]. in case the initial condition is describes as [0 0 1] for the
market, it implies that the brand d3 hold the entire market while for a customer, it would imply that the
given customer has currently bought brand D3.
Q(0) =[0.30 0.45 0.25].

OUTPUTS OF MARKOV CHAIN ANANLYSIS:

There are two outputs; viz. “Specific–state Probabilities” and the “Steady State Probabilities”.

As stated earlier there are two predictions which a Markov analysis provides. The first of these is the
probabilities of the system being in a particular state at a future time

1. Specific–state Probabilities:for calculating the probabilities for the system in specific state, we let
qi(k) to represent the probability (q) of the system being in a certain period (k), called the state
probability since the system would occupy one and only one state at a given point in time. It is
obvious that the sum of all qi values would be equal to 1.

2. Steady State (long run) Probabilities:A significant property of the Markov chains is that the
process tends to stabilize in the long run. A stabilized system is said to be in a steady state or in
equilibrium, so that the system’s operating characteristics can become independent of time.

The phenomenon of equilibrium can be expressed symbolically as Q(k) = Q(k-1), so that the Probabilities
in period k are the same as in the previous period.

Assumption underlying under Markov Analysis:

The markov chain analysis to follow is based on the following Assumptions:

 Page 14 of 22

1) Finite States: The given system has a finite number of states, none of which is “absorbing” in nature.
In our detergent example, there are three states – three brands of detergent which the customer can
switch between they are all non-absorbing.

2) First – order Process: The condition of the system in any given period is dependent only on its
condition prevailing in the previous period and the transition probabilities. In our brand switching case,
it is assumed that the choice of a particular brand of detergent is dependent upon and influenced only
by the choice in the previous month.

3) Stationarity: The transition probabilities are constant over time. As already indicated, it is assumed
that the system has settled down so that the switching among different brands takes place at the given
rates in each time period.

4) Uniform Time Periods: The changes from one state to another take place only once during each time
period, and the time periods are equal in duration. Thus, in our example, it is assumed that the customers
change their brands of detergent on a monthly basis and accordingly.

Other terminology:

Stochastic process:
In probability theory and related fields, a stochastic or random process is a mathematical object usually
defined as a collection of random. The study of how random variable involves over a period of times
includes stochastic process.

Absorbing State:
In our detergent example, if there was one brand of detergent from which a customer would never switch,
then that brand would have represented an absorbing state.

Recurrent State:
A state Iis known as a transient state if there exists a state J which is reachable from I, but I is not reachable
from state J. Thus, a state I is transient if there is a way to leave state I that never returns to state I. If a state
is not transient, it is known as a recurrent state.
Unit-IV: Decision Theory

Decision Theory:
 Page 15 of 22

Decision Theory Provides a rational approach in dealing such problems confronted with partial, imperfect
or uncertain future conditions.

Steps in decision Theory Approach (Process) :

The decision Theory approach generally involves four Steps:

Step1:identification of the various possible outcomes (States of Nature).

Step2:identification of all Courses of Actions (Strategies).
Step3: Construct a payoff table.
Step4: Choose the criteria that result in the largest pay off.

Decision -Making Environments:

Decisions are made under four types of environments

1. Decision –making under conditions of Certainty
2. Decision –making under conditions of un-certainty
3. Decision –making under conditions of Risk
4. Decision –making under conditions of Conflict

What is Decision Rule? Explain the decision rules under conditions of uncertainty.

DECISION RULE: There are several rules or criteria, on the basis of which decision may be taken. The
selection of an appropriate criterion depends on factor like the nature of decision situation, attitude of
decision maker and so on.

Decision Rules under Conditions of Uncertainty:

The decision situation where there is no way in which maker can assess the probabilities of the various
states of nature is called decision under uncertainty. The several principles employed for taking decision
under such condition.

a) Laplace principle (Criteria of Rationality):The Laplace principle is based on the simple

philosophy that we are uncertain about the various events then we may treat them equally probable.
Under this, the expected value of pay-off for each strategy is determined and the strategy with the
highest mean value is adopted.
 Page 16 of 22

b) Maximin or minimax principle (Pessimistic or Wald): This principle is adopted by pessimistic

decision maker who are conservative in their approach. Under this, the minimum pay-offs resulting
from adoption of various strategies are considered and among these values the maximum one is
selected. When dealing with cost, the maximum cost associated with each alternative is considered
and the alternative which minimizes this maximum cost is chosen.

c) Maximax or Minimin principle (Optimistic): The maximax principle is optimist principle of

choice. It suggests that, the maximum profit should be considered and the strategy with which the
highest value is associated should be chosen. In minimin principle, the minimum cost for each
alternative is considered and then the alternative which minimizes the minimum cost is selected.

d) Hurwicz principle (Criterion of Realism): The hurwicz principle of decision making stipulates
that a decision maker’s view may fall somewhere between the extreme pessimism of the maximin
principle and the extreme optimism of the maximax principle.

e) Savage principle (Minimax Regret Principle): The savage principle is based on the concept of
regret and calls for select if the course of action that minimizes the maximum regret. It is also known
as the principle of minimax regret.

DECISION UNDER RISK

The decision situations where in the decision maker chooses to consider several possible outcomes and the
probabilities of their occurrence can be stated are called decisions under risk.

a) Maximum likelihood principle (MLP): Under this, the decision maker first considers the event
that is most likely to occur. He then decide the course of action which has the maximum conditional
pay-offs.

b) Expectation principle: More generally, the decision making in situation of risk is on the basis of
the expectation principle, with the event probabilities assigned, objectively or subjectively, the
expected pay-off for each strategy is calculated by multiplying the pay-off values with their
respective probabilities and then adding up these products. The strategy with the highest expected
pay-offs represent the optimal choice. Also known as Expected Monetary Value (EMV) criterion.

EMV: (Expected Monetary Value):

 Page 17 of 22

The criteria requires the calculation of the expected values of each decision alternative which is the sum of
weighted payoffs for that alternative, where the weighted are assigned to the states of the nature that can
happen. Also known as EMV. It consists of the following steps:

Step 1: Construct a conditional pay-off table

Step 2: Calculate the EMV

Step 3: Select the alternative that gives the highest EMV.

EOL (Expected Opportunity Loss) Criterion:

An alternative approach (to maximizing EMV approach) is to minimize expected Opportunity Loss. EOL
represents amount by which maximum possible profit will be reduced under various possible stock
action.The procedure to calculate EOL is as follows:

Step 1: Prepare the conditional profit table for each decision-event combination and write the associate
probabilities.

Step 2: For each event determine the conditional opportunity loss (COL) by subtracting the payoff from
highest payoff

Step 3: Calculate the expected Opportunity Loss (EOL) by multiplying COL with associate probability

Step 4: Select the alternative that gives lowest EOL

Expected Profit of perfect Information (EPPI):

EPPI = probabilityX payoff under the perfect information

Expected Value of perfect Information (EVPI):

EVPI = EPPI – Max Expected EMV

DECISION TREE:
 Page 18 of 22

A decision tree is a graphical representation of the decision process indicating decision alternatives, states
of nature, probabilities attached to statues of nature, and conditional benefits and losses. It consists of a
network of nodesand branches. Two types of nodes are used

[I] Decision Nodes: It is represented by “Square”. =>

[II] Chance Nodes: It is represented by “Circle”. =>

Alternative courses of action (Strategies) are originated from the decision node as main braches (decision
branches). At the end of each decision branch, there is a chance node. The probabilities associated with the
chance events are shown alongside these branches. At the terminal of the chance branches are shown the
values of outcome (Payoffs).

ROLLBACK or FOLD BACK PROCESS:

The general approach used in decision tree analysis is to work ‘backward’ through the tree from right to
left,computing the expected value of each chance node. We than, choose the particular branch leaving a
decision node which leads to the chance node with the highest expected values. This is known as roll back
or “fold back process”.

