THIS OPINION WAS INITIALLY ISSUED UNDER PROTECTIVE ORDER AND IS BEING RELEASED TO THE PUBLIC IN REDACTED FORM ON OCTOBER 12, 1994 ___________________________ DENIED: September 26, 1994 ___________________________ GSBCA 12911-P LITTON SYSTEMS, INC., Protester, v. DEPARTMENT OF TRANSPORTATION, Respondent, and COMMUNICATION SYSTEMS TECHNOLOGY, INC., Intervenor. David V. Anthony, Richard P. Rector, Kevin P. Mullen, and Nanci S. Redman of Pettit & Martin, Washington, DC, counsel for Protester. Patricia A. McNall, Gregory C. Carter, George P. Kinsey, and Christian F. P. Jordan, Office of the Chief Counsel, Federal Aviation Administration, Department of Transportation, Washington, DC, counsel for Respondent. Ronald S. Perlman and Ellen F. Randel of Porter, Wright, Morris & Arthur, Washington, DC, counsel for Intervenor. Before Board Judges DEVINE, WILLIAMS, and DeGRAFF. DeGRAFF, Board Judge. In this protest, Litton Systems, Inc. (Litton) alleges that the Department of Transportation, Federal Aviation Administration (FAA) committed three errors in awarding a contract to Communication Systems Technology, Inc. (CSTI). Litton contends that the FAA improperly excluded Litton's alternate proposal from the competitive range, improperly relaxed requirements for CSTI, and did not make a proper best value determination. Because Litton has not established that the FAA violated any statute or regulation in conducting this procurement, the protest is denied. Findings of Fact Radio Control Equipment In this procurement, the FAA is acquiring Radio Control Equipment (RCE). RCE is part of the equipment that the FAA uses to provide communications between air traffic controllers located at FAA control sites and aircraft under their control. The equipment consists of a voice/data communication unit located at the control site and a corresponding unit located at a remote site. Remote sites are typically 200 to 300 miles away from control sites and FAA employees are not usually stationed at remote sites. Some remote sites, for example, are located on tops of mountains. The communication path between the voice switching control equipment (VSCE) at the control site and the remote site equipment is called a channel. Each channel provides the capability to transmit and receive voice and radio signals on up to two frequencies. One frequency (VHF) is used for the control of civilian aircraft and the other frequency (UHF) is used for the control of military aircraft. Each frequency has available a main transmitter, a backup transmitter, a main receiver, and a backup receiver. Protest File, Exhibit 9; Transcript at 30, 811, 1163. In the opinion of an air traffic controller with nearly twenty-three years of civilian and military experience, communications equipment is probably the most important equipment that air traffic controllers use. Without it, they would have no means of controlling aircraft in flight. Transcript at 1108-09. Controllers flip switches in order to change from main transmitters and receivers to backup transmitters and receivers, and they press a push-to-talk button when they need to transmit information to an aircraft. Id. at 1112. If a controller experiences a problem, such as losing a main transmitter on one frequency, the controller will immediately flip the switches to activate the backup transmitters and receivers on both frequencies, and press the push-to-talk button in order to ensure that communications have been re-established. Id. at 1114-15. In certain circumstances, several air traffic controllers control the same frequencies at the same time. Id. at 1117. It is quite usual for air traffic controllers to flip more than one switch at the same time in order to select frequencies, transmitters, and receivers. It would be rare for a controller to select only one switch at a time. Id. at 1119. If air traffic controllers could not flip switches simultaneously and immediately press the push- to-talk button, there would be delayed transmission of information between the controllers and aircraft. One result of a delayed transmission would be that pilots would not be able to hear all of a controller's message, and this would lead to safety problems. Id. at 1122-23. Simultaneous commands, whether from one controller or several controllers, occur all the time in a control facility. Id. at 1123. Pre-Solicitation Events In 1983, the FAA determined that it needed to replace aging, unsupportable RCE with newer, more reliable equipment. In 1985 and 1986, the FAA awarded contracts for the development and production of RCE. In 1991 and 1992, for a variety of reasons, the FAA terminated those contracts. Protest File, Exhibit 6; Protester's Supplemental Protest File, Exhibit 4. After terminating the initial RCE contracts, the FAA reevaluated its needs for RCE and determined that its needs could be satisfied by commercially available products. As a result, the FAA began its Non-Developmental Item (NDI) RCE program. As a part of the NDI RCE program, in March 1992, the FAA distributed a Request for Information (RFI) to vendors. The RFI stated that the FAA wanted to replace aging equipment with new, more capable, equipment. The RFI contained a draft solicitation and asked vendors for their comments and questions. Protest File, Exhibit 6; Protester's Supplemental Protest File, Exhibit 43. Litton responded to the RFI. Part of its response was a list of general questions, including one which asked, in its entirety, "Why is FAA-G-2100e imposed for NDI items available in the commercial marketplace?" Protester's Supplemental Protest File, Exhibit 1. FAA-G-2100e was a standard developed by industry and the FAA and adopted by the FAA that set forth general electrical and mechanical requirements, including power requirements, for FAA equipment. Id., Exhibit 21; Transcript at 363. In April 1992, the FAA responded to vendors' questions concerning the draft solicitation. In response to Litton's question, the FAA responded, "We are soliciting your comments to alternatives. Please be specific in suggesting alternatives, their applications, and the basis for using them." Protester's Supplemental Protest File, Exhibit 44. In May 1992, the FAA decided that the scope of the NDI RCE would have to be severely reduced, due to funding constraints and other agency priorities. As a result, the FAA once again evaluated its need for RCE, and initiated the Down Scoped RCE (DSRCE) program. Protest File, Exhibit 6. The FAA issued a draft solicitation for the DSRCE in April 1993, and Litton responded to that draft solicitation. Concerning one of the power supply requirements (harmonic content), Litton stated, "This requirement adds substantial cost to conventional power supply techniques. Request you reconsider the need for this capability." Protester's Supplemental Protest File, Exhibit 8. The FAA responded to Litton on August 5, 1993, stating, "This is a National Airspace System (NAS) requirement. Specifically, this requirement can be found in FAA-G-2100e, Electronic Equipment, General Requirements. Therefore, this requirement will not be modified." Id., Exhibit 11. The Solicitation On September 1, 1993, the FAA issued solicitation DTAF01-93- R-16456 for DSRCE. Protest File, Exhibit 9. Section B of the solicitation contains a list of contract line item numbers (CLINs). Each CLIN describes an item that the successful offeror would be required to deliver to the FAA, and establishes the number of items to be delivered. As part of their proposals, offerors were to state in Section B their prices for each CLIN. For example, CLIN 1004(a) describes the items to be delivered as "Control Facility DSRCE Channels," establishes the quantity as twelve, and asks offerors to state a price per unit. Protest File, Exhibit 9. Section F of the solicitation contains the schedule for delivering the items required by the contract. Section F explains that delivery of equipment is to begin one month after the FAA orders the equipment. Section F also explains that a DSRCE "system" consists of the equipment required to support one channel. Protest File, Exhibit 9. Currently, the FAA has approximately 3,800 channels operating nationwide, in en route locations, terminal/tower locations, and flight service station locations. Transcript at 594, 618. The solicitation will permit the FAA to purchase the equipment needed for approximately 5,800 channels. Section F also provides that the contract term is five years. Section H.19 of the solicitation requires the successful offeror to provide a ten-year warranty of its equipment. Protest File, Exhibit 9. Section L of the solicitation explains that the FAA would assess each offeror's technical merit by using the offeror's technical equipment proposal in conjunction with an operational capabilities test (OCT). The OCT test plan, which is part of the solicitation, explains that the OCT would consist of automated testing conducted by the FAA using a testbed in a laboratory environment, and would last approximately ten days for each offeror's proposed equipment. Most of the specification requirements were to be verified at OCT, and the remaining requirements were to be addressed in the technical equipment proposal. Section L states that offerors were required to submit three complete systems (control and remote site equipment) for the OCT. The systems had to be submitted by the time written proposals were due. Only systems delivered for the OCT would be evaluated by the FAA, and the FAA would not accept a system as part of an offeror's best and final offer. Offerors were told that the systems they submitted for OCT were to be the same as the systems described in their proposals and the same as the systems that would be produced upon award of the contract. In addition to verifying that an offeror's system met certain specification requirements, the OCT would also be used by the FAA to assess the technical risks associated with the offeror's proposal. Protest File, Exhibit 9. Section L also explains that each offeror was to submit its proposal in seven separate volumes. Volume II was to contain the offeror's technical equipment proposal, and Volume VII was to contain the offeror's price proposal. Other volumes were to contain the offeror's technical services and management proposals. Protest File, Exhibit 9. Section L explains that the technical specification contained "shalls" and "shoulds." "Shalls" were mandatory requirements, and "shoulds" were desired features. All of the "shalls" and "shoulds" are numbered in the technical specification. In addition, the "shalls" are listed in Table M-1 of Section M of the solicitation, and the "shoulds" are listed in Tables M-2, M-3, and M-4 of the solicitation, according to the priority given to the desired features by the FAA. The "shoulds" listed in Table M-2 were the highest priority (Priority 1), those listed in Table M-3 were the next highest priority (Priority 2), and those listed in Table M-4 were the lowest priority (Priority 3). Section M of the solicitation explains that the value of a "should" listed in Table M-2 was equal to the combined value of a "should" listed in Table M-3 and one listed in Table M-4. Section M of the solicitation explains that the "shalls" would be scored on a pass/fail basis, and the "shoulds" would be numerically scored. The tables in Section M list nearly 200 "shalls" and 42 "shoulds." Protest File, Exhibit 9. Section L.6.2.3 explains that each offeror was required to provide a list of all solicitation provisions to which the offeror took exception or from which the offeror deviated. The offeror was supposed to provide a full discussion and detailed explanation of the exceptions and deviations, including an explanation of the benefit to the FAA of each exception and deviation, in order to permit "evaluation" by the FAA. Exceptions and deviations would not make a proposal automatically unacceptable, although a large number of exceptions or significant exceptions providing no benefit to the FAA might result in rejection of a proposal. Protest File, Exhibit 9. Section L.6.2.3. of the draft solicitation issued by the FAA had not contained any comparable language. Protester's Supplemental Protest File, Exhibit 9. Section M.5.2.1 states that the FAA did not intend to make an award to an offeror that failed to meet any of the mandatory requirements. However, the FAA reserved the right to waive any mandatory requirement, if it would be in the FAA's best interest to do so. Protest File, Exhibit 9. In the draft RFP, the FAA had not reserved the right to waive any mandatory requirements. Protester's Supplemental Protest File, Exhibit 9. Section L.10 states that the FAA would award a contract to the responsible offeror whose offer would be most advantageous to the Government, considering cost or price and other factors set forth in the solicitation. Section M.2 provides: This acquisition will utilize the Best Value Approach for selecting an offeror for award. The Best Value Approach is a method of selecting the proposal that represents the greatest value to the Government, based on the evaluation of cost or price and other factors specified in the solicitation. This approach provides the opportunity for a technical/price trade-off and does not require that the award be made to either the offeror submitting the highest rated technical proposal or the offeror submitting the lowest price, although the ultimate award decision may be to either of these offerors. Protest File, Exhibit 9. Section M.2 also explains that each proposal would be evaluated based upon the offeror's written proposals, responses to written questions, and BAFOs. Section M.2 states that the evaluation would address technical equipment/OCT, technical services, management, and price, as well as the vendor's quality control and small business plans. Section M.3 provides that technical equipment/OCT was more important than technical services, technical services were more important than management, and all of these were more important to the FAA than price. Section M.3 states, "Price may become increasingly more important as the difference in technical scores decreases." Protest File, Exhibit 9. Section M.3 explains that in determining which proposal was most advantageous, the FAA would assess the risk presented by each proposal. Specifically, the assessment of risk would consider the degree of uncertainty as to whether each offeror could meet the technical and schedule requirements set forth in the solicitation, and the impact upon the DSRCE program due to the failure to meet any such requirements. Section M.4 establishes three categories of risk. High risk meant that an offeror's system required significant engineering development work in order to be accepted for deployment, and that serious schedule disruption, an increase in cost, or degradation in system performance was likely to occur. Medium risk meant that an offeror's system required some modification or tailoring, and that schedule disruptions, cost increases, or degraded performance might occur. Low risk meant that an offeror's system met the FAA's requirements. Protest File, Exhibit 9. On October 5, 1993, the FAA responded to questions and comments concerning the solicitation. One of Litton's comments expressed the view that it would be more cost-effective for the FAA to acquire equipment that was capable of housing more than one channel instead of acquiring equipment that was capable of housing only one channel. Litton asked the FAA to provide information concerning the number of channels needed at each control and remote site, and to amend the solicitation "to allow pricing both single-channel and multi-channel chassis/power supply configurations." Litton suggested adding CLINs in order to permit pricing of multi-channel units. The FAA responded that it did not have information concerning the number of channels needed at each site, and that it would not amend the solicitation. The FAA also stated that offerors could submit alternate proposals and/or solicitation deviations in accordance with section L.6.2.3. Protest File, Exhibit 10. On October 25, 1993, the FAA responded to additional questions and comments concerning the solicitation. Litton again asked the FAA to provide information so that offerors could calculate prices for supplying multi-channel units. The FAA referred to its October 5, 1993 response, and stated, "If an offeror is interested in submitting an alternate proposal covering such configurations, they are welcome to do so. Any alternate proposal that meets the government [sic] requirements will be evaluated in accordance with Section M of the solicitation." Protest File, Exhibit 13. Specification Provisions The following specification sections are relevant to this protest: Section 3. SYSTEM REQUIREMENTS[foot #] 1 3.1 Definitions. . . . The DSRCE channel shall5 allow control over either frequency one at a time, or over both frequencies simultaneously. . . . The term "frequency," as used throughout this specification, refers to a voice path and its associated control signals that connects the VSCE to a designated transmitter/ receiver. . . . . 3.2 System Characteristics . . . . 3.2.2.1 Channel Operation. A channel shall10 provide for a maximum of two radio frequencies to utilize one four-wire transmission path. Each channel shall11 provide for transmission on one selected frequency, or simultaneously on both frequencies (audio summed), without arbitration, that is, enabling the ----------- FOOTNOTE BEGINS --------- [foot #] 1 Each "shall" is numbered with a subscript. ----------- FOOTNOTE ENDS ----------- corresponding voice path in direct response to a PTT.[foot #] 2 . . . . 3.2.2.3 Radio Control. The control facility DSRCE shall18 accept radio control signals from the VSCE and the DSRCE/VSCE interface, encode these signals, and transmit them to the remote site DSRCE. . . . . 3.2.4.2 DSRCE/Analog Unconditioned Transmission Path. . . . [T]he DSRCE shall103 meet all requirements except end to end system timing if the transmission path introduces a propagation delay of up to 300 ms, and shall104 meet all requirements, including end to end system timing, in the presence of impulse noise as defined below: [definition omitted]. The DSRCE shall 105 meet the following signal level, isolation, and impedance matching characteristics: [characteristics omitted]. . . . . 3.2.5.1 Dimensions. A single DSRCE unit, i.e., a remote site or a control facility unit (each comprising one end of a DSRCE channel), with all associated cables connected shall113 fit into a standard 19" wide rack panel opening, shall114 not exceed a depth of 19" including all connecting cables, and shall115 not exceed a height of 12 1/4". ----------- FOOTNOTE BEGINS --------- [foot #] 2 "PTT" is the abbreviation for push-to-talk. A PTT is a switch that an air traffic controller or pilot activates in order to transmit audio signals. Protest File, Exhibit 9. ----------- FOOTNOTE ENDS ----------- . . . . 3.3 Design and Construction. . . . . 3.3.1.1.1.1 Control Facility DSRCE Power Requirements. The control facility DSRCE shall136 comply with FAA Order 6950.2C with the exception of the inrush current requirement and the power factor requirement. 3.3.1.1.1.1.1 Harmonic Content. The total harmonic content of the input current caused by the DSRCE and fed back into the AC power source shall137 not exceed five percent of the fundamental (60 Hz), and no single harmonic shall138 be greater than three percent of the fundamental. 3.3.1.1.1.1.2 Inrush Current Limiting. The DSRCE inrush current characteristics shall139 fall below the Maximum Inrush/Nominal Peak vs. Duration curve shown in figure 3-3. . . . . 3.3.1.1.1.1.3 Power Factor. The control facility DSRCE shall141 present a power factor to the AC power source of not less than 0.85 leading or lagging when operating under steady state conditions, from 25% to 100% of full load at the nominal line voltage. 3.3.1.1.1.2 AC Line Controls. . . . Each DSRCE assembly which is energized by direct connection to the AC line shall144 have, as a minimum, a front-panel mounted "main power" switch or circuit breaker, a front-panel mounted AC line indicator light, and a front-panel mounted AC line indicating type fuse-holder if a circuit breaker is not provided. Protest File, Exhibit 9. Initial Proposals and Competitive Range[foot #] 3 On November 2, 1993, Litton submitted two complete written proposals. In its basic proposal, Litton offered to supply single-channel units. Protest File, Exhibits 16, 17. Litton failed mandatory requirements and met desired features. Based upon the FAA's review of Litton's basic technical proposal and the results of the OCT, Litton's basic ----------- FOOTNOTE BEGINS --------- [foot #] 3 In addition to Litton and CSTI, two other vendors submitted proposals. These two other vendors' proposals are not material to the resolution of the merits and, therefore, we have not included any information concerning the proposals in this opinion. ----------- FOOTNOTE ENDS ----------- proposal was rated as a risk and Litton's basic proposal's technical score was . Id., Exhibits 22, 26. In addition to its basic proposal, Litton submitted an alternate proposal titled, "DSRCE Proposal For A Multichannel System." Litton's alternate technical proposal explains that, according to Litton's calculations, it would be more economical to install a multi-channel unit than to install single-channel units if more than four channels are required at a control site or a remote site. Protest File, Exhibit 18. Litton made certain assumptions concerning the number of sites where multi-channel units would be installed and the number of sites where single- channel units would be installed, developed its price based upon its assumptions, and described its assumptions in its pricing proposal. Id., Exhibit 19; Transcript at 45-47. Litton delivered a multi-channel unit, but not a single-channel unit, for the OCT of its alternate proposal. Transcript at 347-48. At the time evaluations took place, the FAA's technical evaluation team chairperson did not realize that Litton's alternate proposal offered to supply a mix of multi-channel units and single-channel units. Id. at 339. Based upon this OCT and the alternate technical proposal, the FAA determined that Litton failed mandatory requirements and met desired features. Based upon the FAA's review of Litton's alternate technical proposal and the results of the OCT, Litton's alternate proposal was rated as a risk and its technical score was . Protest File, Exhibits 22, 26. On November 2, 1993, CSTI submitted its proposal. Although CSTI included alternate pricing information for two multi-channel units, it did not provide information in its technical proposal about these units and did not submit them for OCT. Protest File, Exhibits 14, 15. CSTI's proposal was rated as a risk. CSTI failed mandatory requirements and met desired features. CSTI's technical score was . Id., Exhibits 22, 26. The FAA performed cost evaluations, in addition to technical evaluations. The results of both evaluations were as follows: Offeror Tech. Score Risk Total Price CSTI $ mill Litton/basic $ mill Litton/alt $ mill Protest File, Exhibits 22, 26. When the FAA performed its cost evaluation, it did not realize that Litton's alternate proposal price was based upon supplying a particular number of multi- channel units and a particular number of single-channel units. If Litton had supplied all single-channel units at the prices set forth in its alternate proposal, its price would have been approximately $ million. If Litton had supplied all multi- channel units at the prices set forth in its alternate proposal, its price would have been approximately $ million. Transcript at 670-71, 685-90. On May 6, 1994, the FAA notified CSTI that its proposal was being included within the competitive range, although the FAA was not interested in CSTI's alternate proposal for multi-channel units. Protest File, Exhibit 27. That same date, the FAA notified Litton that its basic proposal was being included within the competitive range. Id., Exhibit 28. Also on May 6, 1994, the FAA notified Litton that its alternate proposal would not be included