Decision

making under
Uncertainty
Dr. K. Anbumani
Associate Professor
Decision Making Under Uncertainty

In the absence of knowledge about the probability of any state of

nature (future) occuring, the decision maker must arrive at a decision
only on the actual conditional payoff values, together with a policy
(attitude). There are several different criteria of decision making in this
situation as given below;

(a) Optimism criterion (maximax or minimin)

(b) Pessimism criterion (maximin or minimax)
(c) Equal probabilities criterion (Laplace)
(d) Co-efficient of optimism criterion (Hurweiz)
(e) Regret criterion (Salvage)
2
 (a) Optimistic criterion (maximax or minimin)

In this criterion the decision maker ensures that he should

not miss the opportunity to achieve the largest possible profit
(maximax) or the lowest possible cost (minimin) by selecting
the maximum of maxima or minimum of minima payoffs
(consequences or outcomes). The steps involved are as follows;

(a) Locate the maximum (or minimum payoff value

corresponding to each alternative courses of action.
(b) Select the maximum for profit and minimum for the
cost. It is optimistic in approach.

3
 (b) Pessimistic or Wald’s Criterion (maximin or minimax)

In this criterion the decision maker ensures that he would earn not
less than (or more than) some specified amount. Thus he selects the
alternative that represents the maximum of the minima (or the minimum
of the maxima) payoffs (consequences or outcomes). The steps involved
are as follows;

(a) Locate the minimum ( or minimum) corresponding to each

alternatives.
(b) Select the minimum for profit and maximum for the cost.
It is an pessimistic in approach.
4
(c) Equal probabilities Criterion (Laplace)

Since the probabilities of states of nature are not known, it

is assumed that all states of nature will occur with equal
probability. i.e., each state of nature is assigned an eqaul
probaility. The states of nature are mutually exclusive and
collectivelky exhaustive. The steps involved are as follows;

(a) Assign equal probability to each state of nature as the

probability of each of these must be equal to 1 / No. of states of
nature

5
(c) Equal probabilities Criterion (Laplace)

(b) Compute the expected payoff for each alternative course of

action by adding all payoffs and dividing by the number of
possible states of nature;
Probability of states of nature (j) x
Payoff value for the combination of alternative I and state of
nature j

(c) Select the best expected payoff value (maximum for profit
and minimum for cost).

6
 (d) Co-efficient of optimism criterion (Hurwicz)

It says that a rational decision maker should be neither

completely optimistic nor pessimistic. Hurwicz used the coefficient
of optimism (α alpha) to measure the decision maker’s degree of
optimism. This coefficient lies between 0 and 1 where 0 represents
complete pessimism and 1 represents complete optimism about the
future outcome. i.e., since alpha (α) refers to coefficient of
optimism
(1- α) refers to coefficient of pessimism. Decision maker would
select an alternative that maximizes H where,
H (criterion of realism) = α (max in column) +
(1-α) (min in coloumn)
7
 (d) Co-efficient of optimism criterion (Hurwicz)

The steps involved are as follows;

(a) Decide the co-efficient of optimism (α) and pessimism (1-α)
(b) Select the largest and smallest payoff values for each alternative.
(c) Multiply these values with (α) and (1-α) respectively
(d) Calculate the weighted average (H) using the
above formula
(e) Select the alternative with best anticipated
weighted average payoff value.

8
(e) Regret criterion (Salvage)

This crieterion is also known as opportunity loss

criterion or minimax regret decision criterion. This is
because the decision maker regrets the fact that he adopted
a wrong alternative (course of action) resulting in an
opportunity loss of payoff. Thus he always intends to
minimize this regret. The steps involved are as follows;

(a) From the given pay-off matrix, develop an opportunity loss (or)
regret matrix as follows;

9
(e) Regret criterion (Salvage)

i. Find the best payoff corresponding to each state of nature

ii. Subtract all other entries (payoff values) in that row from this
value

(b) For each alternative, identify the worst or maximum regret

value. Record this number in a new row.

(c) Select the alternative with the smallest anticipated opportunity

loss (regret) value.

10
 Example 1

A food products company is contemplating the

Introduction of a new product with new packaging replacing
the existing product at much higher price (S1). It may even
make a moderate change in the composition of the existing
product with a new packaging at a small increase in price (S 2),
or it may make a small change in the composition of the
existing product advertising with a word “new” and a negligible
increase in the price (S3).

