January 31, 1997

P.S. Protest No. 96-22

GRAND RAPIDS LABEL COMPANY, INC.

Solicitation No. 102590-96-A-0097

Decision

Grand Rapids Label Company, Inc., (GRL) protests the award of a contract to SEI for the supply of flat mail forwarding terminals.

On April 9, 1996, the Postal Service announced the requirement in the Commerce Business Daily and asked vendors to submit prequalification statements. Of the seven vendors which responded, four, including GRL, were considered prequalified.

Solicitation 102590-96-A-0097 was issued by Automation Purchasing, Merrifield, VA, on June 25, 1996, seeking offers to supply 1,450 Flats Forwarding Terminals (FFT), each of which includes a printer, label applicator, desk, flats tray rack, chair and footrest.⁽¹⁾ The procurement was conducted in three stages. Each offeror was to present a prototype of its FFT system by August 5, after which demonstrations and tests of the prototypes were conducted separately with each offeror's system. Oral presentations were scheduled the week of August 12 and written cost proposals were due August 19.

Solicitation section J.5 B. provided in part that offerors were to "deliver a prototype FFT consisting of (1) a Printer/Label Applicator (P/LA) and a desk or section of a desk; (2) Labels and printing expendable supplies; and (3) offeror's development PC and keyboard . . . . The USPS test will evaluate: 1. P/LA Functionality . . . 2. Label . . . [and] 3. Printing Supplies. . . ."

Section J.5 3.C., Oral Presentation Instructions, stated that the "oral presentation and interview will be a test of the offeror's knowledge, ability, and competence to meet the USPS' requirements and program objectives." It also advised offerors to "structure the oral presentation in accordance with the evaluation factors included in Section M of the solicitation and address each section of the [Statement of Work (SOW)]." Offerors were required to explain in detail their "understanding, technical design and approach to meet or exceed the SOW requirements for their proposed hardware, software, and labels"; their implementation strategy; and their proposed support services.

Section K.4 informed offerors that the Postal Service intended to make award on the basis of initial proposals, without discussions. Section K.6, Alternate Intellectual Property Rights Proposals, advised offerors that the Postal Service was seeking "[t]itle to any patents resulting from contract performance"; and "[u]nlimited rights in certain data (technical data and computer software) delivered to the Postal Service during contract performance." That section allowed offerors to "propose alternate [such as licensing agreements] intellectual property rights arrangements" only for intellectual property "not first produced in performance of this requirement."

Section M.1 a. provided for award "to the responsible offeror whose proposal offers the best value to the Postal Service. . . . This will be determined by comparing the differences in the value of technical/management features with differences in cost to the USPS. Cost/Price will be considered in the award decision, although the award may not necessarily be made to the offeror submitting the lowest[] evaluated price. The USPS is more concerned with obtaining superior technical/management features than with making award at the lowest overall cost to the USPS."

The solicitation provided that the offerors would be evaluated "on the demonstrated performance of the prototype as well as the quality of the responses provided in the oral presentation. The cost/price proposals will be evaluated for reasonableness and realism." The technical evaluation factors, listed in descending order of importance, were:

I. Design and Performance

1- Hardware

2- Software

3- Labels

II. Implementation Strategy

* * *

III. Support Services

* * *

Section 2.4 of the SOW stated:

The P/LA shall be capable of printing, cutting, and applying labels at a rate greater than 400 labels per hour. This is a minimum rate. A good operator can key at a sustained rate of over 350 pieces per hour. When processing 'good running mail', or repeat extra codes, the good operator will have spurts near 400 pieces per hour. The operator must not have to wait for the FFT to complete printing before starting the next mailpiece.

The SOW stated at section 2.1.1: "If the P/LA requires air pressure the contractor shall provide an air compressor that will provide a pressure of 100 psi and have a capacity that is 20% greater than the initial site requirement for FFTs being installed in each individual CFS unit."

The contracting officer states that she received complaints from two of the prequalified vendors about the prototype test plan, specifically that its requirements were too "strenuous" and that their units "would not be able to meet all the performance requirements of the SOW." GRL did not comment on or question the test plan. In response to the two offerors' concerns, the contracting officer issued Amendment A02 on July 30. It stated in pertinent part that the prototype test plan

will be used by the FFT Proposal Review Committee as part of the technical evaluation of the contractor's FFT proposal . . . . The demonstration shall show the equipment and explain/show how it functions, including:

1. How the printer prints labels.
2. How the label gets between the printer and label applicator. |
3. How the label applicator applies labels.
4. How the label rolls will be loaded/replaced.
5. How the ribbons, if any, are loaded/replaced.
6. Ergonomic features of the design.

