Page 1 of 22 Page 2 of 22

SUBJECT: DECISION SCIENCE MODELS (DSM) Assumptions of Input-Output Analysis:

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

1. No two products are produced the jointly.

INPUT-OUT PUT ANALYSIS:
2. In a productive process, all inputs are employed in rigidly fixed proportion.
Meaning: The input-output analysis is a technique of studying the production structure of an economy, 3. Factors and commodity prices are given.
considering the mutual interdependence of the various production sectors. 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.
The input-output analysis, thus, seeks to analyze inter-industry relationship in order to understand the
6. Firms enjoy constant returns to scale.
interdependences and complexities of the system and to find the conditions for maintaining balance between
7. There is a pure competition in producing sector.
demand and supply of each industry. The input-output analysis is also known as inter-industry flow
8. Both inputs and outputs are expressed in monetary unites.
analysis.
9. Production relations are linear.
We know that output of each interacting industry is consumed in two ways: 10. Labour is the only primary inputs.

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
Types of Input-Output Model:There are two types of Input-Output Models:
ii. Final Demand: Consisting of consumption demand by households, government and export
sector. Type I: Open Input-Output Model [where, total output is consumed by Intermediate demand and Final
Demand]
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 Type I: Closed Input-Output Model[total output is consumed by Intermediate demand (Final Demand
production schedules. equals zero)]Open Input-Output Model is generally preferred.

Objects of Input-Output Analysis:

1 To reveal the nature of inter-industry dependence Components of Input-Output Model:There are three Components of an Input-Output Model:
2 To ensure consistency among various industries so that there no bottlenecks or unused stocks.
1. Input-output transaction table ; consisting of information on inter-industry lows, final demand
3 To use it as a forecasting device
and primary inputs
4 To have detailed quantitative account of underlying changes in the economy.
2. Technological Coefficient matrix; stating proportionally between the output and individual inputs
Main Features Input-Output Analysis: of a sector.
3. The set of balance equations ;
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 1 of 92
 Page 3 of 22 Page 4 of 22

HAWKINS-SIMON CONDITION OF INPUT-OUTPUT ANALYSIS: 3. Input–output analysis also proves quite helpful in forecasting.

Hawkin - Simon condition have been developed to ensure that the solution of the input-output model does 4. Input- Output analysis is also quite helpful in providing necessary information for formulation of
not contain negative output. Negative output means that more quantity is used as input than what is got as economic policies.
output.
5. It is useful for national income accounting as it provides detailed breakdown of the micro-
For Example:If more than one quintal of seed is required to produce one quintal of wheat, then we get aggregates and the money flow.
negative output, clearly such a situation is economically non-viable. Hawkin-Simon conditions are
6. It is also helpful in providing necessary information for formulation of economic policies.
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 LIMITATIONS OF INPUT-OUTPUT ANALYSIS
that one unit of output of any sector should not use more than 1 unit of its own output as i. In the input-output analysis ‘Labour’ is the only input which is considered scarce. This is not true
Where I = identity matrix ,A = technological Matrix , [I-A] = Leontief Matrix in practice.

ii. The input-output analysis is based on the basic premise that the input coefficients are fixed.
Before solving an Input-Output problem it should be ensured that the system fulfills Hawkin-Simon
conditions.
iii. The assumption of fixed coefficients of production ignores any possibility of factor substitution.
INPUT – OUTPUT ANALYSIS IS DONE IN FOUR STEPS:
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.
Step 1: construction of input – output table
v. Input – output analysis is based on linear equations relating outputs of one industry to inputs of the
Step 2: Calculating technical coefficients others. This appears unrealistic.

Step 3: Building up the set of simultaneous equations vi. Input – output analysis also ignores price changes.

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

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 2 of 92
 Page 5 of 22 Page 6 of 22

CHARECTERISTICS OF GAMES

A competitive game has the following characteristics:

Unit-II:
What Game of
is theory Theory
games (or Game Theory)?
i. There is finite number of participants or competitors. If the number of participants is 2, the game is
called two-person game.

