Dynalantic Corp., Appellant, v. the United States, Appellee, 945 F.2d 416 (Fed. Cir. 1991)

DynaLantic Corporation (DynaLantic) appeals from a decision of the General Services Administration Board of Contract Appeals (the board), docket GSBCA No. 10956-P, affirming a decision of the Department of the Air Force (Air Force or agency) denying DynaLantic's bid protest to an award by the Air Force of a firm fixed priced contract to Environmental Tectonics Corporation (ETC) for the production of a computerized system for training personnel in maintaining F-16 aircraft. The bid of ETC was two million dollars higher than the bid of DynaLantic. We set aside the decision of the board and remand the case for reconsideration and with instructions.

OPINION

The facts show that on May 11, 1990, the Air Force Systems Command, Aeronautical Systems Division, issued a solicitation for bids for the procurement of a computerized system for training personnel at the Lowry Technical Training Center, Lowry Air Force Base, Denver, Colorado, in the operation, maintenance and testing of F-16 aircraft. The procurement was to be negotiated for a firm fixed price contract. The solicitation set forth the evaluation criteria against which proposals of bidders would be measured, and informed them that the selection would be made using Air Force Regulation (AFR) 70-30 entitlement "Streamlined Source Selection Procedures" and the evaluation criteria set out in Section M of the solicitation. Paragraph B of Section M provides in pertinent part as follows:

B. Basis for Award.... Evaluation of proposals will be accomplished by comparing proposal elements against approved standards.

1. ... each offeror's proposal will be evaluated in terms of the following specific criteria listed in descending order of importance:

a. Technical

b. Management

c. Cost to the government

As required by Air Force Regulation 70-30, the technical team of the Air Force established objective evaluation standards for the technical and management evaluation areas prior to receipt of the bids. It also established certain standards for use in cost analysis. These cost standards and guidelines included the estimate of the Air Force that 10,000 lines of software had to be developed and that it could be produced at an average of 3 lines of code (LOC) per person per day. These standards were not set forth in the solicitation and were not divulged to the offerors before they made their bids, although they had been previously established by the Source Selection Authority (SSA), a Brigadier General.

Six offerors submitted bids and four of them, including DynaLantic and ETC, were determined to be in the competitive range. Thereafter, the Air Force conducted discussions with these four offerors. ETC bid $4,875,556, and estimated that it could perform the contract by producing 3 lines of code per programmer per day. DynaLantic bid $1,889,807, and estimated that it could perform the contract by producing 10 lines of code per programmer per day. The other two offerors estimated they could perform the contract by producing, respectively, 9 and 12 lines of code per programmer per day. The bid of ETC was higher than that of DynaLantic, because by performing the contract at the rate of 3 LOC per programmer per day it would take ETC longer to produce the 10,000 LOC required by the contract than it would take DynaLantic to perform the contract by producing 10 LOC per programmer per day.

After these proposals were made, the Air Force conducted "face to face" discussions with each of the offerors who submitted proposals within the competitive range, including DynaLantic. These discussions were held so that the agency could advise the offerors of any deficiencies in their proposals and also of any major concerns of the agency with the bids. Under this system, the offerors were to be given an opportunity to satisfy the government's requirements, and to resolve any uncertainties in the terms and conditions of their proposals. The purpose of these discussions was to enable the government to get the most advantageous contract possible. The Air Force was concerned with the low cost of DynaLantic's estimated man-hour software development proposal. At one point they told DynaLantic that "its hardware alone could cost almost as much as its proposed price for the entire contract". The officers said they thought that DynaLantic could reasonably understand from this statement that their concern was software, although it was not discussed. In any event, DynaLantic never knew nor understood that the low cost of software in its bid was a major concern of the Air Force. Consequently, it did not change its software proposal.

The evaluators who were working with General Graham, the Source Selection Authority (SSA), who was the deciding official, advised him that they had concluded that DynaLantic did not understand the contract because its bid was low. The General then selected ETC to receive the award. He stated that the selection was based on the criteria established in Section M of the Request for Proposal "Evaluation Factors for Award". These criteria contained the unannounced "approved" standard of 3 lines of code per programmer per day, which was what ETC had bid, and which was 2 million dollars higher than the bid of DynaLantic. In making the selection, the General stated that all proposals of all bidders in the competitive range were adequate when measured against the above criteria.

DynaLantic filed a protest with the Air Force, which was rejected. It then appealed to the board. The board assigned the protest to an administrative judge (AJ) who conducted a trial. DynaLantic proved at the trial that the standard of 3 lines of code per programmer per day used by the agency had never been approved by anyone having expert knowledge of the computer science or of the production of lines of code such as those contemplated by the contract. This standard was "established" by General Graham who admitted that he was not a computer expert and did not know the computer language to be used to create the software in the contract. He relied on a Lt. Gantt, who admitted he had no expertise in the computer field. Lt. Gantt obtained the 3 lines of code (LOC) standard in a telephone call to an unnamed person in another unidentified division of the Air Force who stated that the 3 line standard was proper because it was being used in another contract in progress. There was no showing that the other contract was similar to the one here. Lt. Gantt then advised General Graham that 3 LOC was the proper standard, but he did not tell the General how or where he got this information nor how he reached this conclusion. The General then established the 3 lines of code per programmer per day as the "approved" standard by which the bid proposals would be measured and under which ETC was ultimately selected to receive the contract award.

DynaLantic produced two computer expert witnesses at the trial who testified that an acceptable and reasonable range of development and production of computer software of the complexity involved in the contract in this case is between 3.8 and 15 lines of code per programmer per day, and that 3 lines of code is unreasonable and 10 lines of code is reasonable. This expert testimony was unchallenged and uncontradicted by the Air Force. It did not put on any testimony in support of its standard of 3 lines of code per programmer per day. The AJ stated in his opinion that he was impressed by both the credentials and the candor of DynaLantic's expert witnesses, yet, for all practical purposes, he ignored their undisputed testimony.

At the close of the evidence, the AJ denied the protest of DynaLantic. He found that "it obviously chose to 'low ball' its proposal in an attempt to obtain the award". We disagree. This finding was uncalled-for, unjustified and not supported by substantial evidence. There is no evidence that DynaLantic's bid was other than genuine. Even the AJ found that General Graham, the deciding authority, considered its proposal to be acceptable. The Air Force admits that DynaLantic was an eligible bidder and that its bid was within the competitive range and was an adequate proposal.

The AJ held that the contract was awarded to ETC in accordance with the evaluation criteria set out in the solicitation even though its bid was higher than that of the low offeror. Again we disagree. The main and controlling evaluation criteria used by the agency in the selection process was its undisclosed standard of 3 lines of code per programmer per day. It says this was the "approved" standard that was referred to in Section M of the solicitation. In our opinion it was not an approved standard within the meaning of the solicitation. This is so because it had never been approved by anyone skilled in computer science or with expert knowledge of the production and development of computer lines of code such as that involved in this case. We hold that the so-called standard was faulty, arbitrary and unreasonable, and the agency erred in using it as the selection standard. The decision of the AJ to the contrary is not supported by substantial evidence.

The AJ held further that DynaLantic's protest was denied because it had not proven that the agency had violated an applicable statute or regulation in the selection process. This holding runs counter to and contradicts the evidence which shows that the agency violated 10 U.S.C. § 2305(b) (4) (A) and 48 C.F.R. § 15.610 by failing to conduct meaningful "face-to-face" discussions with DynaLantic after it submitted its proposal. We point out that 10 U.S.C. § 2305(b) (4) (A) provides in pertinent part:

The head of an agency shall evaluate competitive proposals in accordance with paragraph (1) and may award a contract--

(1) after discussions with the offerors, provided that written or oral discussions have been conducted with all responsible offerors who submit proposals within the competitive range.

and Federal Acquisition Regulation (FAR), 48 C.F.R. § 15.610 states in pertinent part:

(c) The contracting officer shall--

(2) Advise the offeror of deficiencies in its proposal so that the offeror is given an opportunity to satisfy the government's requirements;

(3) Attempt to resolve any uncertainties concerning the technical proposal and other terms and conditions of the proposal.

The evidence shows that the agency failed to comply with these requirements. While it is true that the agency conducted discussions with DynaLantic after it submitted its proposals, they were not meaningful discussions that complied with the requirements of the above authorities. For instance, the agency was concerned that DynaLantic's bid on software was too low, yet it never advised it of this deficiency as required by the regulation. Furthermore, DynaLantic was not given an opportunity to satisfy the agency's requirements because it did not know what its software cost requirements were. No attempt was made by the agency to resolve these uncertainties. Actually, the agency never mentioned software costs during the discussions even though it knew that software costs would be the basis on which the contract award would be made. It says it did not discuss software costs for fear of being guilty of improper technical leveling. This argument is not persuasive. The evidence shows that it freely discussed hardware costs without fear of technical leveling. We see no difference between discussing hardware costs and discussing software costs as far as technical leveling is concerned.

We hold that the Air Force did not hold meaningful discussions with DynaLantic regarding its proposal and that this failure violated the statute and the regulation. The decision of the AJ and the board is arbitrary, an abuse of discretion, contrary to law and unsupported by substantial evidence.

The Air Force says that it did not select low-bidder DynaLantic for the contract award because it concluded that DynaLantic did not understand the contract, since its bid was so low. This was a subjective conclusion that is not supported by substantial evidence. It is our view that if the amount of a bidder's proposal were to be the test by which his understanding of the contract would be measured, ETC's understanding would be much less than that of DynaLantic. This is so because its estimate of 3 LOC per programmer per day was outside of the range of reasonableness that was shown by the uncontradicted testimony of the expert witnesses to be between 3.8 LOC and 15 LOC per programmer per day.

The board concurred with the decision

We hold that the board's decision denying the protest is arbitrary, an abuse of discretion, contrary to law and is not supported by substantial evidence. The decision is set aside and the case is remanded to the board for reconsideration and with instructions to award DynaLantic its bid and proposal costs and reasonable attorney fees.