within the competitive range for two reasons. First, the FAA told Litton, the alternate proposal "deviates from the FAA solicitation since it proposes a multi-channel unit rather than the solicited single channel unit. A multi-channel unit does not support/address the FAA's program strategy, which is primarily to replace aging and unsupportable equipment on a channel by channel basis . . . ." Also, the FAA told Litton, the multi-channel equipment "does not meet the specification equipment dimensions . . . ." Protest File, Exhibit 29. Litton was "kind of surprised" by the FAA's statements that Litton's alternate proposal did not meet the FAA's program strategy. Litton "couldn't find anyplace that they had advised us of their program strategy . . . . We didn't know what that was and didn't know why we didn't meet it. And further, that [sic] the portion that talks about unsupportable equipment on a channel-by-channel basis is incorrect. We do have the ability to add channels on a channel-by-channel basis . . . ." Transcript at 60. Litton believed that the specification permitted either single-channel or multi-channel equipment. Id. at 60-61. Litton knew that its alternate proposal did not meet the size limitations imposed by specification section 3.2.5.1. Id. at 62. Litton's alternate proposal did not take exception either to the specification's size limitations or to any other specification section that affected Litton's ability to meet the size limitations. Nor did Litton's alternate proposal suggest why it might be to the FAA's advantage to accept an oversized system. Protest File, Exhibit 18. Discussions On May 6, 1994, when the FAA notified CSTI and Litton that their single-channel proposals were being included within the competitive range, the FAA also sent CSTI and Litton lists of discussion questions to which they were to respond in writing. CSTI and Litton asked the FAA to clarify some of the discussion questions and, after the FAA did so, CSTI and Litton provided their responses. Protest File, Exhibits 27, 28, 30-35. CSTI Among the discussion questions presented to CSTI were questions concerning six mandatory requirements: Shalls 103 and 104 ( ), Shalls 137 and 138 ( ), Shall 139 ( ), and Shall 141 ( ). Protest File, Exhibit 27. CSTI's proposal took exception to Shalls 136, 137, 138, and 141. CSTI stated that, by taking exception to these requirements, it would be able to use a conventional, commercial, lower cost power supply.[foot #] 4 CSTI's proposal notes that CSTI thought that it met the requirements of Shall 139. Id., Exhibit 15. Although CSTI's proposal takes exception to Shall 144, the FAA did not ask CSTI any discussion questions about this requirement. Id., Exhibits 15, 27. On May 16, 1994, CSTI presented its response to the FAA's discussion questions. CSTI explained that, in order to prepare its response, it leased a test unit like the one used by the FAA at OCT. CSTI provided the FAA with test results from the test unit and from a piece of CSTI test equipment for Shalls 103 and 104. In addition, CSTI explained . Protest File, Exhibit 34. Concerning Shalls 137, 138, 139, and 141, CSTI explained that it had located a power supply which would meet the harmonic content, inrush current, and power factor requirements contained ----------- FOOTNOTE BEGINS --------- [foot #] 4 The cover page of CSTI's initial proposal contains a restrictive legend stating that the proposal contains data that CSTI did not want the FAA to disclose. The cover page to CSTI's list of exceptions, which does not mention any of the exceptions taken by CSTI, is marked with a reference to the restrictive legend. The pages following the cover page to CSTI's list of exceptions, which explain the exceptions taken by CSTI, are not marked with a reference to the restrictive legend. Protest File, Exhibits 14, 15. ----------- FOOTNOTE ENDS ----------- in FAA-G-2100f, and that it would include this replacement power supply in its best and final offer (BAFO). Protest File, Exhibit 34. FAA-G-2100f, released on November 15, 1993, replaced FAA-G- 2100e. FAA-G-2100f revised several power requirements, including requirements for harmonic content, inrush current, and power factor. Protester's Supplemental Protest File, Exhibit 21. FAA- G-2100e contained standards that did not distinguish between equipment that consumes large amounts of power and equipment that consumes small amounts of power. FAA-G-2100f established standards that are based upon the amount of power consumed by equipment. Equipment that consumes a large amount of power must meet certain standards, and equipment, such as CSTI's DSRCE, that consumes less power must meet less stringent standards. Id., Exhibit 21; Transcript at 413. The revision from FAA-G-2100e to FAA-G-2100f was based upon a study conducted by engineers from the FAA and from private industry. Transcript at 1015-17. CSTI stated that the replacement power supply "will be provided designed (sic) by Resonant Power Technology (RPT), Inc. (San Jose, CA). This power supply will be based on a larger, 100W supply currently available from RPT, but will be scaled down in size and in wattage capability. . . . CSTI has tested the larger, 100W power supply to develop confidence in RPT's ability to provide a compliant supply. The 100W unit performed well against tested parameters of power factor and harmonic content." Protest File, Exhibit 34. In fact, the unit as tested by CSTI met the power factor requirement of Shall 141. Id.; Transcript at 1229-30. CSTI included the test results in its response to the discussion questions. The test results explain that, because the unit tested was a 100 watt unit, CSTI had loaded it appropriately to approximate the results that could be expected from a smaller unit. CSTI explained that RPT is a company that specializes in power supplies. CSTI sent RPT a power supply specification that included the FAA's requirements, and RPT provided a price quote. The quote included a one-time development charge, because RPT was going to scale down its 100 watt power supply to a 50 watt power supply and provide the smaller power supply to CSTI. Protest File, Exhibit 34. If a manufacturer builds a power supply for a particular use and then prices the power supply commercially, the FAA considers that the power supply is a commercial power supply. In evaluating risk associated with such a power supply, the FAA considers whether the new product is a derivation of an existing product. Transcript at 387-88. The FAA does not believe that downsizing a power supply from 100 watts to 50 watts should present a problem for a power supply manufacturer. Id. at 935. An expert who testified for the FAA is of the opinion that it is not a very complex task to scale down a 100 watt power supply to a 50 watt power supply. Id. at 1094. Litton Among the discussion questions presented to Litton were questions concerning Shalls 11 and 18 (during OCT, Litton's system experienced delayed responses to multiple command state changes and ignored the mute release command when it was executed simultaneously with a state change of main/standby signal), Shalls 103 and 104 (during OCT, Litton's system experienced an increasing number of signal failures as the transmission delay increased and in the presence of impulse noise), Shall 105 (during OCT, Litton's system did not meet impedance and isolation requirements), Shalls 137 and 138 (during OCT, Litton's system did not meet the harmonic content requirements), and Shall 141 (during OCT, Litton's system did not meet the power factor requirements). Protest File, Exhibit 28. Litton's proposal did not take exception to any mandatory requirements. Id., Exhibits 16, 17. When Litton asked the FAA to clarify its questions concerning Shall 11, Litton stated that its system was designed to "detect simultaneous commands as a system fault and re- configure the system accordingly." It also stated that it could modify its system so that simultaneous commands were processed sequentially. Litton asked the FAA to identify the specification section that required it to accept and act upon simultaneous commands. Concerning Shalls 137, 138, and 141, Litton asked the FAA to provide its test data because, according to Litton, it believed that its power supply met the harmonic content and power factor requirements. Protest File, Exhibit 32. The FAA responded by telling Litton that the simultaneous command generation requirement was contained in Shall 5 and Shall 11, and by providing the test data requested. Id., Exhibit 33. On May 16, 1994, Litton provided its responses to the FAA's discussion questions. Concerning Shalls 5, 11, and 18, Litton took issue with the FAA's interpretation of the specification. Litton did not read the solicitation to require simultaneous processing of a combination of all commands. Litton explained that its system allowed controllers to talk on two frequencies at the same time, in response to a controller pressing the push-to- talk button. Litton stated, "More complex operation would have required considerably more specifications which do [not] exist." Litton explained that its system was designed "to act upon one VSCE stimulus at a time. . . . Any other combination of simultaneous command generation is unlikely and impractical in a human interface environment and is assumed to be an error. . . . Simultaneous commands are accepted. These commands are acted upon one at a time . . . ." Protest File, Exhibit 35. Litton later amended its responses to delete the statement that simultaneous commands are accepted by its system and acted upon one at a time. Id., Exhibit 59. Concerning Shalls 103 and 104, Litton explained that it believed that its system met these requirements. The head of the FAA's technical evaluation team understood Litton's response to Shall 103 to suggest that the FAA's test method had, perhaps, caused Litton not to meet the requirements of Shall 103. The technical evaluator contacted the OCT test facility and confirmed that the FAA's method of testing should not have caused Litton to fail the test for Shall 103. As far as Shall 105 was concerned, Litton stated that its system could meet this requirement if Litton were to make a wiring change. Litton had been permitted to make a change during OCT that enabled Litton to comply with some, but not all, of the requirements of Shall 105. Concerning Shalls 137, 138, and 141, Litton stated that it had conducted its own tests and that it believed that its system met these requirements. Protest File, Exhibit 35; Transcript at 614, 789, 807-08. Best And Final Offers CSTI and Litton submitted their BAFOs on May 24, 1994. Protest File, Exhibits 36, 37. CSTI CSTI offers a modern, state-of-the-art design which is clean, robust, innovative, and unique. The system can be reconfigured by making software changes instead of hardware changes, and this can be done from the control site. CSTI's system is designed to process information digitally, which is more reliable than analog processing. As voice and data signals enter CSTI's DSRCE, they are converted to digital numbers and processed in the digital computer that is part of CSTI's design. Most, if not all, new communications equipment is digital-based. A computer can process digital information more flexibly, more efficiently, more accurately, and more precisely than can analog hardware. Utilizing the digital capability of a computer enables a user to accomplish some tasks, such as adjusting for distortion and remote testing, that are impossible using analog systems. CSTI designed its system to be flexible and expandable to meet the FAA's current and future needs. For example, if the FAA replaces its telephone lines with digital lines, CSTI's system, unlike an analog system, will be able to take advantage of that replacement. CSTI's digital approach permits it to adjust signal levels automatically instead of manually. CSTI's data channel transmits information at bits per second. CSTI has taken the precaution of using a technique called "surface mounting" of its components, which increases reliability. Due to the technology utilized by CSTI in its design, if a CSTI unit needs to be repaired, there is no need to adjust any levels or make any tone measurements before making the repair, and this means the time needed to make a repair will be decreased. CSTI's system incorporates proprietary software that will help abate wide band noise and enhance voice quality. Protest File, Exhibit 40; Transcript at 609, 877-79, 1005, 1040-41, 1044, 1052, 1064, 1070, 1077, 1153, 1167-71, 1174. CSTI's system provides two particular features that were neither mandatory requirements nor desired features in the solicitation. The system is capable of generating a test tone that can be used by the FAA's technicians to align the equipment remotely or to perform remote troubleshooting. The system also provides remote maintenance monitoring of some transmitters. These features were not considered as part of the technical evaluation, although they provide added value to the FAA. CSTI was awarded no points for these features. Protest File, Exhibits 40, 42. Desired Features CSTI's system offered of the forty-two desired features listed in the solicitation. Of the desired features, were designated as Priority 1 features, were Priority 2 features, and was a Priority 3 feature. After evaluating these features, the FAA determined that CSTI's technical proposal scored points.[foot #] 5 This figure is the sum of points for technical equipment, points for technical services, and points for management. Protest File, Exhibit 38. CSTI's system meets the requirements of Should 12, which includes six Priority 1 desired features, one Priority 2 desired feature, and one Priority 3 desired feature. By meeting Should 12, CSTI's system provides the FAA with Centralized Maintenance Terminal (CMT) capability. Instead of sending technicians to remote sites to perform maintenance, alignment, and troubleshooting functions, the FAA will be able to perform these functions remotely by using a personal computer. CMT capability will permit the FAA to reduce the amount of travel to remote sites. Protest File, Exhibits 9, 40; Transcript at 643-44. This capability will save the FAA time and money. Transcript at 611. CSTI's system meets the requirements of Shoulds 6, 7, and 11, which are Priority 1 desired features, and Shoulds 8, 9, and 10, which are Priority 2 desired features. By meeting these requirements, CSTI's system provides the FAA with Remote Site Switching (RSS) capability. This capability provides what the FAA refers to as "diversity," which is a method for ensuring that there is a backup communications path if the primary communications path stops working. This capability includes fault sensing and automatic switching to another telephone line. RSS capability is important to the FAA because it provides a reliable, available, sustainable means for communication between controllers and aircraft. It ensures that controllers and pilots can talk to one another, even if a primary communication line fails. Protest File, Exhibits 9, 40; Transcript at 612, 644-45, 1178, 1207. ----------- FOOTNOTE BEGINS --------- [foot #] 5 As mentioned above, Section M of the solicitation explains that the mandatory requirements would be scored on a pass/fail basis, and only the desired features would be numerically scored. Protest File, Exhibit 9. ----------- FOOTNOTE ENDS ----------- Mandatory Requirements In its BAFO, CSTI states that its proposal does not meet mandatory requirements.[foot #] 6 In its BAFO, CSTI . CSTI took exception to four of the remaining six requirements (Shalls 136, 137, 138, and 141) concerning power supply, and said that it could meet the requirements of FAA-G-2100f, instead of the requirements contained in the solicitation. CSTI stated that it could use a less expensive, smaller power supply if it was permitted to meet the requirements of FAA-G-2100f. Protest File, Exhibit 36. CSTI took exception to Shall 141 because it was not entirely certain that its replacement power supply would meet this requirement, although it had met the requirement during CSTI's post-OCT testing. CSTI included in its BAFO a price quote from RPT, agreeing to supply the replacement power supplies in exchange for payment of certain non-recurring engineering costs plus a fixed price per unit. Id., Exhibits 34, 36; Transcript at 1229-30. CSTI also took exception to Shalls 104 and 144, explaining that it fulfilled most, but not all, of those two requirements, and stating what it believed were the benefits to granting an exception to those requirements. CSTI's BAFO states that it meets the requirements of Shall 139. Protest File, Exhibit 36. In its technical evaluation report regarding CSTI's BAFO, the FAA states that CSTI's failure to meet the mandatory requirements that could be met presented a risk. The FAA also decided that CSTI's failure to meet three shalls (Shalls 137, 138, and 139) regarding power supply presented a risk because these mandatory requirements contained the standards set in FAA-G-2100e which had been revised to FAA-G-2100f, and CSTI could conform to the requirements of FAA-G-2100f. The FAA's technical evaluation report lists Shalls 104 and 141 as being met by CSTI at BAFO, although CSTI took exception to part of Shall 104. The technical evaluation report also states that CSTI's failure to meet Shall 144 presents a risk. Protest File, Exhibits 36, 38; Transcript at 678, 1010. The technical equipment portion of CSTI's proposal was evaluated as being a risk. Protest File, Exhibit 38. Although CSTI made some changes to its system after OCT, CSTI explained and documented these changes in its response to the FAA's discussion questions and in its BAFO. Based upon the information provided by CSTI, the FAA did not believe that CSTI would need to make further changes to its system in order to ----------- FOOTNOTE BEGINS --------- [foot #] 6 The pages of CSTI's BAFO were marked the same as the pages of its initial proposal, described in footnote 4. ----------- FOOTNOTE ENDS ----------- complete necessary testing and to deploy the equipment. Transcript at 209-10, 286. Overall, CSTI's technical proposal was evaluated as being a risk. Protest File, Exhibit 38. In making its risk assessment, the FAA did not assume that CSTI met the mandatory power requirements contained in the solicitation. The FAA did not change the power requirements contained in the solicitation. Transcript at 1010. Price The FAA determined that CSTI's price, including the optional CLINs, was $ . CSTI's price includes $ that Litton's price does not include. Protest File, Exhibit 39. Litton Litton's system design uses "old and outdated techniques." Transcript at 1040. Litton designed its system in a traditional manner that limits the system's flexibility and expandability. The system's design would not aid the FAA in fulfilling future needs. Protest File, Exhibit 40; Transcript at 609. Reconfiguring the system will require making hardware adjustments, and this requires traveling to remote sites. Protest File, Exhibit 40; Transcript at 878-79. In addition, because Litton's system is an analog system and not a digital system, it contains a greater number of component parts than does CSTI's system, which makes Litton's system less reliable than CSTI's system. Litton's system contains approximately 580 parts, and CSTI's system contains approximately 380 parts. The less reliable a system is, the more travel will be required to remote sites to make repairs. Transcript at 877-78, 882, 1040, 1063. Industry is moving away from analog systems. Id. at 1005. Litton's system processes voice and data signals by using hardware that is outside and independent of its computer. Id. at 1042. Litton's system requires manual adjustments to hardware in order to adjust signal levels. Id. at 1044-45. Litton's data channel transmits information at bits per second. Id. at 1052. Due to the technology utilized by Litton in its design, if a Litton unit needs to be repaired, the FAA will need to adjust signal levels and make tone measurements before making the repair, and this means the time needed to make a repair will be increased. Id. at 1072. Desired Features Litton's system offered of the forty-two desired features listed in the solicitation. Of the desired features, were designated as Priority 1 features and were Priority 2 features. After evaluating these features, the FAA determined that Litton's technical proposal scored points. This figure is the sum of points for technical equipment, points for technical services, and points for management. Protest File, Exhibit 38. Litton's system does not provide CMT or RSS capability. Id., Exhibit 40; Transcript at 1009. Mandatory Requirements In its BAFO, Litton stated that it complied with all mandatory requirements, and did not take exception to any requirement. Protest File, Exhibit 37. In its technical evaluation report regarding Litton's BAFO, the FAA determined that Litton's proposal did not meet the requirements of Shalls 11 and 18, which were high risk; Shalls 103 and 105, which were medium risk; and Shall 141, which was high risk. Id., Exhibit 38. The FAA later decided that Litton met the requirements of Shall 141. Transcript at 610. The FAA believed that Litton would have to redesign portions of its system in order to comply with the solicitation's requirements, and that this would increase the risk of being able to deploy the equipment in a timely manner. Protest File, Exhibit 40. Further, the FAA interpreted Litton's responses to the discussion questions to mean that the FAA would have to issue a change order after contract award if the FAA wanted Litton to meet the requirements of the solicitation, as the FAA interpreted them. Transcript at 407-08, 414-16. The technical equipment portion of Litton's proposal was evaluated as being a risk. Overall, Litton's technical proposal was evaluated as being a risk. Protest File, Exhibit 38. The head of the FAA's technical evaluation team interprets Shalls 11 and 18 in light of Shall 5, which requires that the DSRCE system provide control over either frequency one at a time, or over both frequencies simultaneously. Shall 5 also explains that a frequency includes both voice and other commands. The FAA's evaluator considered several features of Litton's system in light of Shalls 5, 11, and 18: If a controller attempts to exert control over a frequency by executing several commands nearly simultaneously, Litton's system processes the first command, waits until it receives confirmation that the first command has been received, then processes the second command, waits until it receives confirmation that the second command has been received, and so on until the fifth command is processed. If a controller switches from the main transmitter and receiver to the standby transmitter and receiver for both frequencies and then presses the push-to-talk button, Litton's system will not allow the controller to talk to aircraft until the first signals, switching from one set of transmitters and receivers to another set of transmitters and receivers, have been sent and confirmed. In addition, if two signals are sent simultaneously instead of nearly simultaneously, Litton's system considers this to be an error and it reconfigures itself, delaying further the enabling of a voice path in response to a push-to-talk command. In the FAA's view, Litton's system does not meet the requirements of Shalls 5, 11, and 18, because it arbitrates among signals, it does not provide complete control over frequencies simultaneously, and it does not enable a corresponding voice path in direct response to a push-to-talk command. Transcript at 819- 23, 828. An expert witness who testified for the FAA interprets Shalls 5, 11, and 18 as does the FAA. He finds the requirement for simultaneous commanding to be stated "right up front" in the specification, in Shall 5. He described this requirement as "a fundamental requirement for air traffic control." Transcript at 1046-48. He explained that, according to Litton's proposal, multiple commands by an operator will cause Litton's system to think that an error has occurred, and Litton's system will reset. In the expert's opinion, this does not meet the requirement for simultaneity. Also, if two signals are nearly simultaneous, Litton's system will process those signals sequentially, which, in the expert's view, does not meet the requirement for performing without arbitration. The FAA expert also explained, clearly and in great detail, why he believes Litton's system is not capable of meeting the requirement for processing signals within a certain required time when the signals are asserted simultaneously or nearly simultaneously. Basically, this occurs because Litton's system transmits information at a very slow rate, and because Litton's system uses a signalling technique that has been in use for thirty-five years and has limited capabilities. Id. at 1049-62. The FAA's expert witness believed that major design changes to Litton's basic architecture would be required in order to meet these mandatory requirements. Id. at 1062. Litton's expert witness interprets Shalls 11 and 18 to require simultaneity only for push-to-talk commands, and not for any combination of any number of commands. He believes that it would be possible for a controller to want to invoke more than one control or command signal at the same time. He does not know whether it is possible for more than one controller to invoke control signals on the same channel at the same time. He believes that Litton's noncompliance with Shall 103 might have been due to a difference in the test procedures used by the FAA and by Litton. He said that Litton needed to make a simple wiring change in order to comply with Shall 105. He believes that Litton could make a software change and comply with the FAA's interpretation of Shall 11. He does not know how much engineering effort would be required in order to make Litton's system compliant with the FAA's interpretation of Shall 11. A Litton employee told Litton's expert what would be required is "the actual coding in software, testing of the software," and incorporating the new software into Litton's single-channel unit. Transcript at 458-59, 473, 486-87, 491-92, 503-04. Price The FAA determined that Litton's price was $ . Best Value Determination The FAA appointed a technical evaluation team to review the technical capabilities of the CSTI and Litton systems. The team prepared a report that summarizes the results of OCT and all parts of the written proposals, other than price. The report explains how each part of the proposals was scored, discusses the risk associated with each proposal, and lists the mandatory and desired features provided by each offeror. Protest File, Exhibit 38. The contract specialist prepared a report, in which the contracting officer and his supervisor concurred, which analyzes the prices offered by CSTI and Litton. Id., Exhibit 39. The FAA's Source Evaluation Board (SEB) reviewed the technical and cost reports, and prepared a report that combines the results of the technical and cost reports, and compares the proposals of the two offerors. Id., Exhibit 40. The SEB presented its report to the source selection official and briefed him concerning its conclusions. Id., Exhibit 41. The source selection official is the FAA's Associate Administrator for National Airspace Development. His organization has a budget of approximately $2 billion, and procures equipment, including communications equipment, for the FAA's national system. He has responsibility for preparing budget estimates for the programs under his control. Transcript at 984-86. The FAA assumed that CSTI's system would cost nearly $ million more than Litton's system. The FAA considered that, in accordance with the provisions of the solicitation, technical factors were more important than price. The FAA also considered that it was supposed to acquire the system that provided the best value, and not the system that cost the least. In comparing the two systems, the FAA was impressed by the flexibility and expandability of the CSTI system, which is software-based, as compared to the rigidity of the Litton system, which is hardware- based. It considered the fact that CSTI's system would provide CMT and RSS capability and Litton's system would not provide either capability. Also, the FAA considered that there were technical and schedule risks associated with the Litton system that were not associated with the CSTI system. In the FAA's view, the award to CSTI was justified even if there were no differences between the offerors' abilities to meet mandatory requirements. The FAA decided that the CSTI system provided the best value because it offered the most advantageous combination of features for the price, with lower risk. Protest File, Exhibits 40, 42; Protester's Supplemental Protest File, Exhibit 37. At the time the FAA decided that the CSTI system provided the best value, the FAA did not make any written calculations of the dollar value of the features offered by CSTI's system and not offered by Litton's system. Transcript at 1023. The source selection official believed that CSTI's system was worth the price, based upon his knowledge of the features of CSTI's system and the importance of those features to the FAA's programs. Id. at 1024-25. As discussed below, after award, the FAA prepared written estimates of the value of the CMT and RSS capabilities offered by CSTI. The head of the FAA's technical evaluation team estimated that the value of the RSS capability offered by CSTI is approximately $6 million to $6.5 million over ten years. This estimate is based upon an acquisition plan that the FAA prepared for a separate procurement of equipment to provide RSS capability, and a budget estimate that was prepared in connection with the acquisition plan. Because Shoulds 6 through 11 basically incorporate the capabilities outlined in the RSS acquisition plan, there should be no need for a separate acquisition of additional equipment to provide RSS capability. Transcript at 854-56; Protest File, Exhibit 49. The FAA's source selection official estimated that the value of the RSS capability offered by CSTI is approximately $12 million over ten years, because the DSRCE contract permits the FAA to acquire approximately twice as much equipment as is called for in the RSS budget estimate. The source selection official also stated that, because of the RSS capability provided by CSTI, the FAA will not need to award a separate contract for RSS capability. Transcript at 994-96. The source selection official did not believe that the RSS capability provided by CSTI will precisely match the requirements of the RSS draft specifications. But, he testified that the RSS procurement can be cancelled if the DSRCE contract is awarded to CSTI. Id. at 1020-22. At the hearing, CSTI's expert witness, a CSTI employee who designed CSTI's system, identified one requirement contained in the draft RSS specification that CSTI's system does not meet. She also testified that CSTI could meet this requirement by making some software changes. Also, she testified that she is not sure whether CSTI's system meets one additional requirement that is contained in the RSS draft specification. Id. at 1266-69. The head of the FAA's technical evaluation team estimated that the value of the CMT capability offered by CSTI is approximately $1.3 million over ten years. CMT capability will permit the FAA to conduct several functions, including the alignment of communications circuits, with one technician at its control sites. Currently, alignment is accomplished by two technicians, one at a control site and the other at a remote site. The estimate is based upon the amount of time that will be saved if one technician can accomplish the alignment task that, without CMT capability, requires two technicians. The estimate takes into account only one of the three types of control sites, and does not take into account either any travel time required by the technician who presently has to travel to the remote site or any travel-related expenses. The $1.3 million estimate does not include the cost of travel time because trips to remote sites might be needed in order to perform maintenance functions other than alignment. Transcript at 970. The head of the technical evaluation team calculated that if all of the travel time could be saved, the value of the CMT capability offered by CSTI is approximately $4.1 million over ten years. Id. at 862-65, 975; Protest File, Exhibit 51. The source selection official estimated that, conservatively, the value of CSTI's CMT capability is $2 million, which includes a savings of some, but not all, travel time. Transcript at 999-1001. The source selection official estimated that this estimate would double if it included all three types of control sites. Id. at 1031. At the hearing, CSTI identified some maintenance functions that its CMT capability cannot now provide. CSTI stated that it could modify its software in order to provide these functions. Id. at 1278-81. Award and Post-Award On July 12, 1994, the FAA awarded the DSRCE contract to CSTI. As part of the award, the FAA accepted CSTI's exceptions to Shalls 104, 136, 137, 138, 141, and 144. Protester's Supplemental Protest File, Exhibit 32. On July 14, 1994, the FAA debriefed Litton concerning the award to CSTI. According to Litton's notes, during the debriefing, the FAA explained that it had "waived" Shalls 137-139 for CSTI. Id., Exhibit 37. On July 29, 1994, the FAA told CSTI to stop work, due to this protest. Id., Exhibit 38. On August 10 and 24, 1994, Litton received quotes from two power supply manufacturers. Protester's Supplemental Protest File, Exhibit 48. According to Litton, if it had been allowed the same exceptions to power supply requirements as was CSTI, it could have purchased power supplies from these vendors, incorporated them into its alternate proposal, and met the specification's size requirements. Litton also contends that, if it had been allowed the same exceptions as CSTI, it could have incorporated power supplies from these two vendors into its basic proposal and reduced its price by approximately $ . Transcript at 65-66. Litton has not conducted any tests of its equipment using a replacement power supply. Id. at 150. Litton asserts, generally, if it had known from the outset that the FAA did not wish to purchase multi-channel units, it could have added more desirable features to its basic proposal. Specifically, Litton says that it could have "investigated" the possibility of adding CMT capability to its basic proposal. Transcript at 67. Litton does not know how much it would have cost to add either CMT or RSS capabilities to its basic proposal. Transcript at 80-81. In late July 1994, CSTI took delivery of the first power supply unit from RPT. Transcript at 1233. RPT has assigned a part number to the scaled-down power supply and intends to market it to other customers. Id. at 1229. RPT's delivery schedule will permit CSTI to meet the contract's delivery schedule. Id. at 1288. If there had been no stop work order, CSTI believes that RPT could have delivered power supplies in time to permit CSTI to meet the contract's delivery schedule. Id. at 1292. Discussion Litton alleges that the FAA committed three errors in conducting this procurement. First, Litton contends that the FAA improperly excluded Litton's alternate proposal from the competitive range. Second, Litton contends that the FAA improperly relaxed the specifications for CSTI, but not for Litton's basic proposal. Third, Litton contends that the FAA did not make a proper best value determination. As explained below, we do not find that the FAA violated statute or regulation, and we reject all of Litton's contentions. 40 U.S.C. 759(f) (1988). I. Litton's Alternate Proposal Litton's first ground of protest is that the FAA improperly eliminated Litton's alternate proposal from the competitive range. The Federal Acquisition Regulation provides that a proposal should be included within the competitive range if, based upon price and other factors stated in the solicitation, the proposal has a reasonable chance of being selected for award. 