What is Game Theory? Explain with suitable example and write its characteristics? ii. Each participant has available to him a list of finite number of action. The list may not be same for
each participant.
The theory of games (or Game Theory) is a mathematical theory that deals with the general features of
iii. Each participant knows all the possible choices available to other but does not know which of them
competitive situations. This theory is helpful when two or more individuals or organizations with conflicting is going to be chosen by them.
objectives try to make decisions. In such situations, a decision made by one decision-maker affects the iv. A play is said to occur when each of the participants choose one of the course of action available to
decision made by one or more remaining decision-makers and final outcome depends upon the decision of him.
all parties.It deals with human process in which an individual decision making unit who can be individual, v. Every combination of courses of action determines outcomes which result in gains to the
a group, a formal or informal organization, society, is not complete control of other decision-making unit, participants. The game may be positive, negative or zero. Negative gain is called a loss.

the opponents, and is addressed to problem involving conflict, co-operation both at various level. Theory of vi. The gains (payoffs) for each play are fixed and specified in advance and are known to each player.
games was developed by Von Neumann (called the father of game theory) in 1921.

Example:

I. Firms struggling to get their market shares Game Models:

II. Launching advertising campaigns by companies marketing competing products

There are various type of game models. They are based on various factors like the number of players
III. Negotiations between organizations and Unions 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
A game refers to a situation in which two or more players are competing. It involves the player who have more then two players, it is name n-number game. An n-person game does not imply on exactly n
different goals and objectives and whose fates are intertwined. In a game situation, each of the players has number of player are involved in it. Rather it means that the participants can be classified into n
a set of strategies available. A strategy refers to the action to be taken by a player in a various contingencies mutually exclusive groups, with all members in a group having identical interest.
in playing games. There is set of outcomes each of which is result of particular choice of strategies made
2. Sum of payoff: if the sum off pay offs (gain or losses) to the player is zero, the game is called zero-
by players on a given play of game, and pay-offs are accorded to each player in each of possible outcomes.
sum or constant sum game, otherwisenon zero game.
Objective
3. Number of strategies: if the number of strategies (move or choice) is finite, the game is called a
The main objective in theory of games is to determine rules of rational behavior in the game situation, in finite game; if not, it is called infinite game.
which the outcome are dependent on action on the action of interdependent players.
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 3 of 92
 Page 7 of 22 Page 8 of 22

with the total market of given size any share of the market gained by one firm must be lost by the other 7. Fair and Unfair Game:
and therefore the sum of gain and loss equal to zero. 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!).
Payoff:
iii. Actually, a game is said to be strictly determinable if Maximin value = minimax value ≠ 0.
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. 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
5. Saddle Point:A saddle point is an element in the game matrix that is both the largest in its column represented by a number with which the course of action is associated.
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 II. Mixed Strategies:It is decision, in advance of all plays, to choose a course of action for each play in
the payoff matrix is
accordance with some probability distribution. Thus, a mixed strategy is a selection among the pure
Player B strategies with some fixed proportions or Ratio(or probabilities).
Player A Strategy IIIIIIIV Row
minimum
I 3 4 -1 2 -2 9. Optimal Strategy:
II 2 5 2 4 2 MAXIMIN
III -5 2 1 0 -5 The strategy that puts the player in most preferred position irrespective of the strategy of his opponents is
called anOptimal Strategy.
Column. 3 5 2 4
maxmum
MI
10. Nonzero-sum game:
NI
Here, a third party (the house or kitty) receives or make some payment is called the Nonzero-sum game.
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

Dominance Rule:
In a two-person zero-sum game, say, we have two players A and B. If we look from A’s perspective, we In the game some time a strategy available to a player might be found to preferable to some other strategy.
may safely assume that A wishes to maximize his gain while B would like to minimize his losses. So, 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.
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 Example:Let’s consider a pay off matrix have two firms A and B,
B’s strategy
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 b1 b2 b3
___________________________________
that we are in a situation where A would go for maximizing his minimum gains while B would strive for a1 12 -8 -2
minimizing his maximum losses. So, A is obtaining a value called the maximin value and the corresponding A’s strategy a2 6 7 3
a3 -10 -6 2
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
Consider the open to firm A. we noticed that every element of the second exceeds the corresponding element
the game. of third row. Clearly a shall never prefer to play a 3, because, in comparison to see strategy, it shall be better

