Decision Making

Analysis

Prof. Eloida C. Dagumboy

Decision Making Formal Procedures

1. Break-even analysis
Identifies how much change in volume or demand is
necessary before a second alternative becomes better
than the first.
2. Preference matrix
Deals with multiple criteria that cannot be evaluated
with a single measure of merit, such as total profit or
cost.
3. Decision Theory
Aids in choosing the best alternative when outcomes are
uncertain.
4. Decision tree
Used for sequential decisions, when today’s best decision
depends on tomorrow’s decisions and events.
Decision Theory

• A general approach to decision making when the

outcomes associated with alternatives are often
in doubt.
• It helps operations managers with decisions on
process, capacity, location, and inventory
because such decisions are about uncertain
future.
• Can be used by managers in functional areas.
Decision Theory

• With decision theory, a manager can make choices using

the following process:
1. List the feasible alternatives.
Including “to do nothing”
2. List events that have an impact on the outcome of the choice
but aren’t under the manager’s control.
Chance events or states of nature
3. Calculate the payoff for each alternative in each event.
Total profit or total cost
4. Estimate the likelihood of each event, using past data,
executive opinion or forecasting methods.
Express it as a probability, which total is 1.0
5. Select a decision rule to evaluate alternatives.
E.g. Lowest expected cost, amount of information, manager’s
attitude towards risk
Decision Making Under Certainty

• The simplest situation is when the manager

knows what will occur.
• Decision rule is to pick the alternative with
• the highest payoff, if expressed in profits
• The lowest payoff, if expressed in costs
Pay Off Table / Decision Table

Possible Future • C1 –
Demands Decision Alternatives
ALTERNATIVES Low High • C2 & C3 –
States of nature or
outcomes
Small Facility 200,000 270,000
• Values – Payoffs
• Profit
Large Facility 160,000 800,000 • Cost
• Distance
Do Nothing - - • Time
Sample Analysis

• Build a small or large

Possible Future facility? The manager is
Demands aware with certainty the
payoffs for each
Alternatives Low High alternative.

Small Facility 200,000 270,000 • What is the best choice if

the future demand will be
low?
Large Facility 160,000 800,000
• Build a small facility to
enjoy a payoff of
Do Nothing - - P200,000.
Decision Making Under Uncertainty

• The manager can list the possible events but

cannot estimate their probabilities.

• The manager can use one of the four decision

rules:
• Maximax (Optimist)
• Best of the best alternative
• Maximin (Pessimist)
• Best of the worst alternative
• Minimax regret
• Worst regret alternative
• Laplace (Realist)
• Best weighted payoff alternative
 Possible Future • Maximax (Optimist)
Demands • Best of the best
• P800,000
ALTERNATIVES Low High
• Build a large
facility
Small Facility 200,000 270,000
• Maximin (Pessimist)
• Best of the worst
Large Facility 160,000 800,000 • P200,000
• Build a small
facility
Do Nothing - -
 Regret (000) Maximum
ALTERNATIVES Low High Regret
200 270
Small Facility 200-200=0 800-270=530 530
160 800
Large Facilty 200-160=40 800-800=0 40
Do Nothing - -
 The minimization of regret that is highest when one decision Minimax regret
has been made instead of another. In a situation in which a • Worst regret alternative
decision has been made that causes the expected payoff of
an event to be less than expected, this criterion encourages • To minimize the
the avoidance of regret. maximum regret, pick
 Regret is also called opportunity loss. the large facility
Alternatives Best Payoff (P000)
Small Facility 0.5(200)+0.5(270)=235
Large Facilty 0.5(160)+0.5(800)=480
Do Nothing
 Laplace (Realist)
○ Best weighted payoff
○ P480,000
○ Build a large facility
Preference Matrix

• A table that allows the manager to rate an

alternative according to several performance
criteria.
• The criteria can be scored on any scale.
• E.g.1- worst possible to 10-best possible
• 0 to1, as long as the same scale is applied to all the
alternatives being compared
• Each score is weighted according to its perceived
importance, with a total of 100.
• The total score is the sum of the weighted scores
(w X s) for all the criteria.
• The manager can compare the scores for the
alternatives against one another or against a
predetermined threshold.
Preference Matrix - Sample Problem

The following table shows the performance criteria, weights, and scores
(1=worst, 10 = best) for new products: air conditioner and electric oven.
If the manager wants to introduce just one new product which of the 2
products would the firm should pursue?
Performance Criterion Weight Score - A Score - O WXA WXO
Market potential 30 8 7 240 210
Unit profit margin 20 10 9 200 180

Operations compatibility 20 6 10 120 200

Competitive Advantage 15 10 9 150 135

Investment Requirement 10 2 5 20 50
Project Risk 5 4 5 20 25
100 750 800

The management should pursue making electric ovens since the sum of
the weighted scores (800) is greater than the sum of the weighted scores
of making air condition (750)
Decision Tree

• Can be used for new product

decisions as well as for variety
of other management
problems.

• Helpful when there are series

of decisions and various
outcomes that lead to
subsequent decisions followed
by other outcomes.
Decision Tree Procedure

• Include all possible alternatives in the tree.

• Nature of each alternative
• Do nothing
• Enter payoffs at the end of the appropriate
branch.
• To develop the payoff of achieving each branch
• Determine the expected value of each course of
action.
• Start at the end of the tree, calculate values at each
step and prune alternatives that are not as good as
others from the same node.
Decision Tree - Problem No.1

• A semiconductor manufacturer is investigating the possibility of

producing and marketing a microprocessor. Undertaking this project
will require either purchasing a sophisticated CAD system or hiring and
training additional engineers. The market for the product could be
either favorable or unfavorable. The company has the option of not
developing the new product at all.

• With favorable acceptance by the market, sales would be 25,000

processors selling for P100 each. With unfavorable acceptance, sales
would be only 8,000 processors selling for P100 each. The cost of CAD
equipment is P500,000, but that of hiring and training new engineers is
only P375,000. However, manufacturing costs should drop from P50
each when manufacturing without CAD to P40 each when manufacturing
with CAD.

• The probability of favorable acceptance of the new microprocessor is

0.40; the probability of unfavorable acceptance is 0.60.

• Use the decision tree to compute the expected monetary value (EMV)
for each case. The one with the highest EMV represents the best
decision.
Decision Tree - Problem No.2

• Dags Inc. has the option of proceeding immediately with

production of a new top of the line TV that has just
completed prototype testing or having the value analysis
team complete a study.
• If Eloi, VP for operations, proceeds with the existing
prototype the firm expect sales to be 100,000 units at
P550 each, with a probability of 0.6 and 0.4 probability
of 75,000 at P550 each. If she uses the value analysis
team, the firm expects sales of 75,000 units at P750 with
a probability of 0.7 and 0.3 probability of 70,000 units at
P750. Cost of the value analysis is P100,000.
• Which option has the highest expected monetary value
(EMV).