Decision-making - Types and Process: formula sheet

Key formulas

Definition

Decision-making = choosing the best alternative from two or more alternatives to achieve an objective

Check for both elements: more than one alternative and a purpose.

Key characteristics

Choice + Goal-oriented + Mental process + Continuous + Pervasive + Dynamic + Commitment

Use this list to match a statement to the correct feature.

Decision vs action

Decision = the choice made; Implementation = putting the choice into effect

A decision is complete when the choice is made, not when it is carried out.

Programmed vs non-programmed

Programmed = repetitive + rule-based; Non-programmed = new + judgement-based

The test is whether a ready procedure exists for the problem.

Level-wise classification

Strategic (top, long-term) → Tactical (middle, medium-term) → Operational (lower, short-term)

Match the decision to the management level and time horizon.

Who decides

Individual = one person, faster; Group = many people, slower but wider views

Group decisions usually gain acceptance and diverse input but cost time.

Routine vs basic

Routine = frequent, minor, procedure-based; Basic = rare, major, long-term commitment

Basic decisions are hard to reverse and need careful study.

Sequence of the decision-making process

Identify problem → Gather information → Develop alternatives → Evaluate alternatives → Select best alternative → Implement → Follow up / evaluate

Wording and the number of steps vary by book. The order stays the same. Follow-up always comes last.

Problem as a gap

Problem = Desired position − Actual position

A problem exists only when there is a gap between what should be and what is.

Rational decision-making assumption

Clear problem + known alternatives + clear criteria → choose the alternative with the best result

Rational decisions assume complete information and an objective manager. Real life often falls short of this.

Expected value (decision under risk)

EV = Σ (probability × payoff)

Probabilities of all outcomes for an alternative must add up to 1. Choose the alternative with the highest EV if payoffs are profits, or the lowest if payoffs are costs.

Maximax (optimistic) criterion

Pick the alternative whose best payoff is the highest

Used under uncertainty by an optimistic decision-maker.

Maximin (Wald, pessimistic) criterion

Pick the alternative whose worst payoff is the highest

Used under uncertainty by a cautious decision-maker.

Minimax regret (Savage) criterion

Regret = best payoff in that state − payoff of the alternative; pick the alternative with the smallest maximum regret

Regret is worked out column by column (state by state).

Laplace (equal probability) criterion

Average payoff = sum of payoffs ÷ number of states; pick the highest average

Treats all states as equally likely when no probabilities are known.

Hurwicz criterion

Weighted value = α × best payoff + (1 − α) × worst payoff

α is the coefficient of optimism, between 0 and 1. Pick the highest weighted value.

Expected value at a chance node

EV = Σ (probability × payoff)

Probabilities at one chance node must add up to 1. Use net payoff if a cost is involved.

Decision tree rule

Choose the branch with the highest expected value (net of cost)

If the payoffs are costs, choose the lowest expected cost instead.

Rational model vs bounded rationality

Rational = optimising (best option); Bounded = satisficing (good enough option)

This contrast is the most common MCQ pair.

Brainstorming vs Delphi

Brainstorming = face-to-face, ideas, no criticism; Delphi = experts, remote, anonymous, repeated rounds

Use the meeting and anonymity cues to separate them.

Value of extra information

Collect more information only if expected benefit > cost of collecting it

A practical rule, not a numeric formula. It explains why decisions are often taken with incomplete data.

Bounded rationality (satisficing)

Decision maker chooses a satisfactory option, not the best option

Idea linked to Herbert Simon. Remember the words: limited information, limited ability, limited time.

Four groups of limitations

Information + Resources (time, cost) + Human (bias, ability) + Organisational (policy, politics, structure)

Use this to classify any factor named in a question.

Quick revision

• A decision is a choice among alternatives; if there is only one option, there is no decision to make.
• Decision-making is a continuous, goal-directed process that involves a choice and a commitment to act.
• Programmed decisions are routine and rule-based; non-programmed decisions are new, unstructured and need judgement.
• Strategic decisions are long-term and made by top management; operational decisions are day-to-day and made at lower levels.
• Remember the steps of the process in order: identify the problem, gather information, develop alternatives, evaluate, choose, implement, follow up.
• Follow-up and feedback is the step that checks whether the decision worked.
• Under certainty, outcomes are known; under risk, probabilities are known; under uncertainty, probabilities are not known.
• Match each technique to its condition before choosing an option in the exam.
• Barriers include lack of information, time pressure, personal bias and resistance to change.
• No negative marking applies, so attempt every question and eliminate options that use absolute words without support.

Common mistakes

• Treating a decision as the same as its implementation. Fix: Remember that the decision is the choice. Carrying it out is a later step.
• Saying decision-making is needed only at top management. Fix: Recall pervasiveness: supervisors and middle managers also decide, with different scope.
• Treating all important decisions as non-programmed. Fix: Ask whether a rule exists. A routine payroll run is important but programmed.
• Confusing tactical with operational decisions. Fix: Tactical is medium-term and departmental; operational is short-term and day-to-day.
• Putting selection of the best alternative before evaluation. Fix: Remember that you cannot choose until alternatives are compared. Evaluate first, then select.
• Treating the decision as complete once the choice is made. Fix: Implementation and follow-up are still part of the process. A decision is judged by its results.
• Calling a decision 'uncertain' just because the future is unknown. Fix: In this topic, risk means probabilities are known and uncertainty means they are not. Check for probabilities.
• Using expected value when no probabilities are given. Fix: Without probabilities, use maximax, maximin, regret, Laplace or Hurwicz, as the question directs.
• Treating bounded rationality as irrational decision-making. Fix: Remember that the manager is still rational, but within limits of information, time and ability. The result is a satisfactory choice.
• Mixing up brainstorming and the Delphi technique. Fix: Brainstorming is a live meeting with free ideas. Delphi uses anonymous experts who never meet and answer in rounds.

Exam tips

• Learn the characteristic list as clue words, because most MCQs describe a situation and ask which feature it shows.
• Watch for options that confuse decision with plan or implementation.
• Be careful with absolute words such as 'only' and 'always' in statements about who decides.
• Link this topic to the types and steps of decisions, since questions often mix them.
• There is no negative marking, so always attempt every question after eliminating two options.
• Most questions give a short scenario. Spot the clue word first, then pick the type.
• Know the three level-wise types with their management level and time horizon; they are asked often.
• Learn one example for each type so you can recognise it in a different disguise.