Decision Making (Chapter 9)

Download as pdf or txt
Download as pdf or txt
You are on page 1of 55

9.

OPERATIONS DECISION-MAKING
CHAPTER OUTLINE
9.1 Introduction
9.2 Management as a Science
9.3 Characteristics of Decisions
9.4 Framework for Decision-Making
9.5 Decision Methodology
9.6 Decision Support System
9.7 Economic Models
9.8 Statistical Models
9.9 Decision Tree
9.1 INTRODUCTION
Thousand of business decisions are made everyday. Not
all the decisions will make or break the organization. But
each one adds a measure of success or failure to the
operations. Hence decision-making essentially involves
choosing a particular course of action, after considering
the possible alternatives. This chapter examines
management as a science and the characteristics of
decisions.
The use of economic and statistical models is discussed
along with decision trees.
9.2 MANAGEMENT AS A SCIENCE
Management scientists hold that, education, scientific
training and experience can improve a person’s ability to
make decisions. Scientific decision-making rests upon
organized principles of knowledge and depends largely
upon the collection of empirical data and analysis of the
data in a way that repeatable results will be obtained.
The association of management with the scientific
method involves drawing objective conclusions from the
facts. Facts come from the analysis of data, which must
be gathered, compiled and digested into meaningful form,
such as graphs and summary statistics.
Computers are helpful in these tasks because they
can easily store data and provide us with the more
sophisticated and statistical analysis. But not all
variables are quantifiable, so decision-makers must
still use some value-based judgments in a decision
process.
Thus management as a science is characterized by:
 Organized principle of knowledge.
 Use of empirical data.
 Systematic analysis of data.
 Repeatable results
9.3 CHARACTERISTICS OF DECISIONS
Operations decision range from simple judgments to
complex analyses, which also involve judgment.
Judgment typically incorporates basic knowledge,
experience, and common sense. They enable to blend
objectives and sub-objective data to arrive at a choice.
The appropriateness of a given type of analysis
depends on:
 The significant or long lasting decisions,
 The time availability and the cost of analysis, and
 The degree of complexity of the decision.
 The significant or long lasting decisions deserve more
considerations than routine ones. Plant investment,
which is a long-range decision, may deserve more
thorough analysis.
 The time availability and the cost of analysis also
influence the amount of analysis.
 The degree of complexity of the decision increases
when many variables are involved, variables are highly
independent and the data describing the variables are
uncertain.
Business decision-makers have always had to work with
incomplete and uncertain data.
Fig. 9.1 below depicts the information environment of
decisions. In some situations a decision maker has
complete information about the decision variables; at
the other extremes, no information is available.
Operations management decisions are made all along
this continuum.
Complete certainty in decision-making requires data on
all elements in the population. If such data are not
available, large samples lend more certainty than do
small ones. Beyond this, subjective information is likely
to be better than no data at all.
Fig. 9.1 Information Continuum
9.4 FRAMEWORK FOR DECISION-MAKING

An analytical and scientific framework for


decision implies the following systematic steps:
 Defining the problem.
 Establish the decision criteria.
 Formulation of a model.
 Generating alternatives.
 Evaluation of the alternatives.
 Implementation and monitoring.
DEFINING THE PROBLEM
Defining the problem enables to identify the relevant
variables and the cause of the problem. Careful definition
of the problem is crucial. Finding the root cause of a
problem needs some questioning and detective work. If a
problem defined is too narrow, relevant variable may be
omitted. If it is broader, many tangible aspects may be
included which leads to the complex relationships.
ESTABLISH THE DECISION CRITERIA
Establishing the decision criterion is important because
the criterion reflects the goals and purpose of the work
efforts.
For many years profits served as a convenient and
accepted goal for many organizations based on
economic theory. Nowadays, organization will have
multiple goals such as employee welfare, high
productivity, stability, market share, growth, industrial
leadership and other social objectives.
FORMULATION OF A MODEL
Formulation of a model lies at the heart of the scientific
decision-making process. Model describes the essence
of a problem or relationship by abstracting relevant
variables from the real world situation.
Models are used to simplify or approximate reality, so
the relationships can be expressed in tangible form and
studied in isolation.
Modeling a decision situation usually requires both
formulating a model and collecting the relevant data to
use in the model. Mathematical and statistical models
are most useful models for understanding the complex
business of the problem. Mathematical models can
incorporate factor that cannot readily be visualized.
With the aid of computers and simulation techniques,
these quantitative models can be made flexible.
GENERATING ALTERNATIVES
Alternatives are generated by varying the values of the
parameters. Mathematical and statistical models are
particularly suitable for generating alternatives because they
can be easily modified. The model builder can experiment
with a model by substituting different values for controllable
and uncontrollable variable.
EVALUATION OF THE ALTERNATIVES
Evaluation of the alternatives is relatively objective in an
analytical decision process because the criteria for
evaluating the alternatives have been precisely defined. The
best alternative is the one that most closely satisfies the
criteria.
Some models like LPP model automatically seek out a
maximizing or minimizing solution. In problems various
heuristic and statistical techniques can be used to suggest the
best course of action.
IMPLEMENTATION AND MONITORING
Implementation and monitoring are essential for completing the
managerial action. The best course of action or the solution to a
problem determined through a model is implemented in the
business world. Other managers have to be convinced of the
merit of the solution. Then the follow-up procedures are
required to ensure about appropriate action taken. This includes
an analysis and evaluation of the solution along with the
recommendations for changes or adjustments.
9.5 DECISION METHODOLOGY