STEPS IN DECISION TREE ANALYSIS:

Step 1: Identify the decision points and alternative courses of action

Step 2: At each decision points calculate the probability and payoff associated with each course of action

Step 3: Commencing from extreme right end calculate EMV

Step 4: Choose course of action that gives best payoff for each of the decision

Step 5: Process backwards to the nest stage of decision points

Step 6: Repeat the above process till the next decision point is reached

Step 7: Finally, identify the courses of action to be adapted from beginning to the end

Advantages and Limitations of Decision Tree approach:

 Page 19 of 22

Advantages:

1. It structures the decision process and helps decision-making in an orderly, systematic, and
sequential manner.

2. It requires the decision maker to examine all possible outcomes, whether desirable or undesirable.

3. It communicates the decision making process to others in an easy and clear manner. Illustrating
each assumption about the future.

4. It displays the logical relationship between the parts of a complex decision.

5. It is especially useful in situations where in the initial decision and an outcome affects the
subsequent decisions.

6. It can be applied in various filed such as introductionof new product, marketing, make or buy
decisions, investment decisions, etc.

Disadvantages (or Limitations):

1. Decision tree diagrams become more complicated as the number of decision alternatives increases.

2. It becomes highly complicated when interdependent alternatives and dependent variables are
present in the problem.

3. It assumes the utility of money in linear with money.

4. It analyses the problem in terms of expected values and thus yield as average valued solution.

5. There is often inconsistency in assigning probabilities for different events.

Unit-V: Simulation
 Page 20 of 22

Simulation:

Definition: Simulation is a representation of reality through the use of a model or other device which will
react in the same manner as reality under a given set of conditions. Simulation has also been defined as” the
use of a system model that has the designed characteristics of reality in order to produce the essence of
actual operation”.

Example:

1. A children cycling park, with various crossings and signals, is a simulated model of the city traffic
system.

2. Planetarium shows represent a beautiful simulation of the planet system

Monte Carlo Simulation:

The Monte Carlo method of Simulation was developed by two mathematicians J.V. Neumann and S.Ulam.
The technique employs random numbers and is used to solve problems that involve probability and wherein
physical experimentation is impracticable and formulation of mathematical model is impossible. It is a
method of simulation by sampling technique. The steps involved in carrying out Monte Carlo Simulation
are:

Step1:Select the measure of effectiveness (objective function) of the problem. It is either maximized or
minimized.

Step 2:Identify the variables that affect the measure of effectiveness significantly.

Step 3:Determine the cumulative probability distribution of each variable selected in step2

Step 4: Get a set of random numbers

Step 5: Consider each random number as a decimal value of the cumulative probability distribution.

Step 6:Record the value generated in step 5. Substitute in the formula chosen for measure of effectiveness
and find it simulated value.

Step 7:Repeat steps 5 and 6until sample is large enough to the satisfaction of the decision maker.

Advantages of Simulation:
 Page 21 of 22

1) Simulation offers to solution to offers the solution by allowing experimentation with model.

2) Estimates the need of costly trial and error methods

3) This help in getting proposed plans accepted and implemented.

4) Simulation models are comparatively flexible.

5) Simulation models are easier to use comparatively mathematical models.

6) Capacity to lend itself to problem that is impossible to handle mathematically using analytical

methods.

7) Used for training the operation and managerial staff in the operation of complex plans.

8) Technique allows the analyst to experiment with the system behavior.

9) It also compresses time to enable the managers visualize the long-term effects in a quick manner.

10) Simulation is often used to test proposed analytical solution as well.

Disadvantages of Simulation:

 It does not represent a methodology for derivation of optimum solutions to the given problems.

 This approach is designed merely to provide a characteristic of the behavior of the system in

question for a given set of inputs.

 It may not prove economical.

 It is a tool of solution evaluation and thus does not generate problem solution.

Application of Simulation:
 Page 22 of 22

 Shop floor management

 Scheduling and production process

 Location of emergency vehicles

 Making inventory policy decisions

 Portfolio selection and capital budgeting

 Design of queuing system and In biomedical system

 In the study of projects involving the risk investments

Programming Languages and Simulation Languages:

Programming Languages:

• FORTRAN
• GASP
• SIMSCRIPT
• GPSS

Simulation Languages:

• DYNAMO
• SIMPAC
• SIMULATE
• SIMULA
• GSMP
• ESP
• CSL

****