Page 4 of 92
 Page 9 of 22 Page 10 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. 1. Can there be “two-person zero-sum” game which has a value of zero? If so, write the pay-
B’s strategy off matrix of one such game and find out the optimal strategies of the players.
___________________________________
b1 b2 b3 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.
A’s strategy a1 12 -8 -2 B’s strategy
a2 6 7 3
__________________________________ b1 b2 Row Minimum
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 a1 0 5 0 MAXIMIN
A’s strategy a2 -8 7 -8
strategy b1 is dominated by strategy b3.
Column Maximum 0 7
i) Dominance Rule for Row:If all the elements of a row (k th), are less than or equal to the
MINIMA
corresponding elements of any other row (r th), then kth row is dominated by the r th row. X
Hence kth row can be deleted / eliminated.
Maximin= Minimax=0
The Optimal Strategies are: Player A= a1, Player B =b1
ii) Dominance Rule for Column: If all the elements of a column (k th), are greater than or equal Value of game = 0
to the corresponding elements of any other column (r th), then kthcolumn is dominated by the
2. There is a famous Italian game called Two-Finger-Morra. This game played by two people
r thcolumn. Hence kthcolumn can be deleted / eliminated. 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
4. Limitations of GameTheory: considered a draw. Obtain the pay-off matrix corresponding to it.

The discussion was restricted to the two-person zero sum games. There are practically no applications of Ans: If (1,1) represent a situation that a player holds one finger and guesses that his opponent will
game theory to the real world situations as of now a partial list of limitations is given below: also show one finger;

[1] The environment in which management decisions are made is rarely a two-person. Similarly, (1,2) indicates that a player holds one finger and his opponent will also show two
fingers;
[2] Some of gains and losses of the opponents may not be zero.
Player B
[3] In real life situations it is very rare that both the parties will have equal information and intelligence. (1, 1) (1, 2) (2, 1) (2, 2)
Player A (1, 1) 0 2 -3 0
[4] It is not easy to find value of game accurately. (1, 2) -2 0 0 3
(2,1) 3 0 0 -4
[5] No consideration about the risk and uncertainty involved (2, 2) 0 -3 4 0

Unit-III: Markov Chain Analysis

Few Questions and Answer:

Page 5 of 92
 Page 11 of 22 Page 12 of 22

What is Markov Processes?

What is Markov Chain Analysis?
As indicated earlier, the detergent brand switching example is a stochastic process and since a customer
In the Markovian analysis, the analysis of a given system is based on the following two set of input data-
brand choice is dependent only on his choice of the immediately preceding month and not on his choice
the transition matrix (containing the transition probabilities) and the initial condition in which the system
two, three or four months ago, this stochastic process is termed as Markov process. Further if the transition
is. Based on these inputs, the model provides for the following predictions:
probabilities if the customers, switching from one brand (known as a state) to another remain constant from
➢ The probability of the system being in a given state at a given future time. time to time, then the markov process is termed as stationary or homogeneous, markov chain.
➢ The steady state (that is, long run or equilibrium) probabilities.

We shall consider first the inputs and then the analysis and the output
INPUTS OF MARKOV CHAIN ANANLYSIS:

Brand Switching Example: There are two inputs; viz. “transition probabilities” and the “Initial Conditions”.

Suppose that there are three brands of detergents D1, D2, and D3selling in the market. The market share 1. Transition Probabilities
has been observed month-to-month for changes in the brand loyalty- that is to say, whether and how The transition probabilities are required for obtaining both type of prediction mentioned above. It may be
customers change their brand of detergent over time. Let us say that the rate of brand switching has settled recalled that the Markov process describes movement of the system from a certain state in the current stage
over time as follows. Every month, the customers are being drawn from brand D1 to brand D2 to extent of (may be current period) to one of the n possible states in the next stage. This probability is known as the
30%, 10% drawn to D3, and remaining 60% staying with D1 itself. However 20% of those using brand transition probability, and expressed as Pij, being the probability that the system moves from current state
D2 in a given month are drawn to D1, 30% shift to D3, 50% remaining stay for D2. Similarly, the behavior i to another state j in the next time period. The transition probabilities may be presented in the tabular or
with regard brand D3, 80% customers strict with Brand D3, 15% shift to brand D1, while remaining 5% matrix form.Example: Refer to the detergent example and construct the transition probabilities matrix
shift to brand D2. The results are shown in following figure:
To Brand (next month)
D1 D1
From Brand (this month) D1 D2 D3