11
 Example 1

The 3 possible states of nature or events are;

(i) High increase in sales (N1)

(ii) No change in sales (N2)
(iii) Decrease in sales (N3)

The marketing department of the company worked out the

payoffs in terms of yearly net profits for each of the strategies
of 3 events or expected sales as shown below;
12
 Example 1

States of Nature

Strategies N1 N2 N3

S1 700000 300000 150000

S2 500000 450000 0

S3 300000 300000 300000

Which strategy should be selected according to;

a) Maximin c) Regret
b) Maximax d) Laplace criterion

13
 MAXImin Criterion

States of Nature

Strategies S1 S2 S3

N1 700000 500000 300000

N2 300000 450000 300000

N3 150000 0 300000

Column min 150000 0 300000

14
 MAXImin Criterion

States of Nature

Strategies S1 S2 S3

N1 700000 500000 300000

N2 300000 450000 300000

N3 150000 0 300000

Column min 150000 0 300000

Maximin 300000

Maximin value is Rs. 300000

So the company should adopt strategy S3

15
 MAXImax Criterion

“
States of Nature

Strategies S1 S2 S3

N1 700000 500000 300000

N2 300000 450000 300000

N3 150000 0 300000

Column max 700000 500000 300000

16
 MAXImax Criterion

“
States of Nature

Strategies S1 S2 S3

N1 700000 500000 300000

N2 300000 450000 300000

N3 150000 0 300000

Column max 700000 500000 300000

Maximax 700000

Maximax value is Rs. 700000

So the company should adopt strategy S1

17
Regret (or) Opportunity Loss Criterion

States of Nature
Strategies S1 S2 S3
N1 700000-700000 700000-500000 700000-300000
0 200000 400000
N2 450000-300000 450000-450000 450000-300000
150000 0 150000
N3 300000-150000 300000-0 300000-300000
150000 300000 0
Column max 150000 300000 400000

18
Regret (or) Opportunity Loss Criterion

States of Nature
Strategies S1 S2 S3
N1 700000-700000 700000-500000 700000-300000
0 200000 400000
N2 450000-300000 450000-450000 450000-300000
150000 0 150000
N3 300000-150000 300000-0 300000-300000
150000 300000 0
Column max 150000 300000 400000
Min Regret 150000

Minimum regret or opportunity loss value is Rs. 150000

So the company should adopt strategy S1
19
 Laplace Criterion

“
States of Nature

Strategies S1 S2 S3

N1 700000 500000 300000

N2 300000 450000 300000

N3 150000 0 300000

Column Total / 3 11,50000 / 3 9,50000 / 3 9,00000 / 3

Ave. Outcome 3,83,333.33 3,16,666.66 300000

Largest Return 3,83,333.33

Since we do not know the probabilities of events that may happen, we

give equal probability 1/3. The largest expected return Rs.383333.33
happens from S1
20
 Example 2

A manufacturer uses principal ingredient X to produce a good by

spending Rs.1000 per year. There is a possibility that the price of X
may increase 4 times. He can use the combination of Y and Z to get
the same effect of X which will together cost Rs.3000 per year.No
change in their price is expected. The co-efficient of optimism is 0.4.
Find the suitable course of action that minimizes the cost;

States of Nature

Strategies S1 (Use Y & Z) S2 (Use X)

N1 (Price of X Increases) - 3000 - 4000

N2 (Price of X doesn’t Increase) - 3000 - 1000

21
 MAXImin Criterion

States of Nature

Strategies S1 (Use Y & Z) S2 (Use X)

N1 (Price of X Increases) - 3000 - 4000

N2 (Price of X doesn’t Increase) - 3000 - 1000

Column min - 3000 - 4000

Maximin - 3000

Maximin value is Rs. 3000 (-)

So the company should adopt strategy S1

22
 Regret (or) Opportunity Loss Criterion

“
States of Nature
Strategies S1 (Use Y & Z) S2 (Use X)
N1 (Price of X Increases) - 3000 (-) - 3000 -3000 (-) – 4000
0 1000
N2 (Price of X doesn’t Increase) - 1000 (-) -3000 - 1000 (-) - 1000
2000 0
Column max 2000 1000
Minimax 1000

Minimax value is Rs. 1000

So the company should adopt strategy S2

23
 Hurwicz Criterion

“
H (criterion of realism) = α (max in column) + (1-α) (min in coloumn)
= α (Best payoff) + (1-α) (Worst payoff)

Since the co-efficient of Optimism is 0.4,

Co-efficient of Pessimism would be 1-0.4 = 0.6
States of Nature
Strategies S1 (Use Y & Z) S2 (Use X)
Best payoff -3000 x 0.4 -1000 x 0.4
-1200 - 400
Worst payoff -3000 x 0.6 - 4000 x 0.6
-1800 - 2400
H - 3000 - 2800
Least cost -2800

24
 Thank you
25