Excellence: Learning from Great Procurement Teams
Quality in the Crossfire: Legal Issues in Product Assurance of Bullet Resistant Vests
by Richard Pennington
During the summer of 2003 in Forest Hills, Pennsylvania, a bullet resistant vest made by Second Chance Body Armor, Inc – its Ultima® armor, that is made of multiple layers of fabric woven from Zylon® yarn – suffered a bullet penetration, injuring a police officer.1 The incident was the first case reported to the National Institute of Justice (NIJ) in which NIJ-compliant body armor appears to have failed to prevent penetration from a bullet it was designed to defeat. This article examines the use of the bullet resistant vest standard in assessing the acceptability of delivered products and ties in the key legal concepts relevant to managing contract performance and exercising appropriate remedies.
As Second Chance – an ISO 9001 certified company – characterized the issue in its October 17, 2003 press release, the synthetic yarn – that was manufactured by Toyobo Corporation of Japan – was flawed. Second Chance publicly attributed its Chapter 11 bankruptcy filing to the defect in the yarn, making this a classic issue of product assurance that intersects quality, procurement, and legal professions.
Procurement and Product Assurance
From a procurement perspective, the objective of product assurance is to achieve an acceptable quality. The overarching issue in product acceptability is the obligation of the buying organization to pay for the product and, as is the case here, the right to demand remediation when a product appears to be defective or performance apparently degrades. In the United States, most product acceptance rights are governed by the Uniform Commercial Code (UCC); routine purchase transactions in goods typically do not involve complicated contracts that provide more remedies. The Uniform Commercial Code essentially permits reasonable inspection at the time of delivery and rejection of goods if there is not “perfect tender,” in other words, the goods do not exactly meet the specifications. In the case of accepted goods – the case with Zylon® yarn vests from any manufacturer that are in service – the Uniform Commercial Code permits “revocation” of acceptance if a “nonconformity substantially impairs its value” after acceptance if “acceptance was reasonably induced either by the difficulty of discovery before acceptance or by the seller’s assurances.”2 It is common for commodity purchases to be governed by the UCC rights of inspection, rejection, and revocation of acceptance. Of course, contracts can change these rights. The federal government has an analogous contract provision required by the Federal Acquisition Regulations that defines the “finality” of acceptance and permits revocation of acceptance in the case of latent defects.
NIJ recognizes in its user guide the wisdom of defining inspection and acceptance protocols in large quantity purchases, but NIJ implies in its guide that routine inspection of vests by testing is inadvisable. “As a guideline, an agency should test extensively only when purchasing a significant quantity of armor. Armor testing is expensive, and departments must plan their actions based on their circumstances.”3
Why is routine testing inadvisable? Because the attributes that define nonconformity here, other than some workmanship characteristics that can be evaluated by visual inspection, are measured by destructive performance testing in accordance with a prescribed testing protocol that uses a sample of six units. Requiring NIJ testing in conjunction with each delivery of a vest that costs several hundred dollars would be cost prohibitive. So what NIJ offers the law enforcement community is a qualified product list approach by publishing products on its Personal Body Armor Consumer Product List (CPL) that meet various threat conditions (e.g. handgun versus rifle).4
Qualified Products Lists
A qualified product list is a common procurement technique, requiring prequalification of suppliers – often using a testing or certification process – in order for the supplier to be eligible to bid. In the case of bullet resistant vests, manufacturers have models tested under the NIJ test protocol.5 Figure 1 shows a typical test range configuration. Vests are tested after a defined “wet conditioning” procedure because there is known degradation in performance as material yarns get wet. After 12 shots at front and back panels in defined sequence (Figure 2), the test establishes performance acceptability in two dimensions. The sample may not suffer a single penetration (the most serious failure) from a shot under specified test conditions that include types of bullets, velocity, specified environmental conditions, and calibration of the clay backing material used to measure the “back face signature” depth. Nor may the back face signature (BFS) measurement, the amount of material deformation that would occur close to the officer’s body, exceed a BFS limit for those bullets that do not penetrate the armor. This is to limit the severity of trauma injury to an officer from impact though the bullet did not penetrate.