D1 0.60 0.30 0.10

D2 D2 0.20 0.50 0.30
D2
D3 0.15 0.05 0.80

Alternately, P = 0.60 0.30 0.10

D2 D2 0.20 0.50 0.30

0.15 0.05 0.80

At present the market shares for bran D1, D2, and D3, are 30%, 40%, and25 % respectively.
It may be noted that the rows of the matrix of transition probabilities indicate theretention and loss of market
There are two inputs; viz. “transition probabilities” and the “Initial Conditions”. And there are two outputs; to other brands, while the columns contain the retention and gain of market.
viz. “Specific–state Probabilities” and the “Steady State Probabilities”.
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 6 of 92
 Page 13 of 22 Page 14 of 22

Now, proceeding column-wise, it may be seen that the column 1 of the matrix indicate that detergent D1 1) Finite States: The given system has a finite number of states, none of which is “absorbing” in nature.
retains 0.60 of its customers while gains 0.20 of the customers of D2 and 0.15 of the D3’s customers. Other In our detergent example, there are three states – three brands of detergent which the customer can
column values can be similarly interpreted. Notice that for every row in the matrix, the sum of the switch between they are all non-absorbing.
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 2) First – order Process: The condition of the system in any given period is dependent only on its
the probabilities would equal one. 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
2. The Initial Conditions:
by the choice in the previous month.
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
3) Stationarity: The transition probabilities are constant over time. As already indicated, it is assumed
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
that the system has settled down so that the switching among different brands takes place at the given
market, it implies that the brand d3 hold the entire market while for a customer, it would imply that the
rates in each time period.
given customer has currently bought brand D3.
Q(0) =[0.30 0.45 0.25].
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
OUTPUTS OF MARKOV CHAIN ANANLYSIS: change their brands of detergent on a monthly basis and accordingly.
There are two outputs; viz. “Specific–state Probabilities” and the “Steady State Probabilities”.
Other terminology:
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 Stochastic process:
In probability theory and related fields, a stochastic or random process is a mathematical object usually
1. Specific–state Probabilities:for calculating the probabilities for the system in specific state, we let
defined as a collection of random. The study of how random variable involves over a period of times
qi(k) to represent the probability (q) of the system being in a certain period (k), called the state
includes stochastic process.
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.
Absorbing State:
In our detergent example, if there was one brand of detergent from which a customer would never switch,
2. Steady State (long run) Probabilities:A significant property of the Markov chains is that the
then that brand would have represented an absorbing state.
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.
Recurrent State:
The phenomenon of equilibrium can be expressed symbolically as Q(k) = Q(k-1), so that the Probabilities A state Iis known as a transient state if there exists a state J which is reachable from I, but I is not reachable
in period k are the same as in the previous period. 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
Assumption underlying under Markov Analysis:
Decision Theory:
The markov chain analysis to follow is based on the following Assumptions:

Page 7 of 92
 Page 15 of 22 Page 16 of 22

Decision Theory Provides a rational approach in dealing such problems confronted with partial, imperfect b) Maximin or minimax principle (Pessimistic or Wald): This principle is adopted by pessimistic
or uncertain future conditions. 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
Steps in decision Theory Approach (Process) : selected. When dealing with cost, the maximum cost associated with each alternative is considered
The decision Theory approach generally involves four Steps: and the alternative which minimizes this maximum cost is chosen.

Step1:identification of the various possible outcomes (States of Nature). c) Maximax or Minimin principle (Optimistic): The maximax principle is optimist principle of
Step2:identification of all Courses of Actions (Strategies). choice. It suggests that, the maximum profit should be considered and the strategy with which the
Step3: Construct a payoff table. highest value is associated should be chosen. In minimin principle, the minimum cost for each
Step4: Choose the criteria that result in the largest pay off. alternative is considered and then the alternative which minimizes the minimum cost is selected.