The kind and amount of information available helps to


determine which analytical methods are most
appropriate for modeling a given decision. Figure 9.2
illustrates some useful quantitative methods that are
classified according to the amount of certainty that
exists with respect to the decision variables and
possible outcomes. These analytical techniques often
serve as the basis for formulating models, which help to
reach operational decisions.
Complete Risk & Extreme
Certainty Uncertainty Uncertainty

(All information) (Some information) (No information)


Algebra: Statistical Analysis: Game theory
• Break-even • Objective and subjective probabilities Flip coin
• Benefit/cost • Estimation and tests of hypothesis
Calculus • Bayesian statistics
Mathematical • Decision theory
programming: • Correlation and regression
• Linear • Analysis of variance
• Non-linear • Non parametric methods
• Integer Queuing Theory
• Dynamic Simulation
• Goal Heuristic Methods
Network Analysis Techniques:
• Decision trees
• PERT and CPM
Utility Theory

Fig. 9.2 Quantitative methods as a function of degree of certainty


The degree of certainty is classified as complete
certainty, risk and uncertainty and extreme uncertainty.
9.5.1 Complete Certainty Methods
Under complete certainty conditions, all relevant
information about the decision variables and outcomes is
known or assumed to be known. Following are some of
the methods used:
 Algebra: This basic mathematical logic is very useful
for both certainty and uncertainty analysis.
With valid assumptions, algebra provides deterministic
solutions such as break-even analysis and benefit cost
analysis.
 Calculus: This branch of mathematics provides a useful
tool for determining optimal value where functions such
as inventory costs, are to be maximized or minimized.
 Mathematical programming: Programming techniques
have found extensive applications in making a product
mix decisions; minimizing transportation costs, planning
and scheduling production and other areas.
9.5.2 Risk and Uncertainty Methods
In risk and uncertainty situations, information about the
decision variables or the outcomes is probabilistic.
Following are some of the useful approaches:
 Statistical analysis: Objective and subjective
probabilities with the use of probability and probability
distribution, Estimation and tests of hypothesis, Bayesian
statistics, Decision theory, Correlation and regression
technique for forecasting demand and Analysis of
variance are some of the techniques used for decision-
making.
 Queuing theory: The analysis of queues in terms of
waiting-time length and mean waiting time is useful in
analyzing service systems, maintenance activities, and
shop floor control activities.
Simulation: Simulation duplicates the essence of an activity.
Computer simulations are valuable tools for the analysis of
investment outcomes, production processes, scheduling and
maintenance activities.
 Heuristic methods: Heuristic methods involve set of rules, which
facilitate solutions of scheduling, layout and distribution problems
when applied in a consistent manner.
 Network analysis techniques: Network approaches include
decision trees, CPM and PERT methods. They are helpful in
identifying alternative course of action and controlling the project
activities.
 Utility theory: Utility theory or preference theory allows decision-
makers to incorporate their own experience and values into a
relatively formalized decision structure.
9.5.3 Extreme Uncertainty Methods
Under extreme uncertainty, no information is available
to assess the likelihood of alternative outcomes.
Following are some of strategies to solve this:
 Game theory: Game theory helps decision-makers
to choose course of action when there is no
information about what conditions will prevail.
 Coin flip: Flipping a coin is sometimes used in
situation where the decision-makers are wholly
indifferent.
9.5.4 Decision-Making Under Uncertainty
No information is available on how likely the various states of
nature are under those conditions. Four possible decision
criteria are Maximin, Maximax, Laplace, and Minimax
regret. These approaches can be defined as follows:
Maximin: Determine the worst possible pay-off for each
alternative, and choose the alternative that has the “best
worst.” The Maximin approach is essentially a pessimistic
one because it takes into account only the worst possible
outcome for each alternative. The actual outcome may not be
as bad as that, but this approach establishes a “guaranteed
minimum.” CHOOSE MAXIMUM OF THE MINIMUM VALUES!
Maximax: Determine the best possible pay-off, and choose the
alternative with that pay-off. The Maximax approach is an optimistic,
“go for it” strategy; it does not take into account any pay-off other than
the best. CHOOSE MAXIMUM OF THE MAXIMUM VALUES!

