Schedule Risk Analysis and Contingency
This domain tests quantitative and qualitative schedule risk analysis, contingency derivation, and the difference between risk reserve and hidden padding. CAPES expects you to use three-point estimates, Monte Carlo simulation, sensitivity analysis, and confidence levels to set a risk-adjusted completion date. You should also know how correlation and path convergence affect the result.
Core concepts
- The risk register identifies threats and opportunities; qualitative analysis ranks them; quantitative analysis models their effect on duration.
- Three-point estimates feed PERT or Monte Carlo simulation. Monte Carlo samples activity durations and logic outcomes many times to produce a distribution of completion dates.
- Sensitivity analysis, often shown as a tornado chart, identifies which activities or risks drive the most variance.
- Contingency or schedule reserve is the difference between a deterministic date and a chosen confidence level, such as P80. It is managed centrally, not buried in activity durations.
- Joint confidence level (JCL) considers cost and schedule confidence together.
Worked example
A deterministic CPM finish is day 100. Monte Carlo simulation with three-point durations and documented risks gives P50 = day 104 and P80 = day 112. If the organization adopts P80, the schedule reserve is 12 days (112 − 100), and the risk-adjusted commitment is day 112. The reserve is not added to every activity; it is held at project level and drawn down through change control. If two critical paths converge on a single milestone, correlation can widen the distribution, so the P80 may move later than a simple independent-activity calculation suggests.
Traps that cost marks
Treating the deterministic date as a P50 commitment; adding contingency to every activity; ignoring correlation among durations; and using Monte Carlo on a schedule with poor logic. Also, a tornado chart ranks sensitivity, not probability; the largest risk may be unlikely but severe.