Here is the first challenge for buyers. If a product specification on a purchase order simply orders a vest that is “NIJ certified,” as is commonly done,6 then that attribute is satisfied at the moment of delivery by a model having been found compliant with the standard, permitting the vendor to submit its manufacturer’s certification. Suppose a department with Zylon® yarn vests in the field takes six vests off the shelf and has them tested by the NIJ testing lab using the wet conditioning tests. What if only one suffers an unacceptable test penetration? Can all the unopened vests in the department’s warehouse be rejected or acceptance revoked? Because the attribute of “certification” was the only specified product requirement, and it was met at the time of delivery, there is a legitimate question about the legal right to revoke acceptance. There may be a tenuous warranty theory under the UCC implied warranties of merchantability or fitness for particular purpose, but many contracts disclaim implied warranties, and in any event the legal content of those warranties is difficult to prove.
Prequalification Versus In-Service Performance Testing
Second Chance’s Ultima® vest was certified as compliant with the NIJ standard. In a pure legal sense, unless those contracts incorporated product performance standards during the service life of the vest, the legal contractual rights may be limited.7
But many manufacturers sell personal body armor with a five-year warranty. According to NIJ Guide 100-01, a warranty means – not that vests in service will pass the wet conditioning tests applicable to new vests – but that they will continue to meet the “ballistic limit” (BL) criteria during the warranty period specified in the agreement.8 The BL test is different from the wet conditioning compliance test. A series of multiple shots are taken at the back and front of the vest to force a penetration failure. The range of bullet velocities at which the bullets achieve “partial” and “compete” penetration are averaged and the BL limit established. This BL represents the velocity at which penetration is expected to occur 50 percent of the time. The BL is considerably higher than the shot velocities used in the wet conditioning compliance test. The BL also differs with each manufacturer’s vests: BL is a baseline from which in-service performance of each test can be evaluated. Generally, if a vest’s performance is called into question, that vest’s BL – upon retesting – must remain at a velocity within three standard deviations of the BL velocity established for that model during original compliance testing.
When the Ultima® vest penetration occurred, Second Chance offered a warranty remediation approach that included replacement of the vests with another model. Had law enforcement agencies been able to prove that the ballistic limits of the vests during their warranty period were not met, there may have been a legal basis to revoke acceptance and pursue UCC or contractual remedies. Because the NIJ standard, however, does not specify a sampling protocol with which to make the revocation decision, the legality of the revocation would have involved factual questions such as, “how many BL failures of which samples would be sufficient to conclude that an entire lot of Zylon® vests could be rejected?”9
Of course, product assurance during the service life of the vest requires testing, or what the NIJ calls “retesting.” But the availability of retests became questionable when Addendum B to the standard was issued in 2003. Previously, NIJ Standard 0101.04, paragraph 5.22, had described its retest as the standard procedure for resolving ballistic performance issues with compliant body armor models.”10 However, Addendum B revised the language to make the retest totally discretionary with NIJ, calling into question just what the appropriate contractual method is for resolving performance issues during service life of the vest. As it stands now, law enforcement agencies have to develop the contractual language to define in-service testing and revocation of acceptance protocols. There is no proposed language in NIJ Guide 100-01 to provide guidance on how to specify in-service remedies.
Procurement Strategies to Answer Questions About In-Service Vest Performance
Compliance certification represents a finding of compliance with the NIJ standard at a single point in time. In May 2004, in response to questions about the warranties and service life of vests, the State of Colorado initiated a multi-state cooperative procurement aimed at validating the acceptability of production vests against the criteria in the NIJ standard. The expectation was that actual finished products, e.g. “out of the box,” would be submitted for testing. When manufacturers are seeking NIJ compliance certification of models to be listed on NIJ’s qualified products list, they submit their six samples to the certified NIJ testing lab, but the method of sample selection is ambiguous. The Colorado solicitation was intended to eliminate the effect of variation between design models and actual production models. Industry’s willingness to participate in the Colorado solicitation implies that there is some confidence that manufacturers can adequately control production variation so the results on production vests are the same as results on vest “samples” submitted in the original NIJ compliance tests.
Lingering Questions About “Models”, Samples, and Size
The Colorado Verification Test (CVT) uses the NIJ criteria for pass/fail but emphasizes the testing of completed products. Confusion remains about how NIJ’s vest samples historically have been selected, how the samples compare to actual production vests, the physical size of each vest sample, and the significance of differences in sizes of vests used for testing. The NIJ standard does not clearly define expectations with respect to how the prequalification sample “models” have to be selected. Paragraph 5.9.2 of the standard says that “all armor samples will be tested in their final design and end use configuration,” suggesting that tested models can be pre-production samples. Paragraph 4.4 of the standard, Traceability, requires that manufacturers will submit to or have on file with NIJ documentation of the method(s) they use to assure configuration control, uniformity of production methods, and materials traceability.