The contractor's FFT is a prototype and is not expected to have all of the SOW features and functionality. This demonstration is meant as a chance for the Committee to see the physical hardware and how it works. It is not meant as a substitution for the oral presentation which should go into more depth about the design, etc.

After one prequalified offeror withdrew from the competition, the three remaining offerors submitted their prototype units for testing on August 6-8, and gave their oral presentations on August 13-15. According to the contracting officer, on August 19-20, the technical evaluation committee convened to discuss the strengths, weaknesses, evaluation scores and overall ranking of the offerors. After the price proposals were received and evaluated, the contracting officer reviewed the proposals and evaluations. She states that SEI was ranked first technically with a score of 7.55, was considered "above average" with "moderate" risk, and its evaluated price was $21,899,182. GRL, on the other hand, was ranked third technically, with a 5.75 score,⁽²⁾ an "average" rating, was considered "high" risk, and had the highest evaluated price of the three offerors,⁽³⁾ $44,486,264.⁽⁴⁾ SEI's proposal was considered the best value and award was made to it on September 6. This protest was timely filed on September 20.

The protest asserts that the FFT prototype test was to evaluate the "throughput" capability of the FFT--P/LA, which the SOW stated at paragraph 2.4. had to be at least 400 labels per hour. The protester points to the prototype testing results, which show under a heading of "productivity test" that SEI's machine applied 711 labels in two hours while GRL's applied 928. Since SEI's result amounted to less than 400 per hour, GRL asserts that SEI's offer should have been rejected as technically unacceptable, and that SEI's proposal could not have been the best value because its "hardware failed to meet the minimum requirements for hardware productivity performance" and the solicitation stated that technical ability was more important than price.

GRL asserts that its evaluation was flawed for the following reasons:

-- Despite GRL's "strengths in the most important evaluation factor," including that it "is one of only two qualified label producers recognized by the USPS," it was downgraded under the Design and Performance factor because it is a "small business" and its "team of specialized subcontractors" are also small businesses. GRL asserts that this "undue preoccupation with the size" of the GRL team "ignores the fact that GRL designed the very machine being procured" and that its software is "far more developed" than that of the other offerors.⁽⁵⁾

-- GRL was improperly downgraded for failure to provide contingency plans in the event its coater becomes inoperative. GRL asserts that it did offer alternative plans during the oral presentation.

-- During the oral presentation, GRL proposed to supply air compressors for those facilities that needed them. The Postal Service indicated that GRL should assume, for both design and pricing purposes, that all of the installations would require the supplying of separate air compressor units. GRL complains that in its debriefing, GRL was told it was downgraded for doing just that.

-- GRL asserts that the contracting officer's concerns over the size of GRL had the effect of "elevat[ing] Factor II over Factor I in order to reach the selection of an offeror other than GRL." GRL states that the latter constituted a change in the evaluation scheme which was arbitrary and capricious and not communicated to the offerors.

-- GRL's proposal was improperly downgraded for being unclear as to what intellectual property rights it was granting the Postal Service. GRL claims that it did not take "a single exception to the extensive and confusing data rights and computer software rights provisions of the solicitation," and the Postal Service could have asked GRL to clarify its position if it had been perceived as unclear.

The protest also complains that in December, 1995, before the solicitation was issued, the Postal Service allowed SEI to "inspect and photograph GRL's patented equipment in operation at the Grand Rapids Post Office." GRL asserts that its patent might have been infringed and that an appearance of impropriety or even bias was created due to the subsequent selection of SEI for this contract. GRL also claims that the solicitation was deficient because it "does not contain the appropriate clauses regarding Patent Rights, Authorization and Consent and Indemnity."⁽⁶⁾

GRL asks that the award to SEI be set aside and that this office either award the contract to GRL or, at minimum, order a re-evaluation of the proposals in accordance with the stated evaluation criteria.

In her statement in response to the protest, the contracting officer asserts that Amendment A02 was issued because she realized that the "hastily made prototypes," which were completed "within six weeks . . . could not be expected to meet all SOW requirements. . . . Because the prototypes were not expected to meet all the performance criteria of the production units, the prototype test was not a pass or fail evaluation of such performance criteria." According to the contracting officer, the prototype demonstration "was an opportunity for the offerors to demonstrate design features of their machine." She also claims that "[t]est results were never reduced to an overall score" for the purposes of the evaluations.

The contracting officer states that the evaluators considered both the oral presentation and the prototype test "as sources of information concerning the nature of the offeror's proposed design and its ability to produce and deploy it," and asserts that the evaluations were conducted pursuant to solicitation section M.

The contracting officer stresses that GRL had the lowest technical score and a price that was more than $22 million greater than that of SEI, which had the highest technical score.