Laplace: Determine the average pay-off for each alternative, and


choose the alternative with the best average. The Laplace approach
treats the states of nature as equally likely.
CHOOSE MAXIMUM OF THE AVERAGE VALUES!
Minimax regret: Determine the worst regret for each alternative,
and choose the alternative with the “best worst.” This approach
seeks to minimize the difference between the pay-off that is realized
and the best pay-off for each state of nature.
CHOOSE MINIMUM OF THE MAXIMUM REGRETS!
ILLUSTRATION 1: Referring to the pay-off table, determine
which alternative would be chosen under each of these
strategies:
(a) Maximin, (b) Maximax, and (c) Laplace.

Possible Future Demand in Birr (Million)

Alternatives Low Moderate High

Small facility 10 10 10

Medium facility 7 12 12

Large facility -4 2 16
SOLUTION
(a) Using Maximin, the worst pay-offs for the alternatives are:
Small facility: Birr 10 million
Medium facility: Birr 7 million
Large facility: Birr –4 million
Hence, since Birr 10 million is the best, choose to build the
small facility using the maximum strategy.
(b) Using Maximax, the best pay-offs are:
Small facility: Birr 10 million
Medium facility: 12 million
Large facility: 16 million
The best overall pay-off is the Birr 16 million in the third row.
Hence, the Maximax criterion leads to building a large facility.
(c) For the Laplace criterion, first find the row totals, and
then divide each of those amounts by the number of states of
nature (three in this case). Thus, we have:

Row Total Row Average


(Birr, Million) (Birr, Million)
Small Facility 30 10.00

Medium Facility 31 10.33

Large Facility 14 4.67

Because the medium facility has the highest average, it


would be chosen under the Laplace criterion.
ILLUSTRATION 2: Determine which alternative would be chosen
using a Minimax regret approach to the capacity-planning
programme.
SOLUTION: The first step in this approach is to prepare a table
of opportunity losses, or regrets. To do this, subtract every
pay-off in each column from the best pay-off in that column.
For instance, in the first column, the best pay-off is 10, so each
of the three numbers in that column must be subtracted from
10. Going down the column, the regrets will be 10 – 10 = 0, 10 –
7 = 3, and 10 – (– 4) = 14. In the second column, the best pay-
off is 12. Subtracting each pay-off from 12 yields 2, 0, and 10.
In the third column, 16 is the best pay-off. The regrets are 6, 4,
and 0. These results are summarized in a regret table:
Regrets (in Birr Million)

Alternative Low Medium High Worst

Small Facility 0 2 6 6

Medium Facility 3 0 4 4

Large Facility 14 10 0 14

The second step is to identify the worst regret for each


alternative. For the first alternative, the worst is 6; for the
second, the worst is 4; and for the third, the worst is 14. The
best of these worst regrets would be chosen using Minimax
regret. The lowest regret is 4, which is for a medium facility.
Hence, that alternative would be chosen.
9.5.5 Decision-Making Under Risk
Between the two extremes of certainty and uncertainty lies the
case of risk: The probability of occurrence for each state of
nature is known. (Note that because the states are mutually
exclusive and collectively exhaustive, these probabilities must
add to 1.00.) A widely used approach under such
circumstances is the expected monetary value criterion.
The expected value is computed for each alternative, and the
one with the highest expected value is selected. The
expected value is the sum of the pay-offs for an alternative
where each pay-off is weighted by the probability for the
relevant state of nature.
ILLUSTRATION 3: Determine the expected pay-off
of each alternative, and choose the alternative that
has the best-expected pay-off. Using the expected
monetary value criterion, identify the best
alternative for the previous pay-off table for these
probabilities: low = 0.30, moderate = 0.50, and high
= 0.20. Find the expected value of each alternative
by multiplying the probability of occurrence.
Possible Future Demand (in Birr, Million)

