4 Decision Making: Concept, Process and Tools

4.1 What is a Decision?

Herbert Simon — Nobel laureate and the central figure in decision theory — described management as “synonymous with decision making” (simon1947?). A decision is a choice among two or more alternatives. Decision making is the broader process that leads to that choice — identifying the problem, generating alternatives, evaluating them, and selecting one.

Koontz and Weihrich define decision making as “the selection of a course of action from among alternatives; it is the core of planning” (koontz2010?). Stephen Robbins points out that every managerial function is, at its heart, a string of decisions — what to plan, how to organise, whom to staff, how to lead, when to control (robbins2018?).

Three Working Definitions

Author	Definition	What it foregrounds
Herbert Simon	“Decision making includes the entire process of choice — identifying alternatives, weighing them, choosing among them, and acting.”	Process
George R. Terry	“The selection based on some criteria from two or more possible alternatives.”	Criteria
Peter Drucker	“Whatever a manager does, he does through making decisions.”	Centrality

4.1.1 Characteristics

A managerial decision is goal-oriented, future-directed, involves choice, is rational (or boundedly so), commits resources, and is irreversible to varying degrees. The hallmark is that doing nothing is also a decision — to keep the present course.

4.2 Types of Decisions

4.2.1 Programmed and Non-programmed (Simon)

Herbert Simon’s most-tested distinction:

Programmed vs Non-programmed Decisions

Feature	Programmed	Non-programmed
Type of problem	Routine, repetitive, well-structured	Novel, unstructured, complex
Procedure	Established — rules, policies, standard operating procedures	Custom — judgment, intuition, creativity
Where it occurs	Lower and middle levels	Top management
Time available	Short	Longer
Information	Mostly complete	Often incomplete
Example	Reordering inventory	Entering a new country

4.2.2 Other useful classifications

Other Classifications of Decisions

Basis	Categories	Cue
Level	Strategic, Tactical, Operational	Time horizon and scope
Scope	Major, Minor	How much it commits
Persons involved	Individual, Group / Collective	Single signature vs committee
Information	Decisions under certainty, risk, uncertainty	How much we know about outcomes
Anthony’s framework	Strategic planning, Management control, Operational control	A staircase from CEO to shop floor

The certainty–risk–uncertainty trio matters most for the tools section.

4.3 The Rational Decision-Making Process

The classical, prescriptive model assumes a rational manager — fully informed, goal-clear, optimising. Robbins’s eight-step version is the most-cited (robbins2018?).

flowchart TB
  P[1. Identify<br/>the problem] --> C[2. Identify<br/>decision criteria]
  C --> W[3. Weight<br/>the criteria]
  W --> A[4. Develop<br/>alternatives]
  A --> AN[5. Analyse<br/>alternatives]
  AN --> S[6. Select an<br/>alternative]
  S --> I[7. Implement<br/>the alternative]
  I --> E[8. Evaluate<br/>effectiveness]
  E -. feedback .-> P
  style P fill:#E3F2FD,stroke:#1565C0
  style E fill:#F1F8E9,stroke:#558B2F

Eight Steps Explained

Step	What happens	Common pitfall
1. Identify the problem	Recognise the gap between actual and desired	Treating a symptom as the problem
2. Identify decision criteria	Specify what matters for this decision	Forgetting a criterion
3. Weight the criteria	Assign relative importance	Equal weights when criteria are not equal
4. Develop alternatives	List feasible courses of action	Premature shortlist
5. Analyse alternatives	Score each alternative on each criterion	Confirmation bias
6. Select an alternative	Choose the highest scorer	Selecting on the loudest voice
7. Implement	Communicate, allocate, execute	Plan without follow-through
8. Evaluate effectiveness	Did it solve the problem?	Skipping evaluation when busy

4.4 How Decisions are Actually Made

The rational model is an ideal. Three descriptive models explain how managers really decide.

4.4.1 Bounded Rationality (Simon)

Real managers face limits on information, time, computational ability and attention. They search until they find an alternative that is good enough — they satisfice rather than maximise (simon1957?). The threshold (the “level of aspiration”) is itself adjusted over time.