The contracting officer asserts that GRL's technical proposal had "several glaring weaknesses":

GRL's manufacturing, implementation and deployment strategy was totally inadequate and considered extremely risky. There was inconsistency when comparing the number of teams and personnel in the deployment workforce and the stated GRL assumptions regarding the amount of time they would spend on site. Under question, they could not explain how the quantity of teams and installers proposed could handle the workload in the timeframe required. Additionally, the unit's reliance on air compressors and the associated noise and necessary piping to be installed at the CFS sites was considered a drawback and extremely costly.

SEI, on the other hand, presented "many significant strengths," including "the overall innovative design features of the FFT such as quiet operation, gentle application to the operator's hand, minimum number of moving parts, ease of maintenance, and high quality labels." The contracting officer asserts that SEI's prototype did, in fact, demonstrate the ability to apply labels as a rate greater than 400 per hour. She uses as proof SEI's prototype test score under the heading, "Repeat Key Test," which showed that SEI's machine completed 23 labels in two minutes. Further, SEI's "training, site preparation and field installation plans were well documented, and they offered site specific flexible scheduling."

The contracting officer denies having waived any solicitation requirements, and asserts that SEI's "contract contains the same 400 labels per hour rate requirement" as the solicitation. She states that the "repeat key test of pure machine speed" was the test that was used to determine compliance with the SOW rather than the "productivity test" because the latter required operators and the SOW did not require "operators maintaining sustained rates of over 400 labels per hour and merely requires a processing capability of 400 labels per hour."

The contracting officer replies to the specific allegations concerning GRL's evaluation as follows:

-- The evaluators "absorbed information from the prototype demonstration/test and the oral presentation and assigned individual scores for each of the subfactors." The evaluators then discussed the strengths and weaknesses of each offeror and "reached a consensus score for each of the evaluation subfactors and then calculated the total weighted technical evaluation score based on the predetermined evaluation weights."

-- GRL was not downgraded because of its size, but instead for "failure to demonstrate an ability to produce and deploy 1450 machines in the required timeframe." The contracting officer states that the "FFT procurement is a large scale employment of 1,450 systems to over 200 field locations," and that GRL's proposed subcontractor "did not demonstrate the required ability perform . . . ." GRL failed to "convince the panel that his company had the necessary skills and expertise to perform the work" and the deployment subcontractor as well as a low number of installers added to the impression that "GRL, who is primarily a label supplier, failed to demonstrate adequate experience in manufacturing equipment of this scale, and did . . . not provide credible manufacturing or quality plans."

-- The evaluators did consider GRL's contingency plan, "but found it to be inadequate." The panel "was unimpressed with their contingency plan as presented. GRL only had one coater and planned to move to a 4 shift operation, possibly overstressing the machine. They offered 2 unnamed firms as potential back-up, but did not provide further details on their contingency plan."

-- The use of compressed air was considered a weakness "because it required extra logistical work and increased noise."

-- The solicitation included a "software" subfactor; therefore, software rights were properly considered in the evaluation. Further, during the oral presentation, "GRL was unable to specify exactly which portions of their software would be subject to limited rights," demonstrating an unacceptable "lack of knowledge." Finally, GRL "confirmed in its cost/price proposal that it was restricting some software."

In response to the contracting officer's statement, GRL reiterates the major points in its protest and makes the following points:

-- Between 1992 and 1995 GRL had several contracts to develop the Printer Label Applicator (P/LA) which is "at the heart of the FFT." The purpose of the FFT has been to save in labor costs; therefore, it is mandatory that any label machine be able to operate faster than a human operator, thereby making performance speed "entirely operator dependent, not machine dependent." The FFT must process more than 400 labels per hour because efficient operators perform at nearly that speed and "the operator must not have to wait for the FFT." Without enforcing that minimum, the Postal Service will not achieve the cost savings which GRL asserts is the purpose of automating this requirement.

-- It was inconsistent for the contracting officer to expect the prospective contractor to use GRL's P/LA or the equivalent and then to state that GRL lacked manufacturing capability.

-- Amendment A02, issued less than one week before the prototype testing was to occur, did not change the prototype test requirements as claimed by the contracting officer. Citing the amendment's language, GRL claims that nothing "deleted the requirement that the prototype hardware meet the requirements of the SOW."

-- Unlike the other offerors, GRL was not privy to discussions about the purpose of Amendment A02, and "nothing in [it put] GRL on notice that the prototype hardware would no longer be required to meet SOW requirements. . . ."

-- Substituting the repeat key exercise for the productivity test to measure productivity throughput is "an attempt by the USPS to 'rationalize' their decision to waive a mandatory minimum requirement by making it 'go away.'. . ." Such substitution is "absurd and jeopardizes the ability of the USPS to demand the productivity upon which [Board of Governors] approval for this procurement was based."

-- Quoting paragraph 2.4 of the SOW, the protester asserts that the contracting officer is pretending that the P/LA capability "of printing, cutting and applying labels at a rate greater than 400 labels per hour" was not a minimum rate that had to be tested without the use of the repeat key test, which, since it relies on the condition that all labels would be identical, is an unrealistic test for productivity.

-- Since GRL's machine processed 928 labels for the two-hour productivity test, while SEI's only produced 711, the contracting officer's use of the repeat key test instead amounts to justification of award to a "nonqualifying offeror."

-- Had GRL known "that the stated evaluation plan would not be used or stated minimum requirements were to be waived or evaluated differently (such as using the two minute repeat key test as a measure of throughput as opposed to the two hour productivity test) its proposal would have been priced and structured differently."

-- To the evaluators, GRL's proven past performance "was not as important as how well GRL's computer analyst subcontractor could extemporaneously answer the questions of the USPS computer analyst consultant" during oral presentation, elevating "presentation style" over substance.

-- "GRL still remains frustrated and annoyed that its proposal was downgraded because it included the cost of an air compressor for each facility, when it believes it was instructed to do so by the USPS."

GRL has submitted an unsigned affidavit from the supervisor at the post office where GRL's equipment was photographed, which states that she felt "uncomfortable" with SEI's "special viewing" [the photo session] of GRL's equipment because this solicitation had not yet been released.

Finally, the protester asserts that the "case for requiring the termination of the award to SEI is compelling, for all the factors cited in TPI International Airways, Inc.[, P.S. Protest No. 87-40, October 30 1987]. In addition, it is clear that the very premise upon which this procurement is based, i.e., the cost savings which justify the investment in the FFT[] system has been compromised by selecting a contractor which has failed to meet the minimum throughput requirements." GRL also asks for reimbursement of its costs of filing this protest.⁽⁷⁾

In rebuttal, the contracting officer reiterates that the solicitation and Amendment A02 "contradict GRL's position." She states that "GRL is simply wrong when it contends that the prototype was required to demonstrate operator driven sustained productivity of over 400 labels per hour." Further, the repeat key test "is the best measure of the machine's productivity" because it measures "the number of labels printed, cut and applied to flat mailpieces. . . and was the best measure for pure machine speed, because it eliminated the keying function" which relies on individual characteristics of operators.

The contracting officer also disputes GRL's claim that style counted over substance:

The oral presentation portion of the source selection process provides the offerors an opportunity to demonstrate their knowledge, skills and abilities to perform the work. If an offeror cannot adequately convey their understanding of the work to be performed or to demonstrate that they have the ability to perform the work, it is eminently reasonable that the USPS evaluators score the related evaluation criteria accordingly.

The contracting officer also denies that factor II was weighted more heavily than factor I, that GRL was asked to price equipment (compressed air) that it did not need to use, and that SEI "gave us less than unlimited rights in the P/LA." She also concludes that "[e]ven if . . . GRL was given a perfect score for [the software rights] subfactor, its technical score would still be lower than that of SEI and its price would still be more than double that of SEI" rendering the deduction for this subfactor "harmless error."

Finally, the contracting officer states that the unsigned affidavit from the post office supervisor "is less than compelling evidence" that the Postal Service was biased in favor of SEI in advance of the solicitation process.

In a protest conference and subsequent written comments, GRL emphasized its view that "expressly stated minimum performance requirements . . . were waived in order to conduct a best value trade-off and ultimately award a contract to a noncompliant offeror." The protester asserts that had the procurement "been run properly, the contract would have been awarded to GRL," which was the only offeror whose prototype met the requirement to "demonstrate the capability to print, cut and apply labels at a rate greater than 400 labels per hour."

GRL reiterated its arguments and also asserted that:

-- Amendment A02 did not affect the 400 label/hour requirement and that proficiency was required to be demonstrated in the prototype testing.

-- If Amendment A02 was supposed to have the effect of changing the productivity requirement, that fact was not communicated to GRL. The rationale for the amendment, that is was a response to the other offerors' complaints that they could not make a fully functional prototype so quickly, was not told to GRL and therefore GRL lacked information which the other offerors had.

-- The only way SEI's prototype could be seen to have met the productivity requirement of 400 labels per hour would be to use the results of the "repeat key test" rather than the test called "productivity." However, GRL maintains that those tests reflect two different measurements and the repeat key test requires a significantly lower level of productivity. GRL, then, "wouldn't have proposed a Cadillac if what Postal Service wanted was a Geo" and would have changed its oral presentation as well.

-- GRL was prejudiced because it would have recognized problems with the solicitation terms and protested them in a timely fashion.

-- Since SEI could not meet the productivity requirement, it should not have made it to the "best value" stage of the selection process.

On the evaluations, GRL made the following points:

-- GRL got insufficient credit for meeting the productivity test requirements.

