AEP-99 thorax injury risk assessment of non-lethal projectiles

What it is

A NATO test method, not a design standard

AEP-99 is a NATO Allied Engineering Publication that sets out a laboratory method for assessing the thorax injury risk of non-lethal, non-penetrative projectiles - a round fired from a non-lethal weapon at a person's chest. The document draws its own boundary clearly at clause 1.1, where AEP-94 "assesses the penetrative impact" of a non-lethal projectile, while AEP-99 covers only the blunt, non-penetrative case. It is Edition B, Version 1, promulgated July 2021.

AEP-99 does not bind anyone by itself. It is covered by STANREC 4744, a NATO Standardization Recommendation rather than a ratified agreement, so a nation is invited to apply the method, not committed to it. Annex F, which records France's reservations against the current edition, notes that "the adoption procedures and the rules of a STANREC do not allow each nation to add its comments or reservations about the document and the assessment procedure" anywhere else, which is why that appendix exists at all. For a company that makes non-lethal projectiles, the document is a further step removed still: it reaches them only when a national test programme or a contract calls for a projectile to be assessed this way.

Who it addresses

The text speaks to the National Authority that oversees an assessment and to the test facility that carries it out, not to a manufacturer's own quality system. Clause 2.2.1 addresses "the test facility employed for the thorax injury risk assessment" directly. Clause 2.2.2 has the National Authority verifying that a substitute launch system's impact conditions match the original weapon. Clause 2.3.1 has the National Authority computing and documenting the statistical result of a surrogate validation, and clause 3.3 leaves it to the National Authority to choose and demonstrate a method for validating a numerical projectile model. A manufacturer's own instructions appear only once, as one input the test facility may consult when setting the launcher-to-target distance (clause 2.2.3).

Two ways to run the assessment

The document gives a test facility two routes to the same injury criterion, defined in clause 1.3. The first, in Chapter 2, uses a mechanical surrogate: a physical dummy thorax that must first be validated and then recalibrated before each test session, with the National Authority deciding whether a failed calibration means the surrogate needs revalidating entirely (clauses 2.3.1 and 2.3.2). An impact only counts toward the assessment if it is a "fair shot": inside the surrogate's validated area, and fired under conditions matching the original weapon system when a substitute launcher is used (clause 2.1.3). The second, in Chapter 3, uses numerical simulation instead of a physical shot: a validated model of the thorax and a validated model of the projectile, with the projectile model itself checked against a rigid wall before it is trusted against the thorax model (clause 3.2.2).

Both routes carry an extensive record-keeping requirement. For the physical route, clause 2.3.3 asks for target and projectile identification, shooting distance, date, ambient conditions, weapon and launcher details, impact location, and, for each shot, its fairness, velocity, angles, the displacement signal and the resulting injury criterion value, plus the calibration and validation results relied on. For the numerical route, clause 3.5 asks for the equivalent: model and projectile identification, validation data, and, per simulated shot, velocity, location, the displacement signal and the injury criterion value.

What it does not cover

Two exclusions are easy to miss on a first read. Clause 1.2.2 states plainly that "the event of repeated shots on a single anatomical region is not covered by this document." And the document is limited throughout to skeletal injury from a frontal, sternal impact; it says nothing about heart injury risk. Annex F's French reservation calls both limits out directly, alongside a broader objection: that the injury criterion this method relies on has not, in France's view, been shown to be valid outside the impact-speed range the underlying research covered, or for a surrogate whose realism above that range is unproven. France states that on those grounds, the current edition "will not be applied by France."

How conformance is checked

There is no certification for AEP-99, and no accredited body assesses a projectile against it. What the document describes instead is a National Authority verifying and documenting a specific test or simulation directly - checking that a substitute launch system behaves like the original weapon (clause 2.2.2), computing and documenting a validation's statistical result (clause 2.3.1), deciding whether a failed calibration means revalidating the surrogate (clause 2.3.2), and choosing how to validate a numerical projectile model (clause 3.3). That is government verification of a specific piece of testing, not a certificate a company or a product can hold.

How it references other publications

AEP-99 hands off to AEP-94 for penetrative impacts, the case this document explicitly excludes (clause 1.1). It is covered by STANREC 4744, the recommendation a nation adopts rather than ratifies, and the instrument Annex F uses to record France's reservation. Its definitions of mean point of impact and R90 are drawn from the NATO glossary that STANAG 3680 carries (clause 2.1.2). And Annex G records that AEP-99's own structure, header, chapters and table of contents were revised to match AEP-103's format, though AEP-103 is not a normative reference for the method itself.

Standards it references

• AEP-94Background
• STANAG 3680Background
• AEP-103Background