4.4.2 Incrementalism (Lindblom)

Charles Lindblom’s “science of muddling through” (1959) argues that real decisions move in small, marginal steps from the status quo, often with comparison limited to alternatives close to current practice (lindblom1959?). It is realistic but conservative.

4.4.3 Garbage-Can Model (Cohen, March, Olsen)

In what Cohen, March and Olsen called organised anarchy — universities, public bureaucracies — problems, solutions, participants and choice opportunities float independently and meet by chance (cohenmarcholsen1972?). The garbage-can model explains why reasonable organisations sometimes adopt a solution that pre-exists the problem.

Three Descriptive Models at a Glance

Model	Author	Core insight	Where it fits
Bounded rationality	Herbert Simon	Satisficing under cognitive limits	Most managerial decisions
Incrementalism	Charles Lindblom	Muddling through, small steps	Public policy
Garbage can	Cohen–March–Olsen	Solutions and problems meet by chance	Loose, organised-anarchy settings

4.5 Decision Making under Certainty, Risk and Uncertainty

Three Decision Conditions

Condition	What is known	Tool of choice
Certainty	The outcome of every alternative is known with full confidence	Linear programming, optimisation
Risk	Outcomes are probabilistic, with known probabilities	Expected value, decision tree
Uncertainty	Outcomes are unknown; even probabilities are unknown	Maximax, maximin, Laplace, Hurwicz, minimax-regret

4.5.1 Tools under risk

Expected Monetary Value (EMV). For each alternative, multiply each possible payoff by its probability and add. Choose the alternative with the highest EMV.

Decision tree. A graphical layout of decision nodes (squares), chance nodes (circles) and end-nodes (triangles), with probabilities on chance branches and payoffs at end-nodes. Solved by folding back (right-to-left) using EMV.

4.5.2 Criteria under uncertainty

Five Classical Criteria

Criterion	Stance	Rule
Maximax	Optimist	Choose alternative whose best outcome is best
Maximin	Pessimist	Choose alternative whose worst outcome is best (Wald’s criterion)
Laplace	Indifferent	Treat outcomes as equally likely; pick the highest average
Hurwicz	Realist	Weight the best and worst outcomes by α and (1–α)
Savage / Minimax-regret	Regret-averse	Build a regret table; minimise the maximum regret

4.6 Group Decision Making

Most managerial decisions are made in groups. The trade-offs are well-documented.

Group Decision Making — Strengths and Weaknesses

Strengths	Weaknesses
More information and knowledge	Time-consuming
More alternatives generated	Pressures to conform — groupthink (Janis)
Higher acceptance of the decision	Domination by a few
Greater legitimacy	Diffused responsibility

Three structured techniques are routinely tested:

Three Group Techniques

Technique	Originator	How it works
Brainstorming	Alex Osborn (1953)	Free generation of ideas, criticism suspended during the idea phase
Nominal Group Technique (NGT)	Delbecq & Van de Ven (1968)	Members write ideas silently → round-robin sharing → discussion → secret rank-order vote
Delphi	RAND Corporation (1950s)	Expert panel responds to anonymous, iterative questionnaires; controlled feedback between rounds

Irving Janis’s groupthink — a deterioration of mental efficiency, reality-testing and moral judgment in highly cohesive groups — is the classic warning (janis1972?). The Bay-of-Pigs invasion is the textbook case.

4.7 Common Quantitative Tools

Quantitative Aids to Decision Making

Tool	Used for	Anchor
Linear Programming	Allocating scarce resources	Operations research
Game Theory	Decisions against rational opponents	Von Neumann & Morgenstern
Queueing theory	Service-system design	Erlang
Simulation (Monte Carlo)	Complex stochastic systems	Statistics
Cost–benefit analysis	Public projects, capex	Economics
Break-even analysis	Sales–cost–profit decisions	Accounting
PERT / CPM	Project scheduling decisions	Operations

These tools are revisited in the Statistics and Operations chapter. The aim here is recognition: which tool fits which problem.

