SEMIS CHAP 7 SEN ANALYSIS Flashcards
an analysis technique that works on the basis of what-if
analysis like how independent factors can affect the dependent factor and is used to predict the outcome when analysis is performed under certain
conditions.
Sensitivity Analysis
is the simplest
method of
sensitivity analysis.
One-way analysis
is a variation of one-way analysis
that allows you to change two factors at the same time and observe how
they affect the outcome.
Two-way analysis
involves changing
one factor at a time
and observing how
it affects the
outcome.
One-way analysis
is a method of sensitivity analysis
that involves creating different
scenarios based on different
combinations of factors and
assumptions.
Scenario analysis
a method of sensitivity analysis that
uses random sampling and simulation to generate many possible outcomes based on the inputs and their probability distributions.
Monte Carlo
is derivative based (numerical or analytical)
Local Sensitivity Analysis
This method is for simple cost functions, but not
feasible for complex models.
Local Sensitivity Analysis
The term ____ indicates
that the derivatives are taken at a single point.
Local
often implemented using Monte Carlo techniques. This approach uses a global set of samples to explore the design space.
Global Sensitivity Analysis
Sensitivity analysis helps in understanding which
variables or factors have the most impact on the outcome or results.
Identifies key driver
Sensitivity analysis helps in assessing the level of risk and uncertainty associated with the project or decision
Evaluates risk and uncertainty
Sensitivity analysis aids in developing robust plans by considering multiple scenarios and variations, ensuring that plans can adapt to different possible outcome
Supports effective planning
Sensitivity analysis relies on assumptions that may not accurately represent the real-world complexity and dynamics, potentially
leading to biased or incomplete result
Assumption limitations:
Sensitivity analysis requires reliable data, and if the data
is insufficient or of poor quality, the analysis may produce unreliable or
misleading outcomes.
Data requirements: