3. Regulatory and internal compliance drivers

Crash simulation models are not validated once and then trusted forever. A CAE model is a calibrated approximation of a specific physical vehicle configuration, and its credibility is bounded by the configuration it was built against. When a design change alters the structure, mass distribution, restraint behavior, or material stack-up that the model was correlated to, the model's predictive validity is no longer guaranteed. Requalification — the structured re-establishment of model credibility after a change — is therefore not a purely technical preference. It is driven by external regulations, consumer rating programs, and internal engineering governance, each of which imposes its own trigger conditions, evidence requirements, and acceptance thresholds.

This chapter examines the three principal compliance drivers that force requalification: U.S. Federal Motor Vehicle Safety Standards (FMVSS), the European New Car Assessment Programme (Euro NCAP), and OEM-specific internal requirements. The key engineering insight is that these drivers do not merely say "revalidate after changes." They define which changes matter, what evidence is acceptable, and how much deviation is tolerable — and those definitions differ substantially between the three.

FMVSS: Certification by Test, Model as Supporting Evidence

FMVSS is a self-certification regime. The manufacturer certifies that each vehicle meets the applicable standard, and the National Highway Traffic Safety Administration (NHTSA) enforces compliance through audit testing of production vehicles. The regulatory obligation therefore attaches to the physical vehicle, not to the simulation model. This distinction shapes how FMVSS drives requalification.

The most consequential FMVSS standards for crash CAE are FMVSS 208 (occupant crash protection, including frontal rigid barrier and offset deformable barrier tests), FMVSS 214 (side impact, including the moving deformable barrier and pole tests), FMVSS 216a (roof crush), and FMVSS 301 (fuel system integrity). For each, the compliance evidence is a physical test result compared against a pass/fail criterion — for example, a Head Injury Criterion (HIC) limit, a chest deflection limit, or a femur load limit.

Where simulation enters is as a certification support tool: CAE is used to predict margin, screen design alternatives, and justify why a physical test is expected to pass. When a design change is introduced, the internal compliance question becomes: does the existing validated model still predict the changed configuration with sufficient fidelity that we can rely on it to certify margin, or must the model be requalified before it is used as certification evidence?

A practical FMVSS-driven requalification trigger set typically includes:

• Changes to the frontal structure load path (crash boxes, rails, bumper beam, shotgun) that affect the deceleration pulse used to drive the occupant restraint model.
• Changes to restraint system parameters — airbag volume, venting, tether configuration, seat belt geometry, pretensioner and load limiter characteristics — because these directly alter occupant kinematics and injury metrics.
• Mass changes exceeding an internally defined threshold (commonly 2–5% of curb mass, or any change to the center of gravity that shifts the pitch behavior in a barrier impact).
• Changes to the body-in-white material grades or gauge in energy-absorbing members, which invalidate the calibrated material and failure models.

Euro NCAP: Rating-Driven, Protocol-Versioned, and Continuously Tightened

Euro NCAP operates differently. It is a consumer rating program, not a legal mandate, but its commercial influence is such that OEMs treat its protocols as binding requirements. Two features make Euro NCAP a particularly strong requalification driver.

First, Euro NCAP protocols are versioned and periodically revised. Each protocol release changes barrier types, impact speeds, dummy variants (for example, the transition to the THOR dummy and the addition of far-side and post-crash assessment), and scoring thresholds. A model validated against one protocol year may not be valid for the next, even with no vehicle design change at all. Requalification is therefore triggered by protocol revision as well as by design change.

Second, Euro NCAP assessment is increasingly simulation-informed. While the star rating is based on physical tests, OEMs use CAE extensively to optimize for the rating and to predict scores before test. Because the rating is a continuous score rather than a binary pass/fail, small model errors can flip a star rating. This raises the fidelity bar for the model and makes the tolerance for post-change model drift much tighter than under FMVSS.

Typical Euro NCAP requalification triggers include any change affecting the frontal offset deformable barrier (ODB) and mobile progressive deformable barrier (MPDB) compatibility assessment, changes to the side impact barrier interaction, and any change to the pedestrian protection active bonnet or bumper geometry. Because MPDB explicitly evaluates compatibility between vehicles, a change to front-end stiffness can alter the compatibility score even if the vehicle's own occupant results are unchanged — a subtle trigger that is easy to miss.

OEM-Specific Requirements: The Tightest and Most Granular Layer

OEM internal requirements are usually the most demanding of the three, because they are written to protect against the specific failure modes the OEM has encountered. They typically take the form of a CAE validation standard or a corporate technical specification that defines model correlation criteria, change classification, and requalification scope.

A representative OEM change-classification scheme distinguishes three tiers:

The correlation criteria themselves are usually quantitative. A common OEM standard requires, for each validated load case, that the simulated occupant injury metrics fall within a defined percentage band of the physical test (for example, HIC within ±10%, chest deflection within ±8%), and that the structural deceleration pulse correlate with an objective metric such as a normalized integral difference or a peak-value and timing match. These thresholds are what convert "the model looks reasonable" into an auditable pass/fail.

How the Three Drivers Interact

The three drivers are not independent. A single design change — say, a 3 mm gauge reduction in the front rail — can simultaneously trigger an FMVSS 208 margin re-check, a Euro NCAP MPDB compatibility re-assessment, and a Class B OEM requalification. The requalification plan must satisfy the strictest applicable requirement, and the evidence must be traceable to each driver separately, because the audit questions differ: NHTSA asks whether the certified vehicle meets the standard; Euro NCAP asks whether the rating is defensible; the OEM asks whether the internal validation standard was followed.

Building the Requalification Trigger Assessment

In practice, the compliance drivers are operationalized through a change-trigger assessment performed at the point of design release. The assessment takes the engineering change notice (ECN) as input and produces a requalification decision with documented justification. A workable structure is:

1. Characterize the change. Record the affected parts, the nature of the change (geometry, material, mass, restraint parameter), and the magnitude. For the front rail gauge example, record the original and new gauge, the affected length, and the resulting mass delta.
2. Map the change to load cases. Determine which validated load cases the change can influence. A front rail gauge reduction affects the frontal ODB and MPDB cases and the FMVSS 208 frontal barrier pulse; it does not affect the side pole case. This mapping is the core engineering judgment and should be recorded explicitly.
3. Classify against each driver. Apply the FMVSS trigger set, the Euro NCAP protocol-revision and compatibility triggers, and the OEM change class. Where classifications conflict, the strictest governs.
4. Define the requalification evidence. Specify which physical test data the model will be re-correlated against, which correlation metrics apply, and what the acceptance thresholds are. For a Class B change, this is typically a targeted re-run of the affected load cases with correlation verified against the existing or newly acquired test data.
5. Record the decision and its basis. The output is a signed assessment that an auditor — whether NHTSA, Euro NCAP, or an internal quality function — can follow from change to decision to evidence.

The practical consequence is that requalification after design change is a compliance activity first and a simulation activity second. The simulation work — re-meshing, re-running, re-correlating — is well understood. The harder and more consequential work is the trigger assessment: correctly determining which changes invalidate the model, against which driver, and to what evidentiary standard. Getting that assessment right is what keeps the model credible in front of regulators, rating organizations, and internal quality audits alike.