-- GRL was improperly downgraded for some intellectual property rights while SEI was not and in its contract it does restrict some of those rights.

-- SEI's prototype used compressed air and the machine that SEI actually is using to perform the contract does not; it was improper and senseless to award a contract for an unprototyped unit.

-- When GRL proposed to use compressed air, the Postal Service asked GRL to price it in its proposal as if compressed air would be needed at every facility without actually knowing how many facilities would need it. GRL was prejudiced because that unnecessary request to price compressed air for all of the facilities artificially inflated its price offer.

-- Further, SEI's price proposal did not include compressed air even though its prototype used it.

-- The flawed productivity evaluation affected the best value determination, because if GRL's machine's productivity is greater, its value is as good as a lower-priced, less productive machine.

Finally, GRL asserts that SEI's exclusive opportunity to photograph GRL's machinery in December, before issuance of the solicitation, gave an "appearance of impropriety" that calls into question the "integrity of the process."

In answer to questions posed in a memorandum from this office, the contracting officer provided the following comments:

-- GRL confuses definitive responsibility criteria for performance requirements. The prototype was never required to be able to process 400 labels per hour by the original test plan or the revised one "following Amendment A02"; the SOW never mentions the word "prototype," referring instead to the P/LA in a "specification requirement that applies to the production units to be delivered in contract performance."

-- The productivity test was "a simplified simulation of the FFT operation [which' was never intended to measure a maximum productivity rate, just stress the prototype in question." The protester "is trying to over-define the term 'productivity' and has given it a meaning it was not intended to convey."

The Proposal Evaluation Committee wanted to know how the prototype operated in the print and label application modes. Did the design create frequent jams? Was the operator restricted in some manner? Did the label and ribbon feed mechanism work properly? And any other intelligence the operator observed about the prototype.

-- For the productivity test, which "was a simplified simulation of the FFT operation," the Postal Service used as an operator a member of the evaluation committee. If the Postal Service had wanted to test "the true productivity of the machine," then keying "would have been performed by an experienced CFS operator, not a committee member." The repeat key test is a better measure of "the capability of the machine alone and not the operation of the machine under operator keying mode." As the SOW stated, "A good operator can key at a sustained rate of over 350 pieces per hour"; therefore, the Postal Service "cannot require that even the production units maintain consistent operation at 400 labels per hour in normal operating mode when an operator keys the label."

-- It was not improper to allow other manufacturers to photograph GRL's equipment at the Grand Rapids Post Office because the Postal Service "purchased the GRL units without any express restrictions on their use. Absent such express restriction, the USPS can permit individuals to view and photograph the equipment at our discretion."

-- She believes that "there must have been a misinterpretation of what was said" with regard to the supervisor's affidavit. In a telephone conversation, the supervisor did not dispute the contents of the unsigned affidavit, and "was quite vague about her recollections of comments made by the SEI representative. . . ." Both the Postal Service officials and SEI representative involved deny having made any comments to the effect that the Postal Service was biased in favor of SEI.

-- The contracting officer asserts that the SOW "describes clear software requirements." If GRL "did not understand the SOW software requirements it should have requested clarification prior to presenting a proposal. To the extent that they are now challenging the specifications of the solicitation, GRL is untimely."

-- Clause H.12 required the contractor "to grant the USPS unlimited rights to use any software first produced in the performance of this contract." If GRL planned to use software that it developed previously, it was required to inform the Postal Service with its proposal. "If GRL [had] not yet developed the software necessary to comply with USPS requirements, then it must grant the USPS unlimited rights."

-- GRL "was downgraded for limiting the USPS' rights and for failure to adequately describe what rights the USPS would receive. . . . [T]he USPS had the right to evaluate GRL's proposal less favorably because it granted less favorable data rights."

-- GRL is "simply wrong" in its assertion that SEI's contract does not mention software rights. SEI "represented that no software [rights were] restricted."

-- If GRL had been downgraded improperly for its limitation of software rights, it would be "harmless error" because "even if GRL received a perfect score on the software category, its total technical score would only be 6.5 out of 10, a 16% lower score than SEI's score and its price would still be $44,486,264 compared to SEI's price of $21,899,182. SEI would still be the best value."

-- GRL's claim that the SEI prototype used compressed air is "completely and unequivocally false." Compressed air "in this conte[x]t means that the machine would need to be connected to a compressed air source. SEI's machine was not, but GRL's was." The SEI production unit (as well as SEI's prototype) is "a high volume, low pressure air driven system. It does not require a separate system to compress the air to a high pressure as is required by the GRL machine."

-- It was proper to ask GRL to price its proposal as if every facility would need it to supply a compressed air source because not all of the facilities have one and even those that do "many not have it conveniently located. . . ." It was "necessary to evaluate such costs because they would be real costs to be incurred by the USPS. . . . The fact of the matter is that SEI designed a more economical machine than the GRL unit."

