Decision Analysis

Basics of Decision Analysis

Decision Analysis refers to the process of making decisions based on research and systematic modelling of tradeoffs.
Contexts in which decisions are made include certainty, risk, and uncertainty.

A decision matrix is a list of values in rows and columns that allows an analyst to systematically identify, analyse, and rate the performance of relationships between sets of values and information.
Elements of decision matrix are weighted according to the importance of the criteria.
Maximax (optimist’s criterion), Maximin (pessimist’s criterion), Laplace (principle of insufficient reason), and Minimax regret (minimising the maximum regret) are criteria used to rank options in decision matrix.

Decision trees are diagram that display an algorithm and includes every possible outcome of a decision.
Decision trees are commonly used in operations research and decision analysis to identify a strategy that would most likely reach a goal.

Expected Value of Perfect Information (EVPI), and Expected Value of Sample Information (EVSI) helps calculate the value of carrying out further considerations or seeking more information before making a decision.
Maximum acceptable cost of further information is the value obtained by subtracting the expected monetary value of a decision from the EVPI.

In situations of uncertainty, utility theory is a method of considering the fact that different decision-makers may have different attitudes to risk.
It is based on the assumption that the decision-maker will choose the option with the greatest expected utility, rather than the greatest expected monetary value.