4.8 Common Decision-Making Errors

Modern behavioural-decision research identifies systematic biases that warp managerial judgment (kahneman2011?). The exam-relevant short list:

Anchoring — over-weighting the first piece of information.
Confirmation — searching for evidence that supports an existing belief.
Availability — judging probability by how easily examples come to mind.
Representativeness — judging by similarity to a stereotype.
Overconfidence — over-estimating one’s own knowledge or accuracy.
Escalation of commitment — sticking with a failing course because of past investment (the sunk-cost fallacy).
Hindsight — believing, after the event, that one would have predicted it.

4.9 Practice Questions

Q 01 Simon Easy

"Management is decision making." The view is most strongly associated with:

AHenri Fayol
BPeter Drucker
CHerbert Simon
DHenry Mintzberg

View solution

Correct Option: C

Herbert Simon — Nobel laureate (1978) — placed decision making at the very centre of administrative behaviour.

Q 02 Descriptive Models Medium

Match the model with its central idea:

(i)	Bounded rationality	(a)	Solutions and problems meet by chance
(ii)	Incrementalism	(b)	Satisficing under cognitive limits
(iii)	Garbage-can model	(c)	Muddling through, small marginal steps

A(i)-(b), (ii)-(c), (iii)-(a)
B(i)-(a), (ii)-(b), (iii)-(c)
C(i)-(c), (ii)-(a), (iii)-(b)
D(i)-(b), (ii)-(a), (iii)-(c)

View solution

Correct Option: A

Bounded rationality → Simon's satisficing; Incrementalism → Lindblom's muddling through; Garbage-can → Cohen–March–Olsen's organised anarchy.

Q 03 Programmed Easy

A decision to reorder a standard inventory item every Friday is best classified as:

AStrategic and non-programmed
BTactical and non-programmed
CProgrammed
DDecision under uncertainty

View solution

Correct Option: C

Routine, repetitive decisions handled by an established rule are programmed.

Q 04 Risk Tools Easy

Under which condition are decision-tree analysis and Expected Monetary Value most appropriate?

ACertainty
BRisk
CUncertainty
DConflict

View solution

Correct Option: B

EMV and decision trees are made for risk — outcomes are probabilistic with known probabilities.

Q 05 Uncertainty Criteria Medium

Which criterion picks the alternative whose worst outcome is the best — i.e., the pessimist's rule?

AMaximax
BMaximin
CLaplace
DHurwicz

View solution

Correct Option: B

Maximin (Wald's criterion) is the conservative pessimist's rule — maximise the minimum payoff.

Q 06 Delphi Easy

Which group-decision technique relies on anonymous, iterative questionnaires sent to a panel of experts?

ABrainstorming
BNominal Group Technique
CDelphi
DFocus group

View solution

Correct Option: C

Delphi, developed at RAND in the 1950s, uses anonymity + controlled feedback + iteration to converge expert judgment.

Q 07 Biases Medium

"Persisting with a failing course of action because of resources already invested" is the bias known as:

AAnchoring
BConfirmation
CEscalation of commitment
DHindsight

View solution

Correct Option: C

Escalation of commitment — a form of the sunk-cost fallacy — keeps a manager throwing good money after bad.

Q 08 Groupthink Easy

Groupthink — the deterioration of judgment in highly cohesive groups — was studied by:

AIrving Janis
BHerbert Simon
CCharles Lindblom
DAlex Osborn

View solution

Correct Option: A

Irving Janis's Victims of Groupthink (1972) is the classic study; the Bay-of-Pigs invasion is the textbook case.

Quick recall

Decision making = choice among alternatives. Simon: management is decision making.
Two big types (Simon): Programmed (routine, rules, lower levels) vs Non-programmed (novel, judgment, top levels).
Conditions: Certainty → LP; Risk → EMV / decision tree; Uncertainty → Maximax, Maximin, Laplace, Hurwicz, Savage.
Process (Robbins, 8 steps): Problem → Criteria → Weights → Alternatives → Analyse → Select → Implement → Evaluate.
Descriptive models: Bounded rationality (Simon, satisficing), Incrementalism (Lindblom), Garbage-can (Cohen–March–Olsen).
Group techniques: Brainstorming, NGT, Delphi. Risk: groupthink (Janis).
Common biases: anchoring, confirmation, availability, representativeness, overconfidence, escalation, hindsight.