Business Decision Analysis - Lecture 1 Vocabulary

Vocabulary flashcards defining fundamental concepts, tools, decision environments, and decision criteria under uncertainty from Lecture 1 of Business Decision Analysis.

Quantitative Decision Making

A field of study that uses computers, statistics, and mathematics to construct mathematical models of business problems for experimenting with strategies and identifying optimal decisions.

Operations Research (OR)

A scientific approach to providing quantitative bases for decisions regarding operations under management control, which originated with military planners during World War I and World War II.

INFORMS

The Institute for Operations Research and the Management Sciences; the largest professional society in the world for operations research, management science, and analytics.

Decision Analysis

A structured, analytic, and systematic approach to decision making that helps identify optimal strategies when facing multiple decision alternatives and uncertain future events.

Objective

A primary decision element representing the goal a decision maker aims to achieve, such as maximizing profits or revenues, or minimizing costs or risks.

Decision Alternatives

A primary decision element consisting of the distinct options or courses of action available to decision makers.

States of Nature

A primary decision element representing the possible future outcomes of uncertain or chance events that impact decision payoffs.

Payoff (Consequence)

The measure of the result obtained when a specific decision alternative is chosen and a particular state of nature occurs.

Influence Diagram

A compact graphical representation of a decision situation that highlights the interdependencies between decision alternatives, chance events, and outcomes.

Payoff Table

A tabular tool summarizing the payoffs associated with every possible combination of decision alternatives and states of nature.

Decision Tree

A decision analysis tool that graphically displays all decision elements, chance events, and outcomes in their chronological sequence.

Decision Environment of Certainty

A decision-making context where all relevant information is fully available and each decision alternative leads to a known single outcome.

Decision Environment of Uncertainty

A decision-making context where complete information is lacking and the probabilities of possible states of nature are unknown.

Decision Environment of Risk

A decision-making context where future states of nature are subject to chance, but their probabilities of occurrence are known.

Optimistic Approach (Maximax)

A decision-making criterion under uncertainty that identifies the maximum payoff for each alternative and selects the option with the maximum of these maximum payoffs: $\text{Select } d_i \text{ with } V = \boldsymbol{\text{max}}_i \boldsymbol{\text{max}}_j V_{ij}$.

Conservative Approach (Maximin)

A pessimistic decision-making criterion under uncertainty that identifies the worst payoff for each alternative and selects the option with the maximum of these minimum payoffs: $\text{Select } d_i \text{ with } V = \boldsymbol{\text{max}}_i \boldsymbol{\text{min}}_j V_{ij}$.

Opportunity Loss Approach (Minimax Regret)

A decision-making criterion under uncertainty that evaluates alternatives based on regret values and recommends the option with the minimum of the maximum regrets: $\text{Select } d_i \text{ with } R = \boldsymbol{\text{min}}_i \boldsymbol{\text{max}}_j R_{ij}$.

Regret ($R_{ij}$)

The difference between the best payoff achievable under a given state of nature ($V_j^*$) and the payoff achieved by a specific decision alternative ($V_{ij}$): $R_{ij} = V_j^* - V_{ij}$.

Descriptive Analytics

Analytics methods focused on understanding past or current performance by answering 'What happened?'.

Predictive Analytics

Analytics methods focused on forecasting future outcomes and performance by answering 'What is likely to happen?'.

Prescriptive Analytics

Analytics methods focused on recommending optimal courses of action by answering 'What should we do about it?'.

Stochastic Decision Models

Decision models incorporating probability concepts to evaluate problems where some data are uncertain.

Deterministic Decision Models

Decision models where all input data are known with certainty, allowing the decision maker to optimize within constraints.