B-233709, Apr 3, 1989, 89-1 CPD 342

B-233709, Apr 3, 1989, 89-1 CPD 342

PROCUREMENT - Bid Protests - GAO procedures - Protest timeliness - 10 day rule DIGEST: 1. Where doubt exists concerning the date on which a protester became aware of the basis of its protest, the doubt is resolved in favor of the protester. PROCUREMENT - Contract Management - Contract modification - Cardinal change doctrine - Criteria - Determination 2. Protest that a contract modification was beyond the scope of the contract is denied where the modification did not result in the procurement of services materially different from the services competed under the original contract.

CAD Language Systems, Inc.:

CAD Language Systems, Inc. (CLSI), protests the Department of the Air Force's modification of contract No. F33615-87-C-1463, a cost reimbursement research and development contract awarded to Intermetrics, Inc., for development of software tools. CLSI contends that the work called for under the modification is beyond the scope of work set out in the prime contract, and argues that the work should be obtained by competitive procurement.

We deny the protest.

CLSI designs, develops and markets computer software in the area of computer-aided design (CAD). The protester's products are written to comply with two new major CAD software standards, VHDL and EDIF. VHDL is a Department of Defense (DOD) developed text-oriented language, which stands for VHSIC (very high speed integrated circuits) Hardware Description Language, and is used in designing digital systems. EDIF stands for Electronic Design Interchange Format, an American National Standards Institute (ANSI) standard for exchange of hardware design data. EDIF graphically describes the physical layout of integrated circuits and is in wide use on engineering workstations for the exchange of electronic data, schematics, netlists, and physical designs.

Intermetrics' prime contract was awarded under a 1987 program research development announcement (PRDA) for production of a new VHDL standard Institute of Electrical and Electronic Engineers (IEEE) Draft Standard 1076/A, to replace the then current standard, as well as work in support of a joint United States/Canadian VHDL tool development effort. The PRDA's estimated magnitude of effort was approximately $2 million. The PRDA's scope is broad, generally requiring offerors to produce a VHDL environment that includes at least the following tools: (1) an analyzer, (2) a design library manager, (3) a simulator, and (4) a reverse analyzer. Offerors could include additional tools as options in their proposals. Deliverable items included software test plans and test reports for any tools awarded. The PRDA provided three criteria for proposal evaluation in descending order of importance: (1) new and creative approaches to producing the new standard, (2) scientific and engineering merit of the proposed approach, and (3) cost. The agency received proposals from Intermetrics, Honeywell, and a CLSI/Unisys team. The CLSI/Unisys proposal was rejected because of technical risk, longer development schedule and cost.

In July 1987, the agency awarded Intermetrics the $1,670,999 prime contract entitled "IEEE 1076 toolset implementation and VHDL Canadian rehost support." The contract states (Sec. C, paragraph 1.0 and 1.2) that its purpose, in part, is to "develop and produce the IEEE standard VHDL environment ... in order to ensure the success of the VHDL in the Electronic and CAD industries. ..." The award was less than the estimated dollar value, in part, because of the agency's determination not to exercise the options for additional tools.

On May 27, 1988, the agency modified the prime contract (modification P0004 for $792,270) both to implement two options proposed by Intermetrics in its February 27, 1987, proposal and to require a "graphics interface" tool in addition to the previously required tools.

On September 29, the agency again modified the contract (modification P0005 for $557,760), adding requirements for: (1) validation tests /1/ for 2 of the 4 tools (the analyzer and simulator); (2) training for 30 Department of Defense (DOD) personnel on the VHDL toolset; and (3) a "bi- directional interface with EDIF schematics." Throughout this period, Intermetrics representatives attended government/industry conferences reporting on its work. On November 16, the second modification (P0005) of the Intermetrics contract was synopsized in the Commerce Business Daily (CBD). This protest then was filed on November 28.

As a preliminary matter, the agency urges dismissal of CLSI's protest as untimely filed. The agency contends that CLSI either knew or should have known of the modifications to the prime contract as a result of attending an industry symposium in October 1988, where both the government and Intermetrics briefed the attendees on the nature of the changes to the prime contract, and therefore should have filed its protest within 10 days thereafter. See Bid Protest Regulations, 4 C.F.R. Sec. 21.2(a)(2) (1988). Since CLSI did not file its protest until November 28, the agency argues that the protest is untimely.

The agency's argument is premised upon the assumption that the protester understood that the software developments that Intermetrics described at the symposium were made under the prime contract. The protester admits attending some symposium sessions, but claims to have left the sessions with the impression that the government would like to support work similar to the work described by Intermetrics at some time in the future when funds became available, but that the work described was independent research and development by Intermetrics for commercial applications. The parties have submitted conflicting affidavits supporting their respective versions of what was said, and what each side contends CLSI heard or should have heard at the October symposium. For example, the agency reports that CLSI's president voiced concern about Intermetrics' performance of modification P0005. However, CLSI's president admits only that he questioned the conflict of interest inherent in the government's allowing any commercial vendor - as opposed to an independent agency or standards body-- to develop the VHDL validation suite. Moreover, Intermetrics' proposal, which was submitted in camera, references VHDL work that was "internally developed" using "internal funding." Where, as here, the parties provide plausible but conflicting statements as to the timeliness of a protest, we will resolve doubt in favor of the protester. See GEBE Gebaeude und Betriebstechnik, GmbH, B-231048, July 7, 1988, 88-2 CPD Para. 20. Therefore, we find CLSI's November 28 protest to our Office, filed within 10 days after the CBD announcement, to be timely.

We generally do not consider protests against contract modifications since modifications involve contract administration, which is the responsibility of the contracting agency, not our Office. Northeast Air Group, Inc., B-228210, Jan. 14, 1988, 88-1 CPD Para. 33. We will consider, however, situations where it is alleged that a modification improperly exceeds the scope of the prime contract and therefore should be the subject of a new procurement. Clean Giant, Inc., B-229885, Mar. 17, 1988, 88-1 CPD Para. 281. In weighing the propriety of a modification, we look to whether there is a material difference between the modified contract and the prime contract that was originally competed. Indian and Native American Employment and Training Coalition, 64 Comp.Gen. 460 (1985), 85-1 CPD Para. 432.

In determining the materiality of a modification we consider factors such as the extent of any changes in the type of work, performance period and costs between the modification and the prime contract. See American Air Filter Co., Inc., 57 Comp.Gen. 285 (1978), 78-1 CPD Para. 136, aff'd on reconsideration, B-188408, June 19, 1978, 78-1 CPD Para. 443. In this regard, we also consider whether the prime contract solicitation adequately advised offerors of the potential for the type of changes during the course of the contract that in fact occurred. National Data Corp., B-207340, Sept. 13, 1982, 82-2 CPD Para. 222.

CLSI urges that the work called for under the two modifications is beyond the scope of work set out in the prime contract, and therefore required a new procurement. Based on a thorough review of the record including review by our technical staff, we find that the agency's modifications were within the scope of the prime contract. Accordingly, the agency was not required to conduct a new procurement for the work in question.

As we noted above, modification P0004 required Intermetrics to implement two options proposed in Intermetrics' February 27, 1987 proposal /2/. The modification also contained two additional options proposed by Intermetrics in a second preaward proposal which it made contingent upon the exercise of option 3 of the initial proposal, (1) a base generic workstation interface providing "a graphics interface to the full VHDL 1076/A toolset, running on a Sun workstation" (option 6), and (2) an interactive generic workstation interface providing "full interactive control of the VHDL 1076/A simulator through the graphics interface" (option 7), as well as a proposed schedule.

The protester states that it does not object to modifying the prime contract to include options for additional tools contained in the original Intermetrics proposals. However, the protester questions whether the work covered by the modification is in fact the same as the option work offered in Intermetrics' proposals since the agency considered it necessary to obtain another proposal from Intermetrics before exercising the option. CLSI speculates that the need for a new proposal indicates a substantial change in the work.

Our in camera review of the proposals shows that Intermetrics' April 5, 1988, proposal, which formed the basis of modification P0004, did not change the substance of the earlier offered options; rather, it merely provided the agency with a new schedule and current pricing on the options for use in negotiating modification P0004. Accordingly, we see no basis to object to modification P0004 to the extent that it included options in the form of additional tools which Intermetrics had proposed prior to the award of the prime contract in response to the PRDA's express invitation to propose additional tools.