In a final submission, GRL stated the following:

-- The Postal Service has been inconsistent in stating alternatively that the SOW requires productivity of 350 and 400 labels per hour, and SEI's "productivity test" results show that its system cannot meet the true requirement, 400 labels per hour.

-- Since GRL's system is more productive, its labor costs would be sufficiently less than SEI's and its system, therefore, would be more economical for the Postal Service.

-- The contracting officer's explanation why the repeat key test was the appropriate measure of productivity "seem[s] to be a rationalization to explain the award but not the basis that was actually used . . . . It is obvious and troubling that the Contracting Officer is trying to create a justification for the award to SEI 'after the fact'."

-- The requirement to process in excess of 400 labels per hour "does not pertain to the offeror's responsibility, but does establish a mandatory requirement that must be met during prototype testing."

-- It is "a matter of fundamental fairness that when an agency disseminates written responses to offeror's questions during the course of a procurement . . . the agency is bound by the responses." GRL asserts that "several responses to questions" indicated that the prototype "had to meet SOW hardware requirements."⁽⁸⁾

-- The repeat key test is not an appropriate measure of productivity because the repeat key "is used relatively infrequently. Data accumulated during the FFT test period over the past months reveal that repeat key entries are less than 2.5% of total FFT operations."

-- GRL maintains that SEI has excepted software rights and states that the contracting officer has confused technical data rights with software rights, citing PM 9.3.1 m., which defines "technical data" as "other than computer software . . . ."

-- GRL also maintains that SEI's machine does use compressed air and that GRL's price proposal was unfairly inflated by the Postal Service's erroneous instructions.

SEI commented on the protest, agreeing with the contracting officer and disputing GRL's version of the facts as follows:

-- SEI has granted the Postal Service "unlimited" intellectual property rights.

-- The FFT that SEI proposed is based on a "new invention," using new and "superior" technology, which does not infringe upon GRL's patent.

-- SEI has observed only "impartiality and insistence of a level playing field" by Postal Service personnel. The idea that it "had an USPS insider helping in winning this solicitation" is "[n]ot only untrue . . . it is ridiculous."

Discussion

In claiming that the prototypes were required to demonstrate the ability to apply 400 labels per hour and that SEI's failed to show that capability, GRL is challenging both the contracting officer's interpretation of solicitation Amendment A02 and the contracting officer's affirmative determination of SEI's responsibility.

According to the contracting officer, the prototypes were not required to meet the performance requirements of the SOW because Amendment A02 was issued for the very purpose of relieving the offerors of those requirements on the day of testing. GRL claims that Amendment A02 had no such effect. Thus, we have two conflicting interpretations of a solicitation amendment.

An ambiguity exists if the specifications are "susceptible to two reasonable interpretations." Dataware Systems Lease, Inc., P.S. Protest No. 91-41, October 10, 1991; Nasuf Construction Corporation Reconsideration, Comp. Gen. Dec. B-219733.2, March 18, 1986, 86-1 CPD 263; Compo Corporation, P.S. Protest No. 88-22, May 3, 1988. Here, the amendment's language could have meant anything; the amendment definitely was ambiguous. However, what dooms GRL's challenge to the amendment is that it is untimely. The fact that the amendment could have had numerous possible meanings was apparent on the face of the amendment and could have been challenged before the prototype test date. See Procurement Manual (PM) 4.6.4 b; see also, Lista International Corporation, P.S. Protest No. 90-47, September 11, 1990.

GRL asserts that it could not know that Amendment A02 was issued to relax the prototype requirements, and had it known, GRL claims that it would have protested or would have proposed a less expensive model. However, GRL was, at the very least, on notice to inquire about a possible ambiguity prior to the prototype due date. Pitney Bowes, Inc., P.S. Protest No. 89-24, June 20, 1989. While it is true that the other offerors knew the purpose of the amendment and GRL did not, GRL knew that it did not know the purpose and it should have asked the contracting officer. It was not improper for the contracting officer to respond to the questions of the other offerors by issuing the amendment, and the contracting officer was not obligated to protect GRL's status as the only offeror with a fully functional prototype. Further, while the amendment's language certainly could have been more explicit, it is not clear that a timely challenge to its underlying purpose would have been successful. In any event, GRL's protest against the amendment and its application is untimely and must be dismissed.

As the contracting officer correctly states, GRL's contention that SEI's system cannot meet the 400 labels per hour requirement of the SOW is a challenge to her affirmative determination of SEI's responsibility, since responsibility relates to a firm's ability to comply with the terms of the solicitation. Transnorm System, Inc., P.S. Protest No. 90-58, October 26, 1990. "An affirmative determination of responsibility is a matter within the broad discretion of the contracting officer and is not subject to being overturned by this office in the course of a protest absent fraud, abuse of discretion, or failure to apply definitive responsibility criteria." Id., quoting Gage Constructors, P.S. Protest No. 87-11, July 13, 1987. Fraud has not been alleged here and would not be supported by the record.

