Sensitivity Analysis Calculator

Sensitivity analysis is a quantitative risk analysis technique that examines how variations in individual input variables affect a project's overall outcome. The PMBOK Guide positions sensitivity analysis within the Perform Quantitative Risk Analysis process as a tool that helps project managers identify which risks and uncertain variables have the greatest potential impact on project objectives. By systematically varying one variable at a time while holding all others constant, sensitivity analysis reveals the hierarchy of risk drivers that should command the most management attention.

The most common visual output of sensitivity analysis is the tornado diagram -- a horizontal bar chart that ranks variables from most to least sensitive. The widest bars at the top of the tornado represent the variables whose variation causes the largest swing in project outcomes. These are your critical risk drivers, and they deserve the lion's share of your risk mitigation budget and monitoring effort. Variables at the bottom of the tornado, with narrow bars, have relatively minor influence and can be managed with lighter-touch controls.

There are two primary approaches to sensitivity analysis: one-at-a-time (OAT) analysis, which varies a single variable while holding all others at their base values, and multi-variable analysis, which varies multiple inputs simultaneously to explore interaction effects. OAT analysis is simpler and produces tornado diagrams directly. Multi-variable analysis captures more complexity but requires tools like Monte Carlo simulation. For most project management applications, OAT sensitivity analysis provides sufficient insight for prioritizing risk responses.

The Sensitivity Index expresses how much the project outcome changes as a percentage of the base case when a variable swings from its pessimistic to its optimistic value. A high sensitivity index means that variable has a dramatic effect on the outcome. A low index means the outcome is relatively stable regardless of that variable's value.

The Base Case represents the most likely project outcome with all variables at their expected values. The Optimistic Outcome assumes the variable is at its most favorable value, while the Pessimistic Outcome assumes the least favorable value. The absolute difference between these extremes, divided by the base case, yields the sensitivity percentage that drives the tornado diagram ranking.

• Confusing sensitivity analysis with scenario analysis -- Sensitivity analysis varies one variable at a time. Scenario analysis varies multiple variables simultaneously to create coherent "what-if" stories (best case, worst case, most likely). They are complementary but distinct techniques.
• Setting unrealistic bounds -- If optimistic and pessimistic values are too extreme, every variable appears highly sensitive. If bounds are too narrow, important risks are missed. Use data-driven estimates, not gut feelings.
• Ignoring variable interactions -- OAT sensitivity analysis assumes variables are independent. In reality, material costs and labor costs may be correlated. For projects with significant variable interactions, consider supplementing with Monte Carlo simulation.
• Over-mitigating low-sensitivity variables -- A common mistake is to apply equal risk mitigation effort across all variables. Sensitivity analysis exists precisely to tell you where to focus. Follow the tornado diagram's ranking.

Sensitivity analysis and tornado diagrams are explicitly listed as tools in the PMBOK Guide's Perform Quantitative Risk Analysis process. On the exam, you should immediately associate these terms with quantitative risk analysis, not qualitative. Qualitative analysis uses probability-impact matrices and risk categorization, while quantitative analysis uses numeric techniques like sensitivity analysis, expected monetary value, decision tree analysis, and Monte Carlo simulation.

Expect questions that show you a tornado diagram and ask which variable deserves the most management attention. The answer is always the variable at the top of the tornado -- the one with the widest bar. Also be prepared for questions about the relationship between sensitivity analysis and contingency reserves: variables with high sensitivity indices should receive proportionally larger contingency allocations.

Know that sensitivity analysis is a diagnostic tool, not a prescriptive one. It tells you which variables matter most, but it does not tell you what to do about them. The next step after sensitivity analysis is developing targeted risk responses for the high-sensitivity variables -- avoidance, mitigation, transfer, or acceptance, depending on the nature of the risk and your organization's risk appetite.