Concerning modification P0005, CLSI's principal objection is to Intermetrics' development of "a set of tests suitable for validation of VHDL IEEE 1076 Analyzer, simulator pairs." In addition to tasking Intermetrics with the development of the preliminary validation tests for the analyzer and simulator, modification P0005 required Intermetrics to:

(1) coordinate with the various companies developing VHDL tools to obtain donated tests;

(2) evaluate the donated tests and integrate the appropriate tests into the test suite;

(3) make the test suite as host-independent and as self- checking as possible;

(4) distribute the tests to the VHDL tool vendors and users;

(5) coordinate with the IEEE on the conformance of the tests to the language reference manual; and

(6) coordinate with the National Institute of Standards and Technology (NIST) formerly the National Bureau of Standards on the format of the tests.

The protester's argument is twofold. On one hand it contends that it is improper for one vendor to develop the tests by which both its and other vendors' products will be judged for conformance to the IEEE standard. Specifically, CLSI argues that the modification will give Intermetrics (1) an opportunity to manipulate the suite to include tests that its software passes and exclude tests that its software cannot pass, and (2) an early view of the suite and more time to conform its software to suite requirements. CLSI contends that the work is better performed by an independent organization like NIST. On the other hand, recognizing that the PRDA requires software testing for the tools developed under the contract, the protester argues that its protest is not against additional validation tests for Intermetrics' software, but against validation tests for industry-wide tools (non Intermetrics software). The crux of the protester's argument rests upon the contention that the prime contract's standardization effort and the tools resulting therefrom should be restricted to government use. We find no merit in this ground of protest.

The agency advises that Intermetrics will not be certifying the accuracy of any software tools except to insure that the analyzer and simulator developed under this contract comply with specifications. The agency intends at some later date to develop a full, formal suite using the increased quantity of tests resulting from this effort, and agrees that NIST has in the past been responsible for testing of tools for certification for validation suites.

Clearly the intent of the prime contract was to "develop and produce the IEEE standard VHDL environment ... in order to ensure the success of the VHDL in the Electronic and CAD industries. ..." This is an evolving technical area with new approaches, new tools, and new tests all simultaneously in the process of development. A critical part of standards efforts is validating that vendors' products conform to the standard, and validation test suites are part of a community effort involving vendors, the government and standards bodies. Further, modification P0005 does not call for a finished product in the form of a final test suite; rather, it is limited to the validation of two tools to serve as the basis for discussing and developing a formal validation suite. In this regard we think the extensive industry/government coordination required of the contractor and the fact that the National Bureau of Standards will work with the contractor to develop the test format safeguards the process from the abuses feared by the protester. Bearing in mind that the work called for is of a research nature and concerns an evolving technical area, we do not find it unreasonable to view the development of VHDL validation tests to instill confidence in industry users as falling within the scope of the prime contract's twin goals of developing the VHDL environment and ensuring the success of VHDL with industry. This is particularly so where the modification is limited to tasking the contractor with the provision of preliminary tests. Further, the modification did not expand the delivery schedule nor add unreasonably to the costs given the uncertain nature of the effort.

We also see no merit in the protester's assertion that modifying the prime contract to provide training for 30 DOD personnel in addition to the training already provided for Canadian personnel is improper. The agency reports that failing to provide for the training of its own personnel was a minor oversight since the contract already includes the development of necessary training and course materials. The protested training merely adds instructor hours and instructor travel to DOD training sites. In our view such training is clearly within the scope of the contract. Moreover, we think it would be illogical for the government to expend over $2 million developing the new VHDL standard and supporting tools and then train only foreign nationals in their use.

Likewise, there is no merit to the protester's objections to the modification of the prime contract to require the development, as an integral part of the VHDL design workstation software, of a bi directional interface to EDIF schematics. This software transforms VHDL structural descriptions into EDIF schematic view data and vice versa. CLSI contends that work with the EDIF standard cannot be within the scope of the prime contract because when the contract was awarded the EDIF standard did not exist. We disagree. The agency acknowledges that EDIF came into existence after the prime contract, but urges that the purpose of the contract is to develop a new standard and that if that new standard does not work with other current computer engineering design standards, such as EDIF, it is not as useful as it could be. In our view, it is reasonable to require that the agency's new standard when developed be capable of use with another newly developed standard.

The protest is denied.

/1/ Validation tests are sets of test cases constructed to demonstrate that a product conforms to requirements. In the context of formal standards validation test suites are used to determine whether products conform to standards.

/2/ Intermetrics proposed Option 1 for an enhanced inter-active simulator (mainly rewriting one of the tool's codes in another software programming language), and Option 3 for Sun/UNIX Rehost (installing a VHDL toolset on a Sun UNIX workstation).