The Predictive Capability Maturity Model
A framework that rates how mature a simulation capability is across several axes, rather than declaring it credible or not.
Maturity, Not a Verdict
Whether a simulation is trustworthy is rarely a yes-or-no answer. The Predictive Capability Maturity Model (PCMM) frames it as maturity across several axes, each rated on a scale. Instead of stamping a prediction credible or not, it produces a profile showing where the capability is strong and where it is weak, which is far more useful for deciding how much to rely on it and where to invest.
The Axes
- Representation and geometric fidelity: how faithfully the model captures the real configuration.
- Physics and material model fidelity: how completely the relevant phenomena are modeled.
- Code verification: the evidence the software solves its equations correctly.
- Solution verification: the evidence the specific calculations have bounded numerical error.
- Validation: the evidence the model matches experiment.
- Uncertainty quantification: the completeness of the uncertainty accounting.
Reading a Profile
Each axis is rated from low, meaning little formal evidence, to high, meaning thorough documented evidence. A realistic profile is uneven: a capability may be strong on code verification but weak on validation because the experiments do not yet exist. The uneven profile is the point; it directs attention and resources to the weakest axis rather than hiding it behind an averaged overall score.
Honest Use
PCMM is a tool for honest self-assessment and communication, not a certificate. Its value depends on rating each axis by the evidence actually in hand, not by aspiration. A low rating on validation is not a failure to conceal; it is accurate information that a design decision should account for, perhaps by adding margin or by prioritizing the experiment that would raise the rating.
For designs where full-system validation is inherently future work, such as fusion systems before first operation, a maturity profile is the honest way to communicate credibility: strong where the evidence is strong, explicitly provisional where the confirming experiment has not yet run.