48 CFR 15.609(a) (1993) (FAR 15.609(a)). Litton's alternate proposal was excluded from the competitive range for two reasons. First, the FAA stated, because Litton offered a multi-channel unit, its proposal deviated from the solicitation and did not support the FAA's program strategy for replacing equipment on a channel-by-channel basis. Second, the FAA stated, Litton's multi-channel unit exceeded the dimension requirements of the specification. Litton argues that neither of the reasons for excluding its alternate proposal from the competitive range is valid. It argues that the first reason is invalid because the solicitation does not explicitly state that multi-channel units were not acceptable, and does not explain the agency's strategy for replacing equipment. It argues that the second reason is invalid because the FAA misled Litton into believing that the FAA would not permit exceptions to the specification's power requirements. Litton contends that it did not realize that the FAA would permit exceptions to the power requirements until it learned, post- award, that the FAA had allowed CSTI to take exception to some of the power requirements. Litton contends that, if it had been allowed to take the same exceptions as CSTI, it could have purchased a smaller power supply for its alternate proposal and met the specification's dimension requirements. In addition, Litton asserts that its alternate proposal should have been included within the competitive range because it was - highest technically by a small margin and in price by a significant margin. FAA and CSTI argue that Litton's protest of its exclusion from the competitive range is untimely. They also argue that, if Litton's protest is timely, Litton's proposal was properly excluded from the competitive range. Timeliness Our rules provide that protests challenging the terms of a solicitation must be filed before proposals are due, and protests raising other challenges to the procurement process must be filed "no later than 10 working days after the basis for the ground of protest is known or should have been known, whichever is earlier." Rule 5(b)(3).[foot #] 7 Litton received a letter on May 6, 1994, notifying it that the FAA had decided to exclude Litton's alternate proposal from the competitive range. The letter clearly states the two reasons upon which the FAA based its decision. Litton did not protest the agency's action until it filed this protest on July 22, 1994. Litton's challenge to the first reason given by the FAA for excluding Litton's alternate proposal from the competitive range -- that Litton's multi-channel units were not acceptable and did not meet the FAA's program strategy -- is untimely. Litton was surprised by the statement contained in the FAA's May 6, 1994 letter, that Litton's alternate proposal did not meet the FAA's program strategy. Litton knew that the FAA had stated that it would evaluate multi-channel proposals. Litton did not believe that the FAA had ever informed vendors of its program strategy, and Litton did not understand what program strategy it allegedly did not meet. Litton disagreed with the May 6, 1994 letter's conclusion that Litton's alternate proposal would not allow the FAA to replace equipment on a channel-by-channel basis. Also, Litton believed that the FAA's letter was wrong when it stated that the solicitation permitted only single-channel units. Thus, on May 6, 1994, Litton believed that the FAA had imposed a program strategy requirement not found in the solicitation, had misunderstood Litton's proposal, and had misread the solicitation. On May 6, 1994, Litton knew all that it needed to know if it wished to file a protest challenging the first reason given by the FAA for excluding Litton's alternate proposal from the competitive range. Because Litton waited nearly three months after receipt of the May 6, 1994 letter to file this protest, the portion of this protest that seeks to challenge the first reason given by the FAA for excluding Litton's alternate proposal from the competitive range is untimely. Litton's challenge to the second reason given by the FAA for excluding Litton's alternate proposal from the competitive range -- that Litton's multi-channel units exceeded the specification's dimension requirements -- is also untimely. Litton argues that this ground of protest is timely because it filed its complaint within ten days after learning that the FAA, as part of the award of the contract, permitted CSTI to take exception to some of the specification's power requirements. In order to determine whether this ground of protest is timely, it is necessary to ----------- FOOTNOTE BEGINS --------- [foot #] 7 58 CFR 69246, 69254 (Dec. 30, 1993) (to be codified at 48 CFR 6101). ----------- FOOTNOTE ENDS ----------- determine what is the "basis for the ground of protest." Rule 5(b)(3). Litton's basis for protest is not that the FAA permitted CSTI an exception to the power requirements. Rather, Litton's basis for protest is that Litton was not excepted from the power requirements. Litton's challenge to the second reason given by the FAA for excluding Litton's alternate proposal from the competitive range is untimely because Litton knew on May 6, 1994, that the FAA had determined that Litton's multi-channel unit did not meet the specification's dimension requirements, and that the FAA had not excepted Litton from those requirements. Litton asserts that it was misled by the FAA into believing that there could be no exceptions to the power requirements, and that this should relieve Litton from the responsibility for filing a protest within ten days after receipt of the FAA's May 6, 1994 letter. Litton's assertion that it was misled by the FAA is based upon questions asked by Litton and responded to by the FAA. In response to the RFI, Litton asked why FAA-G-2100e was being imposed, and the FAA responded that it was soliciting vendors' suggestions for alternatives, their applications, and the basis for using the alternatives. In response to the draft solicitation, Litton asked that the FAA reconsider its need for one of the power requirements, because it would add cost. The FAA replied that the requirement, contained in FAA-G-2100e, would not be modified. Unlike the draft solicitation, the final solicitation contains a provision that makes it clear that vendors could take exception to mandatory requirements, so long as they provided a full discussion and explanation of the exception and explained the benefit to the FAA. Litton never provided the FAA with any request for an exception to any power requirement, any dimension requirement, or any other requirement of the solicitation. Litton says that it did not request an exception because, although it recognized that the solicitation provides that the FAA can allow exceptions to mandatory requirements, Litton did not believe that this provision applied to the mandatory power requirements, due to the FAA's misleading pre-solicitation statements. Litton was not misled by the FAA. Pre-solicitation, the FAA stated that the FAA-G-2100e requirements would not be modified, and that offerors had to meet all mandatory requirements. Also pre-solicitation, the FAA stated, in response to a question about FAA-G-2100e, that it was soliciting vendors' suggestions for alternatives. The solicitation itself states quite plainly that vendors could propose exceptions to mandatory requirements, so long as they justified the exceptions, and that the FAA would evaluate the exceptions. What Litton says it learned, after award, is that the FAA was willing to allow exceptions to the specification's power requirements. Litton should have known of the FAA's willingness to permit exceptions since the date the solicitation was issued. The power requirements were mandatory requirements, which the FAA had not modified. If Litton wished to propose an exception to those requirements, the solicitation provided that the FAA would evaluate the exception. If Litton did not understand the plain language of the solicitation, it is not the result of being misled by the FAA. By reading the solicitation to say that the FAA would not allow exceptions to the power requirements, Litton added language to the solicitation which is not there. In support of its position that its challenge to the second reason given by the FAA for excluding its alternate proposal from the competitive range is timely, Litton cites to ST Systems Corp., GSBCA 11207-P, 91-3 BCA 24,201, 1991 BPD 154.[foot #] 8 There, we permitted ST Systems to protest its exclusion from the competitive range, after contract award. Our decision was based upon the fact that, in order to know whether it had a basis for challenging the agency's decision, ST Systems needed to know the other vendors' probable costs. Such information could not have been made available to ST Systems at the time the competitive range determination was made, and so the Board did not require ST Systems to protest within ten days after being excluded from the competitive range. In our case, on May 6, 1994, Litton knew that its alternate proposal had been excluded from the competitive range for failing to meet a mandatory requirement, that the solicitation provided that vendors could propose exceptions to mandatory requirements, that it had not proposed an exception to any mandatory requirement, and that the FAA had not permitted Litton an exception to any mandatory requirement. Litton did not need to learn anything about any other vendor's proposal in order to challenge the FAA's decision. This Board has consistently required offerors to challenge competitive range decisions within ten days after they learn of those decisions, especially when those decisions are based upon the technical acceptability of proposals. Integrated Systems Group, Inc. v. Department of Labor, GSBCA 12581-P, et al., 94-1 BCA 26,439, 1993 BPD 284. We have explained that a post- award protest of a competitive range decision could severely prejudice the procurement process by potentially requiring agencies to redo months of negotiations and evaluations should such a protester prevail. Such a system would prejudice other vendors as well and unnecessarily prolong -- at additional expense -- the procurement process in contravention of the Brooks Act's stated support for the "goals of economic and efficient procurement." 