Management Optimization
Techniques
DECISION THEORY
Presented By,
ANNIE JOSE - 215403
Decision making is a process of selecting a best course of action
from a number of alternatives for achieving the objectives.
In organisations decisions may be on the basis of cost, or on the
basis of competitiveness or at times on intuitions.
Statistical Decision theory is used for solving decision problems
where Uncertainty plays a crucial role.
Statistical methods try to minimise the degree to which the person is
likely to regret the decision.
Components of a decision problem
1) Decision Maker
2) Acts(courses of action)
Different possible strategies that the decision maker can choose
Problem is to choose the best alternative to achieve an objective
Under the control of Decision maker.
3) Event (State of Nature)
These are set of exclusive outcomes consequent to the decisions made.
These events are often called states of nature or outcome.
he decision maker has no control over which the event will take place
4) Outcomes
When the Decision maker Select a particular act under a particular state of nature, the
result obtained is called outcome [Consequence]
a) Payoff :- Outcome in quantitative form.
monetary gain or loss.
b) Opportunity Loss (Regret) :- Loss incurred because of the failure to take best
possible decision
Types of Decision Making Situations
Q. A food products company is planning the introduction of a revolutionary
new product with new packing to replace the existing product at much higher
price (S1) or a moderate change in the composition of the existing product
with a new packaging at a small increase in price (S₂) or a small change in
the composition of the existing except the word, "New" with a negligible
increase in the price (S3). The three possible states of nature of 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 pay offs in terms
of yearly new profits for each of the strategies on these events.
This is represented in the following table.
States of nature
Strategies N1 N2 N3
S1 700 300 150
S2 500 450 0
S3 300 300 300
Which strategy should the executive concerned choose on the basis of
(a) Maximin Criterion (b) Maximax Criterion (c) Minimax Criterion (d) Laplace
Criterion (e) Harwicz alpha criterion
SOLUTION:-
STEP 1 :- Rewriting the Pay off table (with strategies in columns)
“ Let every RAW of pay off table represent a STATE OF NATURE &
every COLUMN represent a COURSE OF ACTION “
states of nature
[Link] Criterion
1)Find out minimum payoff (column wise)
S1 S2 S3
2) Max of the minima
N1 700 500 300
N2 300 300450 300
N3 150The Executive should choose
0 strategy S3 300
2. Maximax criterion
1)Find out maximum payoff (column wise)
S1 S2 S3
N1 700 500 300
N2 300 450 300
N3 150 0 300
2) Maximum of the maxima
700
The executive can choose Act S1
3. Minimax regret criterion
1)Payoff table
S1 S2 S3
N1 700 500 300
N2 300 450 300
N3 150 0 300
2) Prepare Regret table / Opportunity Loss Table from payoff table
S1 S2 S3
N1 700-700 = 0 700-500 = 200 700-300 = 400
N2 450-300 = 150 450-450 = 0 450-300 = 150
N3 300-150 = 150 300-0 = 300 300-300 = 0
3) Apply Minimax criteria
I. Find out maximum opportunity Loss
II. Minimum of these maxima
150
The executive should choose strategy S1 since it minimizes the maximum
[Link] Criterion
1)Calculate expected pay off from pay off table by using the formula
1/n[P1+P2 + ……+ P n]
S1 S2 S3
N1 700 500 300
N2S1=1/3[700+300+150] = 383.3
S2=1/3[500+450+0] = 316.67
300 450 300
S2=1/3[300+300+300] = 900
N3 150 0 300
2) Select maximum expected payoff as optimal Act
TOTAL S1=383.3 1150
The executive may select strategy 1
950 900
Average (divided 383.3(1150/3) 316.67(950/3) 300(900/3)
by 3)
[Link] Criterion
1)Find out weighted outcome (D1,D2,D3 etc.)connected with all strategies
where Di = α Mi + (1- α) mi
α = the coefficient of optimism. Varies between 0 and 1.
Mi = Maximum payoff of ‘i’ th strategy
mi = minimum payoff of ‘i’ th strategy
S1 S2 S3
[ given α = 0.6 ]
N1 700 M = 420+60 = 480
D1= (0.6* 700)+(1-0.6)150 500 M 300 M
D2= (0.6* 500)+(1-0.6)0 = 300+0 = 300
D3= (0.6* 300)+(1-0.6)300 = 180+120 = 300
N2 300 450 300
2) Select
N3an alternative with highest
150 weighted
m outcome ( Di) 0m 300 m
Since, D1 is Max, select S1 as optimal strategy
Thank you !...
