3. Case 03: Thorax Injury Criteria Selection for a Frontal Impact Diagnostic Study

Case context and the criteria decision problem

This case fixes a single frontal impact diagnostic scenario and asks you to choose the thorax injury criteria that belong in the analysis, then justify each choice against the study's purpose. The scenario is a 50 km/h frontal sled test with a belted Hybrid III 50th-percentile male occupant, a force-limiting retractor set to 3.5 kN, a pretensioner, and a lap-and-shoulder belt with the shoulder belt routed over the mid-clavicle. The diagnostic question is not "did the occupant pass?" but "what does the thorax loading tell us about restraint performance, and which metrics can answer that?" Those are different questions, and they map to different criteria.

The distinction that drives the whole case is diagnostic versus regulatory. A regulatory criterion exists to produce a pass/fail number against a legally defined limit under a prescribed test procedure. A diagnostic criterion exists to explain the loading mechanism, localize it in time, and support engineering decisions. A metric can be both, but only if the test conditions match the regulation's scope. In this study the sled pulse, the dummy, and the restraint hardware are all fixed by the diagnostic program, not by a homologation procedure, so any regulatory number computed here is informative but not a compliance result. State that explicitly before you present any value.

Candidate criteria and what each one actually measures

Four families are relevant to a belt-loaded frontal thorax. Each answers a different question, and selecting all four is not automatically correct — the study's purpose determines which ones earn a place.

• Chest deflection (absolute and normalized). Peak sternum-to-spine compression, in millimetres, and as a fraction of the initial chest depth. This is the primary measure of how much the chest was compressed and is the most direct link to belt penetration and rib loading. It is the workhorse diagnostic metric for restraint tuning.
• Deflection rate. The time derivative of deflection, typically the peak value in m/s. Rate separates a slow, well-distributed compression from a fast, concentrated one at the same peak deflection. Two tests with identical peak deflection but different rates imply different loading paths and different injury risk.
• Viscous criterion (VC). The product of instantaneous deflection velocity and instantaneous normalized compression, VC = v(t) · C(t)/C_max, reported as its peak in m/s. VC captures the combined effect of compression and rate and is the standard soft-tissue injury correlate for the thorax. It is sensitive to filtering and to the normalization reference, so it must be computed from the same validated deflection channel used for the peak value.
• Rib fracture metrics. Number of fractures, fracture location relative to the belt path, and — where a rib-level model or detailed post-test inspection is available — strain or deflection at specific rib levels. These are the most mechanism-specific metrics and the ones that connect a shoulder-belt contact patch to a specific injury pattern.

Selecting the criteria set for this scenario

The study's purpose is restraint diagnosis: explaining how the 3.5 kN force limiter and the shoulder-belt path shape thorax loading. That purpose sets the selection logic. Peak chest deflection is mandatory because it is the direct output of belt penetration and the quantity the force limiter most strongly controls. Deflection rate is mandatory because the force limiter's effect is partly a rate effect — a lower limit pays out belt earlier and spreads the loading over a longer window, which shows up in the rate trace even when peak deflection changes little. VC is included because it is the accepted combined compression-and-rate correlate and because it is the metric most likely to reveal a loading concentration that peak deflection alone would hide. Rib fracture metrics are included in a reduced form: fracture count and location relative to the mapped belt path, taken from post-test inspection, without a rib-level strain model, because no such model is part of this test setup.

What is deliberately excluded, and why, matters as much as what is included. A regulatory chest deflection limit is not used as a pass/fail gate, because the sled pulse and restraint configuration are diagnostic-program conditions rather than homologation conditions. A rib-level strain criterion is not used because the instrumentation does not support it. Including either would create a false impression of compliance or of mechanism resolution that the data cannot deliver.

Justification logic you must be able to defend

A defensible criteria selection is not a list; it is an argument with three parts. First, traceability: every selected criterion must be computable from a channel that exists in this test, with a stated filter and normalization reference. Peak deflection, rate, and VC all come from the same validated deflection channel; rib fracture metrics come from inspection. Second, purpose fit: each criterion must answer a question the study actually asks. Deflection and rate answer "how did the force limiter shape the loading?"; VC answers "was the loading concentrated enough to matter for soft tissue?"; fracture metrics answer "where on the rib cage did the belt path deliver its load?" Third, scope honesty: any criterion whose limit is defined by a regulation must be flagged as non-compliance in this configuration, and any criterion requiring instrumentation the test does not have must be excluded rather than approximated.

A common failure in this case is selecting criteria by reputation — including VC and a regulatory deflection limit because they are well known, without checking whether the test supports them. The check is mechanical: for each candidate, name the source channel, the processing step, and the question it answers. If any of the three is missing, the criterion does not belong in the set.

Worked selection record

The table below is the deliverable for this case: a criteria selection record that a reviewer can audit line by line. Values in the "expected role" column describe what the criterion is used for, not measured results — no test data are asserted here.

Verification before the criteria are used downstream

Before any of these criteria feed a loading diagnosis, confirm three things. The deflection channel must pass a signal quality check — stable baseline, no clipping at peak, and a filter applied consistently across deflection, rate, and VC. The normalization reference must be the measured initial chest depth for this occupant, not a nominal value. And the fracture mapping must reference the same belt contact patch documentation used in the geometric setup, so that a fracture location can be tied to a specific point on the shoulder-belt path. If any of these fails, the criteria set is not yet usable, regardless of how complete the table looks.