40 U.S.C. 759(f)(5)(A) (1988). Trimble Navigation Ltd. v. Department of Transportation, GSBCA 11692-P, 93-1 BCA 25,341, at 126,242, 1992 BPD 62, at 6-7. ----------- FOOTNOTE BEGINS --------- [foot #] 8 Litton does not present any argument as to why its challenge to the first reason for excluding its proposal from the competitive range is timely. ----------- FOOTNOTE ENDS ----------- Litton's challenge to the FAA's competitive range determination is untimely. Litton's Alternate Proposal's Chance For Award Even if Litton had filed a timely protest of its exclusion from the competitive range, it has not convinced us that its protest would succeed. Litton's assertion that its alternate proposal was the -highest technically by a small margin and in price by a significant margin is incorrect. When the FAA evaluated Litton's alternate proposal, the FAA did not realize that the proposal included both multi-channel units and single-channel units. The FAA's confusion is understandable because Litton's technical proposal does not explain that Litton had formulated a blend of two types of units, and Litton provided only a multi-channel unit to the FAA for OCT of its alternate proposal. Computing Litton's price based upon supplying only single-channel or only multi-channel units, the FAA established that the true price range of Litton's alternate proposal was than Litton's basic proposal and than CSTI's proposal. Also, the range of technical scores for the alternate proposal would have been lower if the FAA had realized that Litton's alternate proposal had contained Litton's single-channel unit, which received a technical score than Litton's multi- channel unit. Considering that Litton's alternate proposal was actually -priced and of technical quality than the FAA realized when it excluded the proposal from the competitive range, Litton has not established that its alternate proposal had a reasonable chance of being selected for award. In addition, Litton has not established that its alternate proposal could have had a reasonable chance for award, given that it did not meet at least solicitation requirements. First, the solicitation clearly required offerors to supply at OCT all of the equipment that they intended to provide after contract award. The OCT was important because it was the means that the FAA was using to evaluate most of the features of the offerors' systems. The solicitation explains that offerors would not be permitted to supply equipment as part of their BAFOs. Litton did not comply with this provision because it did not supply the FAA with a single-channel unit as part of its alternate proposal's OCT. Second, the specification required that equipment meet certain dimension requirements. Litton's multi-channel units exceeded the specified dimensions. The solicitation required that offerors who took exception to solicitation requirements were required to identify those exceptions explicitly and to justify those exceptions, so that the FAA could evaluate the exceptions. In its proposal, Litton did not take exception to the requirement that it supply all of its proposed equipment at OCT, to the dimension requirement, or to any other requirement. Litton did not discuss or explain its failure to provide a single-channel unit for OCT of its alternate proposal or its failure to meet the dimension requirements, and did not explain why it would be beneficial to the FAA to permit exceptions to these requirements. Litton does not explain how its proposal would have had a reasonable chance for award, given that it did not meet all of the solicitation requirements and did not identify or justify any exceptions to those requirements. Litton has not established that, considering price and the other evaluation factors that were stated in the solicitation, its alternate proposal had a reasonable chance of being selected for award. For this reason, even if Litton's first ground of protest were timely, it would not succeed. II. Litton's Basic Proposal Litton's second ground of protest is that the FAA allowed CSTI to take exception to the specification's power requirements without offering Litton an opportunity to amend its basic proposal, and that this violated FAR 15.606(c). This FAR provision reads: 15.606 Changes in Government requirements. (c) If the proposal considered to be most advantageous to the Government (as determined according to the established evaluation criteria) involves a departure from the stated requirements, the contracting officer shall provide all offerors an opportunity to submit new or amended proposals on the basis of the revised requirements; provided that this can be done without revealing to the other offerors the solution proposed in the original departure or any other information that is entitled to protection (see 15.407(c)(8) and 15.610(d)). FAR 15.407(c)(8) directs the reader to FAR 52.215-12, which provides that offerors who include in their proposals data that they do not want disclosed shall place a restrictive legend on the title page of their proposals, and mark each page which contains restricted data. According to Litton, the solicitation permitted offerors to take exception to mandatory requirements. But, Litton argues, the FAA should have told Litton the particular mandatory power requirements to which CSTI took exception in its initial proposal and in its BAFO, and amended the solicitation accordingly, either before or after receipt of BAFOs. Litton asserts that the FAA could have taken these actions without disclosing to Litton any of CSTI's proposed solution. Litton claims that it was prejudiced by the FAA's actions because, if the FAA had disclosed the exceptions requested by CSTI, Litton could have changed its basic proposal to lower its price or to provide more desired features. Although the parties cited us to numerous decisions addressing FAR 15.606(c), the precedents are not persuasive because, in the cases cited, the solicitations provided that offerors were required to meet all mandatory requirements. Here, the solicitation provides that exceptions to mandatory requirements would not make a proposal unacceptable, so long as the offeror provided a full discussion, a detailed explanation, and a justification so that the FAA could evaluate the exception. The purpose of FAR 15.606(c) is to ensure that all offerors are provided an equal opportunity to compete for award, and we will interpret the regulation consistently with its purpose, in light of the language in the solicitation issued in this procurement. It is difficult to conclude that CSTI's proposal departed from the stated requirements of the solicitation, because CSTI followed precisely the directions contained in the solicitation. CSTI identified the requirements to which it took exception, and provided the discussion, explanation, and justification required. CSTI's proposal did not identify a large number of exceptions, and did not take exception to any significant requirement that changed the essential character or conditions of the solicitation. It is also difficult to conclude that the FAA revised its requirements, given that it never eliminated the power requirements or decided that a change in power requirements was warranted for all offerors, without regard to the particular systems that they proposed. Even if the FAA changed its requirements, it would not have been required to tell Litton the precise specification sections to which CSTI took exception. FAR 15.606(c) provides that the FAA could not have revealed to Litton any information that was entitled to protection. Although CSTI did not mark the individual pages which discuss its exceptions as containing restricted data, CSTI did place such a mark on the cover page which it placed in front of those individual pages. The pages which list and discuss CSTI's exceptions were entitled to protection, and FAR 15.606(c) did not permit the FAA to disclose the information contained on those pages. FAR 15.606(c) also provides that, even if the FAA changed its requirements, it could not have revealed to Litton the solution proposed by CSTI. Litton argues that the FAA would not have revealed CSTI's solution if it had only told Litton the particular mandatory requirements to which CSTI took exception. We disagree. The exceptions that CSTI proposed were integral parts of its solution. The notion that disclosing the exceptions would not have revealed CSTI's solution is belied by Litton's assertion that, in order to prepare a better basic proposal, it needed to be told the particular specification sections to which CSTI -- Litton's sole remaining competitor -- took exception. Litton also argues that disclosing CSTI's exceptions would have been permitted by FAR 15.606(c) because the FAA would not have released any innovative or ingenious technical solution of CSTI. FAR 15.606(c) does not prohibit agencies from disclosing only innovative and ingenious technical solutions. Rather, the regulation prohibits disclosure of one offeror's solution to another offeror, without regard to the simplicity or the inventiveness of the solution. Disclosing to Litton the precise exceptions taken by CSTI would have provided Litton with much more than an equal opportunity for award, and this is not consistent with the language or the purpose of FAR 15.606(c). Finally, Litton has not persuaded us that it was prejudiced by the fact that it did not know what exceptions CSTI took to the mandatory requirements. Litton asserts that it could have used a replacement power supply and either provided more desired features or lowered the price of its basic proposal. Litton does not know whether, in fact, it could have found a replacement power supply for the DSRCE that it offered in its basic proposal. After Litton learned what exceptions CSTI took to the mandatory requirements, Litton located what it claims are two possible replacement power supplies. Litton, however, has never tested either of these power supplies in its DSRCE, and there is no evidence concerning how or whether these power supplies will function in Litton's equipment. In addition, there is no evidence that Litton would have been able to provide more desired features in its basic proposal, if it had learned of CSTI's exceptions. Finally, Litton's ability to lower its price by approximately $ would not have made Litton's proposal the best value to the FAA, as discussed in the following section. III. Best Value Determination Litton's third basis for protest is that the FAA did not make a proper best value determination. Litton argues that the agency improperly evaluated the risks associated with Litton's and CSTI's proposals, did not adequately document or justify its best value determination before award, and cannot presently justify its best value determination. Litton asserts that the FAA's actions violated 41 U.S.C. 253b(a) and 253b(d)(4) (1988), and FAR 15.603, 15.608(a), and 15.611(d). The statutes and regulations cited by Litton require agencies to evaluate proposals based solely upon the factors set forth in the solicitation; to award a contract to the offeror whose proposal is most advantageous to the agency, considering price and other factors stated in the solicitation; to evaluate proposals comprehensively and select an awardee who will best meet the agency's stated requirements; and to perform cost and technical evaluations and document the results of those evaluations. Because the FAA has not violated any of these statutory or regulatory provisions, and because the FAA conducted its evaluation consistently with the terms of the solicitation, Litton's arguments are rejected. Risk Assessments Litton's Basic Proposal Litton asserts that it complied with the requirements of all of the mandatory requirements, and so its basic proposal should have been assessed as presenting a low risk. Regarding Shalls 11 and 18, Litton says that a plain reading of those requirements establishes that they were met by Litton's basic proposal. Litton says that its failure to meet the requirement of Shall 103 was the result of an "incongruity" between its equipment and the FAA's test facility, and the result of an unspecified minimum event time. Regarding Shall 105, Litton says that the FAA misunderstood Litton's BAFO, and Litton also claims that its inability to pass the OCT was due to a simple wiring error that has been corrected. Finally, Litton argues, even if the FAA properly determined that Litton did not meet Shalls 11, 18, and 105, Litton has a simple solution for meeting these requirements. The result, Litton says, is that its basic proposal should have been rated as a low risk instead of a risk. With regard to Shalls 11 and 18, we are not convinced that a plain reading of these requirements establishes that they were met by Litton. The analysis and interpretation of these technical requirements that is offered by the FAA and by its expert is much more convincing than the analysis offered by Litton. It is not appropriate to ignore Shall 5, as does Litton. Shall 5 is a definition, and it provides that a DSRCE channel shall allow control over both frequencies simultaneously. Shall 5 explains that "frequency" refers to a voice path and its associated control signals. Shall 11 requires that each channel allow controllers to transmit without arbitration. Shall 18 requires that the control facility DSRCE shall accept radio control signals, encode those signals, and transmit them to the remote site DSRCE. Litton's system does not permit complete control over both frequencies at the same time, it arbitrates among signals, and it does not accept signals over both frequencies simultaneously. Based upon the air traffic controller's testimony as to how controllers perform their work, we agree with the FAA's expert that the requirement for simultaneous control is a fundamental requirement for air traffic control, and this requirement is, as the expert stated, found "right up front" in the solicitation, in Shall 5. Litton states that it has a simple solution for complying with Shalls 11 and 18. If this is true, there is no evidence of it in the record. Litton's expert testified that he had no personal knowledge of the engineering effort that would be required for Litton to comply with Shall 11. He explained that he was told by a Litton employee that Litton would have to code and test software and insert that software into its DSRCE. The FAA's expert, who exhibited a thorough familiarity with the solicitation's requirements and the offerors' proposals, is of the opinion that major design changes to Litton's basic architecture would be required in order to meet these mandatory requirements. Litton's basic proposal does not meet the requirements of Shalls 11 and 18. Litton has not established that it has any solution, simple or complex, for meeting these requirements. For these reasons, the FAA correctly concluded that Litton's basic proposal presented a risk that the DSRCE program's technical and schedule requirements could not be met. With regard to Shall 103, Litton asserts, as it has since it responded to the FAA's discussion questions, that there is a problem with the FAA's test procedure. Litton's best evidence of this is that Litton was not able to pass the test for Shall 103 at OCT, although Litton could pass the test in its own laboratory, using its own test procedures. The FAA's lead technical evaluator contacted the FAA's test facility and confirmed that there was nothing wrong with the FAA's test procedure. The record does not support a conclusion that Litton's basic proposal failed the OCT for Shall 103 due to some deficiency in the FAA's test procedure. The FAA appropriately determined that Litton's failure to meet the requirements of Shall 103 presented a risk. Concerning Shall 105, Litton contends that the FAA misunderstood Litton's BAFO to state that, although Litton had a solution to Shall 105, it would not implement that solution unless directed to do so by the FAA. In response to the FAA's discussion questions, Litton stated that it could comply with Shall 105 if it were to make a wiring change. Litton did not state that this change had been made or tested. Litton had been permitted to make a wiring change during OCT, and this had solved only part of Litton's problem with Shall 105. Litton stated in its BAFO that it met the requirements of Shall 105. We are not so sure that the FAA's interpretation of Litton's BAFO was as far-fetched as Litton contends. It is not unreasonable to read Litton's response to the discussion questions and Litton's BAFO as meaning that Litton had made one change, which had not solved Litton's problem, and that Litton had not yet made or tested other changes which it believed would solve its problem with Shall 105. Litton, however, contends that it was able to make a simple change and meet the requirements of Shall 105. Even assuming that the FAA should have realized that the change Litton needed to make was a simple one, if Litton is rated as a low risk for Shall 105, Litton has not persuaded us that this would result in a change to its overall risk rating, because of the high risk associated with Shalls 11 and 18, and the risk associated with Shall 103. CSTI's Basic Proposal Litton asserts that CSTI's proposal should have been rated as a risk instead of a risk. Litton says that the FAA should have considered CSTI's inability to meet the requirements of Shalls 104 and 141. Litton also says that the FAA did not attribute enough risk to permitting CSTI to meet the requirements of FAA-G-2100f instead of FAA-G-2100e, and to change its power supply after the OCT. Concerning Shall 104, Litton did not establish that CSTI's failure to comply with part of this mandatory requirement created a risk to the FAA. Concerning Shall 141, although CSTI took exception to this requirement as a precaution, the test results that CSTI provided to the FAA in response to discussion questions establish that CSTI met this requirement. Litton did not establish that CSTI should be rated anything other than risk for the exceptions that it took to Shalls 136, 137, 138, and 141, which contain power requirements. For these Shalls, CSTI explained that it would meet the standards of FAA-G-2100f instead of the standards of FAA-G-2100e. FAA-G-2100f is a revision of FAA-G-2100e. The revised standards incorporated in FAA-G-2100f were developed as the result of a study by industry and the FAA, and adopted by the FAA. The standards established by FAA-G-2100e applied to all equipment, regardless of the amount of power that it consumed. FAA-G-2100f simply recognizes that equipment which consumes a small amount of power does not need to be held to the same standards as equipment which consumes a large amount of power. CSTI's DSRCE consumes a small amount of power, and it was appropriate for the FAA to assess CSTI's exceptions to Shalls 136, 137, 138, and 141 as posing a risk. Similarly, Litton did not establish that CSTI's replacement of its power supply poses anything more than a risk to deployment of its DSRCE. CSTI located a power supply manufacturer who could scale down one of its power supplies and provide CSTI with a replacement power supply. Scaling down a power supply is not a very complicated task. CSTI appropriately loaded the power supply provided by the manufacturer, tested it, and provided the test results to the FAA as part of its responses to the FAA's discussion questions. CSTI received a price quote from the power supply manufacturer, who has assigned a part number to the scaled down power supply and intends to market it commercially. The manufacturer's delivery schedule will not interfere with CSTI's ability to meet the contract's delivery schedule. For these reasons, the FAA correctly assessed the risk associated with CSTI's replacement of its power supply. Best Value Analysis CSTI's equipment is clearly technically superior to Litton's equipment in many ways, and Litton does not argue otherwise. Litton argues that CSTI's technical superiority is not worth the approximately $ million difference in price between CSTI's proposal and Litton's proposal. After balancing the evidence in the record, we disagree.[foot #] 9 Considering the cost savings associated with only the RSS and CMT capabilities offered by CSTI, CSTI's proposal is the best value to the Government. The FAA estimated that CSTI's RSS capability is worth between $6 million and $12 million over a ten-year period. Both estimates are based upon a projection made by the FAA of the cost of acquiring the equipment needed to provide RSS capability in a separate procurement. The lower estimate is that of the FAA's lead technical evaluator, and the higher estimate is that of the FAA's Associate Administrator for National Airspace Development, who considered that the FAA will acquire approximately twice as much DSRCE as is called for in the FAA's projected cost of separately acquiring RSS capability. In the opinion of both of these witnesses, because of the RSS capability provided by CSTI's DSRCE, the FAA will no longer need to conduct a separate procurement for RSS capability, and their estimates are based upon the amount that the FAA will save as the result of not acquiring RSS capability separately. Litton says that the FAA's estimates of the cost savings associated with RSS capability are inflated and unrealistic because CSTI's system cannot provide one of the functions that was included in a draft RSS specification. CSTI's expert witness, who designed CSTI's system, testified that CSTI could provide this additional function by modifying its software. Litton introduced no evidence to suggest that FAA needs the function set forth in the draft RSS specification, or that the software modification will be costly. Litton has not persuaded us, by a preponderance of the evidence, that the FAA's analysis of the savings associated with CSTI's RSS capability is in error. The FAA estimated that CSTI's CMT capability is worth a minimum of $1.3 million and a maximum of $4 million. The lower estimate, which includes no amount for the cost savings associated with reduced travel, represents the amount that the ----------- FOOTNOTE BEGINS --------- [foot #] 9 We summarily reject Litton's argument that we should grant this protest because the FAA did not adequately document its best value decision prior to the protest. Even if we agreed with Litton, the fact remains that we do not review contracting officer decisions based upon the record that was before the agency. We hold hearings and find facts de novo. __ ____ Grumman Data Systems Corp. v. Widnall, 15 F.3d 1044 (Fed. Cir. _______________________________________ 1994). ----------- FOOTNOTE ENDS ----------- FAA will save by being able to align communications circuits using only one technician at one type of control site. The higher estimate represents a conservative estimate of the amount that the FAA will save by being able to align communications circuits remotely at all three types of control sites, including a small amount for the cost savings associated with reduced travel. Litton says that these estimates are unreliable because the FAA estimated its savings for ten years. The contract provides that the DSRCE is warranted for ten years, so it was appropriate for the FAA to estimate its savings for ten years. Also, Litton says that there are some maintenance functions that CSTI's CMT capability cannot provide unless its software is modified. This is true. However, the FAA's estimate does not assume that CSTI's CMT capability will permit the FAA to conduct all of its maintenance using CSTI's CMT capability. Litton has not established that the FAA incorrectly estimated the cost savings associated with CSTI's CMT capability. The FAA estimates that the RSS and CMT capabilities are worth between $7.3 million and $16 million. The FAA has not estimated the value of the various other additional features, some of which are listed in our findings of fact, that are provided by CSTI's system and not by Litton's system. The FAA has not attempted to place a value upon the lower risk associated with CSTI's system. Considering the reliability of the FAA's estimates of the value of the RSS and CMT capabilities, and the fact that the other features and lower risk of CSTI's system are not worthless, Litton has failed to establish that CSTI's system is not worth the added $7 million price. Litton argues that the FAA has not demonstrated with "reasonable certainty" that the added value of CSTI's DSRCE is worth the higher price. Litton finds support for its argument in Lockheed Missiles & Space Co. v. Bentsen, 4 F.3d 955 (Fed. Cir. 1993). In that case, the court of appeals stated, "Accordingly, a proposal which is one point better than another but costs millions of dollars more may be selected if the agency can demonstrate within a reasonable certainty that the added value of the proposal is worth the higher price." 4 F.3d at 960 (emphasis in original). Apparently, Litton interprets this language as shifting the burden of proof from the protester to the agency. This is not the case. The court of appeals was illustrating a point, and did not hold that the agency bears the burden of proof. In order to prevail in a protest case before this Board, the burden remains upon the protester to establish, by a preponderance of the evidence, a violation of statute or regulation by an agency. Because Litton has not established any such violation by a preponderance of the evidence, it does not prevail in this protest. Decision Because the protester did not prove that the agency violated any statute or regulation in the course of conducting this procurement, the protest is DENIED. The suspension of respondent's procurement authority lapses by its own terms. _______________________________ MARTHA H. DeGRAFF Board Judge We concur: ___________________________ _______________________________ DONALD W. DEVINE MARY ELLEN COSTER WILLIAMS Board Judge Board Judge