Integrated Risk — Parametric + Monte Carlo of a CPM Model

What combined parametric + Monte Carlo CPM is

This is the same architecture as Lesson 19 — parametric for systemic, a second method for specific — but with the specific-risk engine upgraded from expected value to full Monte Carlo simulation of the schedule (the risk-driver approach of Lesson 16). The reward: the specific-risk side now respects the schedule network, merge bias, and correlation, and produces a confidence curve rather than a single mean. The parametric layer still supplies the systemic risk that simulation alone would miss.

How the combined analysis runs

• Parametric estimate of systemic risk — Score the project's definition, complexity, and technology to get the systemic contingency baseline (Lesson 12).
• Build the cost-loaded CPM and risk drivers — Take a sound schedule with costs, and define risk drivers for the specific risks (Lesson 16).
• Monte Carlo the schedule — Simulate the network thousands of times for specific-risk cost and schedule distributions, with merge bias and correlation captured.
• Combine without double-counting — Merge the systemic (parametric) and specific (simulated) layers, ensuring no risk is captured in both.
• Read integrated contingency at a confidence level — Get combined cost and schedule contingency to your chosen P-value — systemic plus specific, complete.

More power, more demand

This method is among the most rigorous in the AACE framework — and among the most demanding. It needs a sound, cost-loaded CPM schedule, a validated parametric model, a well-built risk register, specialist simulation software, and the expertise to combine the layers correctly. The payoff is a genuinely complete, network-aware, confidence-based integrated contingency.

Running the combined method well

Get the schedule right first (the warnings of Lesson 17 apply in full), validate the parametric model against relevant history, and define risk drivers carefully from the register. Above all, document the boundary between the systemic and specific layers so the combine step is clean. Present the integrated result as cost and schedule contingency at a stated confidence level, and reconcile against a simpler method as a sanity check.

Nine things to remember

1. 117R-21 pairs a parametric baseline with Monte Carlo of the CPM.
2. Same systemic-vs-specific architecture as Lesson 19, stronger engine.
3. Parametric covers systemic risk simulation can't see.
4. Monte Carlo CPM captures network, merge bias, and correlation.
5. It yields a confidence curve, not a single mean.
6. Double-counting is the central risk — define each layer precisely.
7. It's among the most rigorous — and most demanding — methods.
8. Match the method to the stakes — use it where the effort pays off.
9. Get the schedule right first — Lesson 17's warnings apply in full.