Alternatives Low Moderate High

Small facility 10 10 10

Medium facility 7 12 12

Large facility -4 2 16
SOLUTION: For each state of nature by the pay-off for that state of
nature and summing them:
EVsmall = 0.30 (Birr 10) + 0.50 (Birr 10) + 0.20 (Birr 10) = Birr 10
EVmedium = 0.30 (Birr 7) + 0.50 (Birr 12) + 0.20 (Birr 12) = Birr 10.5
EVlarge = 0.30 (–4) + 0.50 (Birr 2) + 0.20 (Birr 16) = Birr 3
Hence, choose the medium facility because it has the highest
expected value.
ILLUSTRATION 4: Global Telecom Corp. must choose one
of three partnering firms (X Co., Y Co., or Z Co.) with which
to develop a personal communicator. The possible states of
nature are that future demand may be low, medium, or high.
Estimated dollar pay-offs (in net present value terms) for
each alternative under each state of nature are shown in the
accompanying Table. Which partnering firm should be
chosen under the criterion of:
a) maximax,
b) Maximin, and
c) Laplace?
Pay-off Table
Profit (Birr, Million) If Future
Alternatives Demand Is

Low Medium High

X Co. Partner 10 50 70

Y Co. Partner -10 44 120

Z Co. Partner 0 20 140


SOLUTION
a) Maximax: Choose the alternative that offers the best
possible (highest) pay-off. The highest pay-off is Birr 140
million with Z Co. Therefore choose Z Co.
b) Maximin: For each alternative, locate the worst possible
pay-off. Then choose the best of these “worst” values. The
best of the worst is X Co. with a minimum profit of Birr 10
million.
c) Laplace: Find the average pay-off for each alternative, and
choose the alternative with the best. The highest average is
Z Co. [(0 + 20 + 140)/3 = Birr 0.533 m]. Choose Z Co.
When probability values, P(X), are assigned to
the states of nature of a pay-off table, two
additional decision criteria can be considered:
maximum probability and expected monetary
value. The expected monetary value approach
utilizes the highest average, or expected value,
E (X):
E(X)=Σ [X.P(X)]
ILLUSTRATION 5: For the data in illustration 4,
assume probabilities of:
P (low demand) = 0.3, P(medium demand) = 0.5
& P(high demand) = 0.2.
Which partnering firm should be chosen under the
criterion of:
(a) Maximum Probability and
(b) Expected Monetary Value (EMV)?
SOLUTION
a) Maximum Probability: Find the most likely state of nature, and
choose the best alternative within that state. Most likely is
[P(medium demand) = 0.5]. Choose X Co., where pay-off is Birr 50
m.
b) Expected Monetary Value: For each alternative, multiply the value of
each possible outcome (X) by its probability of occurrence, P (X).
Then sum across all outcomes, i.e., ∑ [X. P(X)] to get the expected
value, E(X), of each alternative. The alternative with the highest
expected outcome is designated the expected monetary value, EMV.
E(X Co.) = 10(0.3) + 50(0.5) + 70(0.2) = 42
E(Y Co.) = –10(0.3) + 44(0.5) + 120(0.2) = 43
E(Z Co.) = 0(0.3) + 20(0.5) + 140(0.2) = 38
EMV = Birr 43 m
9.6 DECISION SUPPORT SYSTEM
Decision support system (DSS) is computer-based systems
designed to aid decision-makers of any stage of the decision
process in the development of alternatives and evaluation of
possible course of action. Their purpose is to provide the
information and analytical support that enables managers to
better control and guide the decision process. Emphasis is
given for giving useful information and appropriate quantitative
models that support the manager’s skills. Thus, DSS are a
logical extension of the managerial decision processes. This
helps the managers to learn better, how to apply data
processing and modeling capabilities of computers to the
analysis of ill-structured and value based decisions.
9.7 ECONOMIC MODELS
Break-even Analysis is an economic model describing
cost-price-volume relationships. It is a complete certainty
type of model because costs and revenues are known
quantities.
9.7.1 Break-even Analysis
One of the techniques to study the total cost, total revenue
and output relationship is known as Break-even Analysis.
‘A Break-even Analysis indicates at what level of output,
cost and revenue are in equilibrium’. In other words, it
determines the level of operations in an enterprise where
the undertaking neither gains a profit nor incurs a loss.
NOTATIONS AND TERMINOLOGY AS USED IN BREAK-EVEN
ANALYSIS
Break-even Chart (BEC): It is a graph showing the variation in
total costs at different levels of output (cost line) as well as the
variation in the total revenues at various levels of output.
Break-even Point: It is that point of activity (sales volume) where
total revenues and total expenses are equal. It is point of zero
profit, i.e. stage of no profit and no loss. BEP can be used to study
the impact of variations in volume of sales and cost of production
on profits.
Angle of Incidence: It is an angle at which total revenue line
intersects total cost line. The magnitude, of this angle indicates the
level of profit. Larger the angle of incidence, higher will be the
profits per unit increase in sales and vice versa.
Margin of safety: It is excess of budgeted or actual sales over the
break-even sales volume i.e. Margin of Safety = (Actual Sales Minus
Sales at BEP)/Actual Sales. A high margin of safety would mean
that even with a lean period, where sales go down, the company
would not come in loss area. A small margin of safety means a
small reduction in sale would take company to cross BEP and
come in red zone.
CALCULATION OF BEP
Relationship between costs and activity level (AL) is also assumed
to be linear. For every elemental cost, actual cost figures at
different activity levels are plotted, and by ‘least square analysis’ a
‘line of best fit’ is obtained. This would give a fixed cost component
and a variable cost component for the elemental cost.
This analysis is carried out for all elemental costs. The
total cost function would give total fixed cost and total
variable cost for the company. The Break-even Point is
that volume where the fixed and variable costs are
covered. But no profit exists. Thus at BEP, the total
revenues equal to the total costs.
If F = Fixed Costs, which are independent on quantity
produced,
a = Variable Cost per unit
b = Selling Price per unit
Q = Quantity (Volume of output)
The total costs are given by
Total Cost (TC) = Fixed Cost + Variable Cost
TC = F + a⋅ Q
Sales Revenue (SR) = Selling price per unit × Quantity
SR = b.Q
The point of Intersection of Total Cost line and the sales
revenue is the Break-even Point i.e. at Break-even Point,
Total Cost (TC) = Sales Revenue (SR)
F+a⋅Q=b⋅Q
F = Q (b – a)
Q = F units ………….. 1
b a

Profit volume ratio (PVR) is defined as the ratio between


Contribution Margin and Sales Revenue.
i.e. Profit Volume Ratio (Ф) = Contribution Margin
Sales Revenue

= Sales Revenue – Total Variable Cost


Sales Revenue
Fig. 5.3 Break – even Chart
Margin Of Safety (MOS) is defined as the ratio between Operating
Profit and Contribution Margin. It signifies the fractional reduction
in the current activity level required to reach the break-even point.
Sales Turnover (STO) is defined as ratio between Sales Revenue
and the Capital Employed. It represents the number of times
capital employed is turned over to reach the sales revenue level
that is called Operating Management Performance [OMP].
IMPROVING OMP
A company interested in improving its OMP will have to improve
its operating profit. Following any of the strategies given below or
a combination of them can do this:
(a) By reducing variable costs
(b) By reducing fixed costs
(c) By increasing sales price
(d) By increasing the activity level.

(a) A reduction in variable costs will bring


down BEP, increase PV ratio and increase
margin of safety. To achieve a required
Targeted Profit (Z), variable cost would have to
be controlled at:
V=SR – (F+Z )
Fig. 5.4 Break-even Chart
(b) A reduction in fixed costs will bring
down BEP and increase margin of safety. It
will have no effect on PV ratio. To achieve
a required TP by controlling fixed cost
alone, the fixed cost would have to be
controlled as:
F = (SR – V) – Z
Fig. 5.5 Break-even Chart
(c) An increase in selling price will bring BEP
down, it will increase PV ratio and it will also
increase the margin of safety. To get the
targeted profit level the increase required in
selling price is given by:
Fig. 5.6 Break-even Chart
(d) An increase in activity level will of course have no effect on BEP,
it will not change PV ratio, but will increase the margin of safety.
The new activity level required to achieve the desired TP is given
by:

If now is the existing activity level, the activity level required in


terms of number of units Nnew to achieve a targeted profit is given
by:

With the help of this analysis, discreet decisions regarding control


of variable costs, fixed costs, fixing sales price and activity level can
be taken to achieve the desired targeted profit.
Dear Student !
Please, use the hard copy (11
pages) for the continuation of this
lecture note on Economic &
Statistical Models as well as on
Decision Trees!!!

You might also like