The requirement that the FFT process more than 400 labels per hour is not a definitive responsibility criterion, which is a requirement for the awardee to demonstrate certain factors about itself (such as its experience with large-scale procurements). It is, instead, a performance requirement, which relates to the offeror's ability to meet specification requirements for the product to be furnished. Id.

Moreover, the contracting officer maintains that SEI's machine does have the required performance capability, if we accept that the repeat key test is a better measure than the productivity test. Despite the unfortunately confusing choice of names for these tests (it might have been better to refer to what was called the "productivity test" as the "operating test" and call the other measure the "repeat key productivity test"), the contracting officer's explanation why the repeat key test was more determinative of whether the machine itself could process more than 400 labels per hour is reasonable. The purpose was to have a machine that could stay ahead of the best operator's capability. The repeat key test showed the machine's speed unencumbered by the intervention of a human operator. Thus, it was not unreasonable or an abuse of discretion for the contracting officer to conclude that SEI's machine, which applied 23 labels in two minutes, had the required capability.

The determination of the relative merits of technical proposals is the responsibility of the contracting office, which has considerable discretion in making that determination. It is not the function of our office to evaluate technical proposals or resolve disputes on the scoring of technical proposals. In reviewing a technical evaluation, we will not evaluate the proposal de novo, but instead will only examine the contracting officer's evaluation to ensure that it had a reasonable basis. The protester bears the burden of showing that the technical evaluation was unreasonable. Standard Register; Moore Business Forms, Inc., P.S. Protest No. 92-68, November 23, 1992. GRL has not met that burden.

We do not understand the evaluators' concerns about GRL's manufacturing capability as relating, as GRL suggests, to its ability to manufacture the product itself, but instead, to manufacture and coordinate the delivery/installation of 1,450 of them. As stated in footnote 5, supra., GRL's previous contracts called for at most eight P/LAs. It was not unreasonable or inconsistent with the fact that GRL manufactured the first P/LA to question GRL's ability to manage a procurement of this size.

If the evaluators were confused, through no fault of GRL, about the intellectual property rights that GRL proposed to grant, the error was harmless. As the contracting officer points out, even without a technical deduction in that area, GRL still would not have equalled SEI's technical scores.

While it is possible that GRL received confusing instructions about the number of compressed air units which it would have to provide, and that its pricing might have been artificially inflated somewhat as a result, we need not resolve that question because again, any error was harmless. The $640,000 which was attributed to the compressed air part of GRL's price proposal does not come close to closing the more than $22 million gap between its price and SEI's.

GRL has claimed that SEI's machine uses compressed air, that SEI limited the intellectual property rights that it granted,⁽⁹⁾

and that the postal officials who let SEI inspect GRL's equipment in December, 1995, acted improperly in doing so and were biased; the contracting officer categorically denies these assertions. "In a factual dispute we adopt the contracting officer's position absent sufficient evidence to overcome the presumption of correctness. . . ." Cohlmia Airline, Inc., P.S. Protest No. 87-118, April 13, 1988. GRL has not submitted evidence sufficient to overcome this presumption of correctness. The contracting officer has made it clear that SEI's proposal represented the best value to the Postal Service because of what the evaluators viewed as superior new technology--a different type of system, which was, among other things, less noisy. GRL, understandably, heatedly disagrees with the contracting officer that its technology is not the superior and most economical. However, mere disagreement with the evaluators on the relative merits of proposals does not amount to evidence sufficient to sustain a protest. See, e.g., Apec Technology Limited, P.S. Protest No. 88-23, June 30, 1988. Our limited review of the record has indicated that the evaluations were conducted even-handedly and in accordance with the solicitation. GRL has given us no grounds to overturn the comparative evaluations based on technical value judgments, for which the evaluators and contracting officer are afforded considerable discretion. Standard Register, supra.⁽¹⁰⁾

GRL has repeatedly complained that the contracting officer insufficiently investigated the December, 1995, photo session, and has suggested that this office must do so to protect the integrity of the process. However, our office does not conduct independent factfinding investigations and we are not set up to conduct adversary proceedings; rather, we resolve protests based upon the written record supplied by the contracting officer, protester and interested parties. See COR, Inc., P.S. Protest No. 90-16, June 22, 1990; Cohlmia, supra. Here, the record does not come close to providing the "irrefragable" proof necessary to conclude that the postal officials acted in bad faith or in a biased manner when they allowed SEI to photograph the equipment at the Grand Rapids post office. See A-Transport Northwest Co., Inc. v. United States, 27 Fed.Cl. 206, 220 (November 25, 1992).