Decision -Making Environments: 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
Decisions are made under four types of environments principle and the extreme optimism of the maximax principle.
1. Decision –making under conditions of Certainty
2. Decision –making under conditions of un-certainty e) Savage principle (Minimax Regret Principle): The savage principle is based on the concept of
3. Decision –making under conditions of Risk regret and calls for select if the course of action that minimizes the maximum regret. It is also known
4. Decision –making under conditions of Conflict as the principle of minimax regret.

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

DECISION RULE: There are several rules or criteria, on the basis of which decision may be taken. The The decision situations where in the decision maker chooses to consider several possible outcomes and the
selection of an appropriate criterion depends on factor like the nature of decision situation, attitude of probabilities of their occurrence can be stated are called decisions under risk.
decision maker and so on.
a) Maximum likelihood principle (MLP): Under this, the decision maker first considers the event
Decision Rules under Conditions of Uncertainty: that is most likely to occur. He then decide the course of action which has the maximum conditional
pay-offs.
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
b) Expectation principle: More generally, the decision making in situation of risk is on the basis of
under such condition.
the expectation principle, with the event probabilities assigned, objectively or subjectively, the
a) Laplace principle (Criteria of Rationality):The Laplace principle is based on the simple expected pay-off for each strategy is calculated by multiplying the pay-off values with their
philosophy that we are uncertain about the various events then we may treat them equally probable. respective probabilities and then adding up these products. The strategy with the highest expected
Under this, the expected value of pay-off for each strategy is determined and the strategy with the pay-offs represent the optimal choice. Also known as Expected Monetary Value (EMV) criterion.
highest mean value is adopted.
EMV: (Expected Monetary Value):

Page 8 of 92
 Page 17 of 22 Page 18 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
A decision tree is a graphical representation of the decision process indicating decision alternatives, states
happen. Also known as EMV. It consists of the following steps:
of nature, probabilities attached to statues of nature, and conditional benefits and losses. It consists of a
Step 1: Construct a conditional pay-off table network of nodesand branches. Two types of nodes are used

Step 2: Calculate the EMV [I] Decision Nodes: It is represented by “Square”. =>

Step 3: Select the alternative that gives the highest EMV. [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
EOL (Expected Opportunity Loss) Criterion:
chance events are shown alongside these branches. At the terminal of the chance branches are shown the
An alternative approach (to maximizing EMV approach) is to minimize expected Opportunity Loss. EOL
values of outcome (Payoffs).
represents amount by which maximum possible profit will be reduced under various possible stock
action.The procedure to calculate EOL is as follows:

ROLLBACK or FOLD BACK PROCESS:

Step 1: Prepare the conditional profit table for each decision-event combination and write the associate
probabilities. The general approach used in decision tree analysis is to work ‘backward’ through the tree from right to

Step 2: For each event determine the conditional opportunity loss (COL) by subtracting the payoff from left,computing the expected value of each chance node. We than, choose the particular branch leaving a

highest payoff decision node which leads to the chance node with the highest expected values. This is known as roll back
or “fold back process”.
Step 3: Calculate the expected Opportunity Loss (EOL) by multiplying COL with associate probability

Step 4: Select the alternative that gives lowest EOL

STEPS IN DECISION TREE ANALYSIS:
Expected Profit of perfect Information (EPPI): Step 1: Identify the decision points and alternative courses of action

EPPI = probabilityX payoff under the perfect information 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

Expected Value of perfect Information (EVPI): Step 4: Choose course of action that gives best payoff for each of the decision

EVPI = EPPI – Max Expected EMV 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

DECISION TREE: Advantages and Limitations of Decision Treeapproach:

Page 9 of 92
 Page 19 of 22 Page 20 of 22

Advantages: Simulation:

1. It structures the decision process and helps decision-making in an orderly, systematic, and Definition:Simulation is a representation of reality through the use of a model or other device which will
sequential manner.
react in the same manner as reality under a given set of conditions. Simulation has also been defined as” the
2. It requires the decision maker to examine all possible outcomes, whether desirable or undesirable. use of a system model that has the designed characteristics of reality in order to produce the essence of
3. It communicates the decision making process to others in an easy and clear manner. Illustrating actual operation”.
each assumption about the future.
Example:
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 1. A children cycling park, with various crossings and signals, is a simulated model of the city traffic
subsequent decisions. system.
6. It can be applied in various filed such as introductionof new product, marketing, make or buy
decisions, investment decisions, etc.
2. Planetarium shows represent a beautiful simulation of the planet system

Disadvantages (or Limitations): Monte Carlo Simulation:

1. Decision tree diagrams become more complicated as the number of decision alternatives increases.
The Monte Carlo method of Simulation was developed by two mathematicians J.V. Neumann and S.Ulam.
2. It becomes highly complicated when interdependent alternatives and dependent variables are The technique employs random numbers and is used to solve problems that involve probability and wherein
present in the problem.
physical experimentation is impracticable and formulation of mathematical model is impossible. It is a
3. It assumes the utility of money in linear with money. method of simulation by sampling technique. The steps involved in carrying out Monte Carlo Simulation
4. It analyses the problem in terms of expected values and thus yield as average valued solution. are:
5. There is often inconsistency in assigning probabilities for different events.
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:

Unit-V: Simulation

Page 10 of 92
 Page 21 of 22 Page 22 of 22

1) Simulation offers to solution to offers the solution by allowing experimentation with model.  Shop floor management

2) Estimates the need of costly trial and error methods  Scheduling and production process

3) This help in getting proposed plans accepted and implemented.  Location of emergency vehicles

4) Simulation models are comparatively flexible.  Making inventory policy decisions

5) Simulation models are easier to use comparatively mathematical models.  Portfolio selection and capital budgeting
6) Capacity to lend itself to problem that is impossible to handle mathematically using analytical  Design of queuing system and In biomedical system
methods.
 In the study of projects involving the risk investments
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. Programming Languages and Simulation Languages:

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. Programming Languages:

• FORTRAN
• GASP
• SIMSCRIPT
Disadvantages of Simulation:
• GPSS

 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 Simulation Languages:

• DYNAMO
question for a given set of inputs.
• SIMPAC
 It may not prove economical.
• SIMULATE
 It is a tool of solution evaluation and thus does not generate problem solution. • SIMULA
• GSMP
• ESP
• CSL

****

Application of Simulation:

Page 11 of 92
Page 12 of 92
Page 13 of 92
Page 14 of 92
Page 15 of 92
Page 16 of 92
Page 17 of 92
Page 18 of 92
Page 19 of 92
Page 20 of 92
Page 21 of 92
Page 22 of 92
Page 23 of 92
Page 24 of 92
Page 25 of 92
Page 26 of 92
Page 27 of 92
Page 28 of 92
Page 29 of 92
Page 30 of 92
Page 31 of 92
Page 32 of 92
Page 33 of 92
Page 34 of 92
Page 35 of 92
Page 36 of 92
Page 37 of 92
Page 38 of 92
Page 39 of 92
Page 40 of 92
Page 41 of 92
Page 42 of 92
Page 43 of 92
Page 44 of 92
Page 45 of 92
Page 46 of 92
Page 47 of 92
Page 48 of 92
Page 49 of 92
Page 50 of 92
Page 51 of 92
Page 52 of 92
Page 53 of 92
Page 54 of 92
Page 55 of 92
Page 56 of 92
Page 57 of 92
Page 58 of 92
Page 59 of 92
Page 60 of 92
Page 61 of 92
Page 62 of 92
Page 63 of 92
Page 64 of 92
Page 65 of 92
Page 66 of 92
Page 67 of 92
Page 68 of 92
Page 69 of 92
Page 70 of 92
Page 71 of 92
Page 72 of 92
Page 73 of 92
Page 74 of 92
Page 75 of 92
Page 76 of 92
Page 77 of 92
Page 78 of 92
Page 79 of 92
Page 80 of 92
Page 81 of 92
Page 82 of 92
Page 83 of 92
Page 84 of 92
Page 85 of 92
Page 86 of 92
Page 87 of 92
Page 88 of 92
Page 89 of 92
Page 90 of 92
Page 91 of 92
Page 92 of 92