AEP-103 head injuries assessment of non-lethal projectiles

What it is

AEP-103 is a NATO Allied Engineering Publication that defines a method for assessing the head-injury risk of a non-lethal projectile when it strikes the frontal, temporal or parietal regions of the head (clause 1.1). It is not a document a company implements as a quality system: it is the technical content covered by STANREC 4744, and it is run by a test facility or laboratory working under a National Authority (clause 2.2.6), not by a manufacturer preparing a product for market.

What the assessment covers

The method applies only to a single, non-repeated strike on the frontal, temporal or parietal regions; the maxillofacial and occipital regions, and any second shot on the same region, are explicitly out of scope (clauses 1.1, 1.2.1, 1.2.2). Its tolerances were developed around two specific non-lethal munitions, the 40 mm Def-Tec eXact iMpact XM1006 and a fragment rubber projectile from the Lacroix GALIX 46 ammunition (clause 1.2.3), rather than non-lethal projectiles in general. It classifies the result against injury-severity categories running from a minimum risk of loss of consciousness up to a significant risk of fracture (clause 1.3), and it leaves the choice of assessment method open, stating plainly that a "preference for any method, presented in the document, could not be confirmed at the time this document was produced" (clause 1.1).

Two ways to run the assessment

AEP-103 gives a test facility two routes to the same kind of result, and treats them as equivalent.

The first is a physical firing test against a mechanical surrogate: a device that has to be validated and calibrated before use, with its biofidelity documented (clause 2.1.1). The surrogate's response is only trusted inside a "validated area" fixed statistically from the mean point of impact and a defined confidence radius (Annex A), and a shot only counts toward the assessment when it is "fair": inside that area, and fired from the original launching system or one a National Authority has confirmed behaves equivalently (clauses 2.1.3, 2.2.2). The test facility has to control and record impact velocity, impact angle and projectile yaw, and to justify and document any signal filtering it applies to the data (clauses 2.2.2 to 2.2.7). Validation and calibration are run to a stated minimum level of statistical confidence, using a reference projectile (clause 2.3.1).

The second is a numerical simulation: a validated numerical model of the human head and a validated numerical model of the projectile, with the models' biofidelity documented and the projectile model validated against a rigid-wall impact before it is used against the head model (clauses 3.1, 3.2). The National Authority chooses and documents the method used to validate the numerical projectile model (clause 3.3).

Either route has to be backed by a defined record: for a physical test, target and projectile identification, ambient conditions, conditioning, shot-by-shot velocity and impact data, and the calibration and validation results relied on (clause 2.3.3); for a simulation, the model identification, its validation data, and the same per-shot readings (clause 3.5).

What it does not cover

The document says nothing about what happens to a reading once it exists: no approval process, no acceptance criterion tied to a programme, and no certification of anything. It excludes repeated impacts on one region and two of the head's anatomical regions outright (clause 1.2), and its tolerances are built around two named munitions rather than non-lethal projectiles in general (clause 1.2.3). Nothing in the document tells a national programme, a customer or a manufacturer what to do with the result; it defines only how the measurement is taken.