The protest is dismissed in part and denied in part.

William J. Jones

Senior Counsel

Contract Protests and Policies

1. The FFTs were to be part of the Postal Service's computerized forwarding system for mail which must be labelled with new addresses.

2. Points were assigned to the three major factors in descending order as follows:

I. Design and Performance 60%

II. Implementation and Strategy 25%

III. Support Services 15%

Within each factors, points were further subdivided by sub-factors and (for hardware and labels) sub-sub-factors.

Each offer was given a raw score on each factor, sub-factor, or sub-sub-factor, apparently on a ten-point scale, although no offeror received a perfect score on any item. The raw score was multiplied by the point score for the factor to arrive at a rating for the item, and the ratings were totaled to arrive at a the total point scores. Since a perfect total point score was 10, the scores can be viewed as representing scores of 57.5% and 75.5%, respectively.

Grand Rapids' and SEI's scores by the major factors and the subfactors of the first major factor were as follows:

GR                       SEI 

I Design and Performance 

1. Hardware                   2.08  2.11 

2. Software                    .75   .90 

3. Labels                     1.05  1.25 

II Implementation Strategy          1.10  2.26 

III Support Services                 .79  1.07 

Totals                        5.77  7.59 

(The contracting officer's totals for all three offers understate them slightly from the totals as we calculate them.)

3. Even though third both technically and with respect to cost, GRL has standing to challenge the manner in which the procurement was conducted. Computerized Screening, Inc., P.S. Protest No. 95-56, March 12, 1996.

4. In the August 21 evaluation memorandum, the evaluators listed in detail the "specific risks" associated with each offeror's system. They noted considerably more questions and problems with GRL's system than with either of the other two offerors' systems, stating:

The prototype needs the printer/label applicator section redesigned (no repackaging needed) to allow the required adjustability . . . . The software demonstrated was only a small part of the required software and the presentation left us with the impression that the software person was confused about the SOW requirements. [The software subcontractor] is an 11 person firm which appears to be a major constraint in producing FFTs, and no alternative or contingency was presented. The manufacturing and deployment plan presentations left us with doubts that Grand Rapids had ever[] done a production or deployment of the FFT magnitude and represented considerable potential risk.

The high risk for Grand Rapids dealt not with the hardware but primarily with the deployment. Grand Rapids' planning seemed very uncertain and they didn't present any strong management tools to control the entire operation -- and with [its subcontractors] in the Albuquerque area a good management control system would be necessary. In addition, they plan to install one compressor in each CFS site with the Postal Service installing the necessary power and compressed air piping to the FFT sites. This Postal Service requirement guarantees a difficult deployment with many scheduling problems and considerable additional costs.

A [rough order of magnitude] estimate for the Postal Service cost of installing the compressed air piping is $3,000 per site, or $654,000 for the 218 CFS sites. This cost needs to be considered in the 'best value' determination.

The risks associated with SEI's proposal related mostly to the fact that the design was new, and "going from 'rough' prototype to production always involves some risk." The evaluators also mentioned that SEI's software was "not completed and only a rough version was presented along with the prototype for testing."

5. GRL, the developer of the first P/LA, has had five previous postal contracts involving the FFT and P/LAs. However, they were not of the scope of the current solicitation, instead involving between 1 and 8 PL/As each.

6. This claim of a solicitation deficiency is untimely since the alleged solicitation defect was apparent before the date set for the receipt of proposals and, therefore, any protest against it was due by that date. PM 4.6.4 b. Further, our protest function does not include jurisdiction over contentions involving matters of patent infringement. Dataview Electronic Systems, Inc., P.S. Protest No. 90-09, March 29, 1990; Facet Transportation, P.S. Protest No. 86-75, October 7, 1986; Tulsa Diamond Manufacturing Corporation, et. al., P.S. Protest Nos. 85-18, 85-20, and 85-23, June 20, 1985.

7. This office lacks authority to reimburse GRL for its protest costs. Sircle Spring Company and Packaging Accessories Company, P.S. Protest Nos. 91-86 and 91-91, January 21, 1992.

8. Amendment A02 was issued after the responses to which the protester refers.

9. PM 9.3.1 n. defines "unlimited rights" as rights in "technical data and computer software . . . ." [emphasis added], which negates GRL's assertion that the contracting officer confused the two.

10. That GRL might have proposed a more expensive unit than it needed to, and that its price might have been further inflated by the compressed air misunderstanding, are arguments unavailing to GRL. We could not overturn the contracting officer's best value determination even if GRL could be seen as completely closing the $22 million gap in price, because the solicitation made it clear that for best value purposes, technology was more important than price.