Integrated Risk — Combined Parametric & Expected Value

What combined parametric + EV is

This is the first of the module's "method-combining" lessons, and it rests on a key distinction you've met in pieces: project risk comes in two flavours. Systemic risk is the broad uncertainty from how mature, complex, and well-defined the project is — best predicted parametrically from history. Project-specific risk comprises the discrete events on your register — best handled by expected value. Combining them gives a fuller picture than either alone.

Systemic vs project-specific risk

• SYSTEMIC → PARAMETRIC — Broad uncertainty from the project's nature: Immature scope definition; Complexity and novel technology; General estimating inaccuracy; The "unknown unknowns"
• SPECIFIC → EXPECTED VALUE — Discrete, identified events on the register: A particular permit may be delayed; A key supplier may fail; A specific design may need rework; The "known unknowns"

Avoiding double-counting

Combining two methods raises one serious hazard: if a risk is captured by both the parametric model and the expected-value register, its contingency gets counted twice — inflating the total. This is the central technical challenge RP 113R-20 addresses.

Done right, the two methods partition the risk cleanly: parametric handles the systemic average, expected value adds only the project's distinctive risks on top. The total is complete without overlap — neither understated (a gap) nor inflated (a double-count).

Combining methods well

Classify your risks: systemic ones go to the parametric model, specific ones to the expected-value register. Crucially, document what the parametric model already includes, and put only genuinely additional, project-specific risks on the register. Sum the two for a complete contingency, and apply the same logic to schedule risk for the integrated time reserve. Reconcile the result against a single-method estimate as a sanity check.

Nine things to remember

1. 113R-20 combines parametric and expected-value methods.
2. Project risk splits into systemic and project-specific.
3. Parametric handles systemic risk — the unknown unknowns.
4. Expected value handles specific risk — the known unknowns.
5. Each method covers the other's blind spot.
6. Double-counting is the central hazard when combining methods.
7. Add only risks beyond what the parametric model already covers.
8. Done right, the methods partition risk cleanly — complete, no overlap.
9. The systemic-vs-specific split organizes all integrated methods.