In March 2005, NIJ issued a “clarification” (effective April 18, 2005) addressing the protocol for getting representative test sample sizes. According to NIJ, the standard was not changed, only clarified. The “clarification” added two elements not appearing in the standard itself. First, the national office of the National Law Enforcement and Corrections Technology Center (NLECTC), not the laboratories, now mark the six shot locations and compare the samples with the newly released “sizing templates.” Second, the clarification added new language expressing the intent to insure that “the sample submitted for testing is representative of the final product manufactured for sale to public safety agencies.” [emphasis added] Coming on the heels of Colorado’s experiences attempting to use a final product sampling method, this development may eliminate any variance between the sample tested and actual production model. But as lawyers who litigate the meanings of documents can attest, industry and trade practices – as well as prior course of dealing – are sometimes relevant to resolving ambiguities. During litigation with respect to products tested under the pre-clarification protocol, testimony about a historical practice on selecting and sizing the test sample might be considered by a court in applying the standard to a product acceptability decision.
This is an area of “clarification” that may require more analysis. The good news about this debate is the visibility is has created. The wet conditioning compliance test evaluates vests in conditions more austere than are commonly encountered in field conditions. The bullet velocities reported in ballistic limit benchmarks for post-acceptance testing show built-in safety margins as well.
Figure 1: Test Range Configuration
Figure 2 – Armor Panel Test Impact Locations
 Status Report to the Attorney General on Body Armor Safety Initiative Testing and Activities, NCJ 207605, December 27, 2004, http://www.ncjrs.org/pdffiles1/nij/207605.pdf.
2 Section 4-2-608, Colorado Revised Statutes. Colorado like most other states has adopted the Uniform Commercial Code.
3 Selection and Application Guide to Personal Body Armor, NIJ Guide 100-01 (National Institute of Justice and Office of Justice Programs, U.S. Department of Justice: November 2001), pp. 41, 62.
4 As a service to law enforcement, corrections, and manufacturers, NIJ’s body armor compliance testing program tests body armor using independent testing laboratories to determine compliance with the requirements of NIJ Standard 0101.04. The models that comply with the requirements of this NIJ standard are added to its Personal Body Armor Consumer Product List (CPL), which is widely distributed to law enforcement agencies as a procurement aid. NIJ Guide 100–01.
5 The tests described in this article and used in the Colorado Verification Test (CVT) are for threat levels IIIA and below. Generally, standards for these threat levels protect against handguns and not rifles. Consult NIJ Guide 100-01 for guidance on selecting appropriate body armor for the types of threats that will be encountered.
6 The NIJ Guide suggests the following contract provision, “The body armor model shall be tested by NLECTC and found to comply with all requirements of NIJ Standard–0101.04 (or NIJ Standard–0115.00) [for stab resistance]. It shall be of Type (specify appropriate threat level and test ammunitions) as defined in that standard, and shall afford full protection to the torso front, torso back, and sides.” NIJ Guide 100-01, p. 53.
7 NIJ Guide 011-01 urges caution here as well, “Unless the department’s purchase contract clearly addresses testing armor in service, lists the tests that will be conducted, and specifies the department’s recourse should armor fail tests, NIJ recommends that the department carefully study its situation before proceeding.” NIJ Guide 100-01, p. 64.
8 The NIJ User’s Guide suggests good language to clarify that this is a performance specification, “Each unit of armor provided under this contract shall be warrantied for a minimum of (number spelled out) (number) years to meet the ballistic-resistant and deformation requirements of NIJ Standard–0101.04 (or NIJ Standard–0115.00 for stab-resistant models).” NIJ Guide 100-01, Appendix D.
9 “If the armor fails the test, the department should not automatically assume that all of the vests of that particular model owned by the department are unsafe. Rather, this suggests that these particular used vests have questionable protection capabilities. The agency may want to consider conducting additional testing of other units of this model from the same material production lot number, which should be indicated on the ballistic panel label. This testing will help determine if the failure was an isolated one or is representative of the entire purchase lot. If further testing results in additional failures, all vests from that lot of material should be replaced.” NIJ Guide 100-01, p. 63.
10 NIJ Standard 0101.04, Paragraphs 3.25 and 5.22.
Copyright © 2004-2018 Richard Pennington. All Rights Reserved. Email SeeingExcellence@gmail.com.