TITLE: B-311344; B-311344.3; B-311344.4; B-311344.6; B-311344.7; B-311344.8; B-311344.10; B-311344.11, The Boeing Company, June 18, 2008
BNUMBER: B-311344; B-311344.3; B-311344.4; B-311344.6; B-311344.7; B-311344.8; B-311344.10; B-311344.11
DATE: June 18, 2008
*********************************************************************************************************************************
B-311344; B-311344.3; B-311344.4; B-311344.6; B-311344.7; B-311344.8; B-311344.10; B-311344.11, The Boeing Company, June 18, 2008

   DOCUMENT FOR PUBLIC RELEASE
   The decision issued on the date below was subject to a GAO Protective
   Order. This redacted version has been approved for public release.

   Decision

   Matter of: The Boeing Company

   File: B-311344; B-311344.3; B-311344.4; B-311344.6; B-311344.7;
   B-311344.8; B-311344.10; B-311344.11

   Date: June 18, 2008

   Rand L. Allen, Esq., Paul F. Khoury, Esq., Scott M. McCaleb, Esq., Martin
   P. Willard, Esq., Nicole J. Owren-Wiest, Esq., Kara M. Sacilotto, Esq.,
   Nicole P. Wishart, Esq., Jon W. Burd, Esq., Stephen J. Obermeier, Esq.,
   and Heidi L. Bourgeois, Esq., Wiley Rein LLP; and Charles J. Cooper, Esq.,
   Michael W. Kirk, Esq., and Howard C. Neilson, Esq., Cooper & Kirk; Lynda
   Guild Simpson, Esq., and Stephen J. Curran, Esq., The Boeing Company, for
   the protester.

   Neil H. O'Donnell, Esq., Allan J. Joseph, Esq., David F. Innis, Esq.,
   Thomas D. Blanford, Esq., Aaron P. Silberman, Esq., Tyson Arbuthnot, Esq.,
   Michelle L. Baker, Esq., James Robert Maxwell, Esq., and Suzanne M.
   Mellard, Esq., Rogers Joseph O'Donnell; and Joseph O. Costello, Esq.,
   Northrop Grumman Systems Corporation, for the intervenor.

   Bryan R. O'Boyle, Esq., Col. Neil S. Whiteman, James A. Hughes, Esq., Col.
   Timothy Cothrel, Robert Balcerek, Esq., Maj. Christopher L. McMahon, W.
   Michael Rose, Esq., Stewart L. Noel, Esq., Gerald L. Trepkowski, Esq.,
   Lynda Troutman O'Sullivan, Esq., John J. Thrasher III, Esq., Lt. Col.
   Thomas F. Doyon, Anthony P. Dattilo, Esq., Bridget E. Lyons, Esq., John R.
   Hart, Esq., Ronald G. Schumann, Esq., Maj. Steven M. Sollinger, Maj.
   Sandra M. DeBalzo, and John M. Taffany, Esq., Department of the Air Force,
   for the agency.

   Guy R. Pietrovito, Esq., and James A. Spangenberg, Esq., Office of the
   General Counsel, GAO, participated in the preparation of the decision.

   DIGEST

   1.      Protest is sustained, where the agency, in making the award
   decision, did not assess the relative merits of the proposals in
   accordance with the evaluation criteria identified in the solicitation,
   which provided for a relative order of importance for the various
   technical requirements, and where the agency did not take into account the
   fact that one of the proposals offered to satisfy more "trade space"
   technical requirements than the other proposal, even though the
   solicitation expressly requested offerors to satisfy as many of these
   technical requirements as possible.

   2.      Protest is sustained, where the agency violated the solicitation's
   evaluation provision that "no consideration will be provided for exceeding
   [key performance parameter] KPP objectives" when it recognized as a key
   discriminator the fact that the awardee proposed to exceed a KPP objective
   relating to aerial refueling to a greater degree than the protester.

   3.      Protest is sustained, where the record does not demonstrate the
   reasonableness of the agency's determination that the awardee's proposed
   aerial refueling tanker could refuel all current Air Force fixed-wing
   tanker-compatible receiver aircraft in accordance with current Air Force
   procedures, as required by the solicitation.

   4.      Protest is sustained, where the agency conducted misleading and
   unequal discussions with the protester, where the agency informed the
   protester that it had fully satisfied a KPP objective relating to
   operational utility, but later determined that the protester only
   partially met this objective, without advising the offeror of this change
   in its assessment and while continuing to conduct discussions with the
   awardee relating to its satisfaction of the same KPP objective.

   5.      Protest is sustained, where the agency unreasonably determined
   that the awardee's refusal to agree to the specific solicitation
   requirement that it plan and support the agency to achieve initial organic
   depot-level maintenance within 2 years after delivery of the first
   full-rate production aircraft was an "administrative oversight," and
   improperly made award, despite this clear exception to a material
   solicitation requirement.

   6.      Protest is sustained, where the agency's evaluation of military
   construction costs in calculating the offerors' most probable life cycle
   costs for their proposed aircraft was unreasonable, where the evaluation
   did not account for the offerors' specific proposals, and where the
   calculation of military construction costs based on a notional
   (hypothetical) plan was not reasonably supported.

   7.      Protest is sustained, where the agency improperly added costs to
   an element of cost (non-recurring engineering costs) in calculating the
   protester's most probable life cycle costs to account for risk associated
   with the protester's failure to satisfactorily explain the basis for how
   it priced this cost element, where the agency did not determine that the
   protester's proposed costs for that element were unrealistically low.

   8.      Protest is sustained, where the agency's use of a "Monte Carlo"
   simulation model to determine the protester's probable cost of
   non-recurring engineering associated with the system demonstration and
   development portion of the acquisition was unreasonable, where the model's
   inputs concerned total weapons systems at an overall program level and
   there is no indication that this is a reliable predictor of anticipated
   growth of the protester's non-recurring engineering costs.

   9.      Protester is not required to file a "defensive protest" when
   during the procurement it is apprised of an agency's evaluation judgments
   with which it disagrees or where it believes the evaluation is
   inconsistent with the solicitation's evaluation scheme, because GAO's Bid
   Protest Regulations, 4 C.F.R. sect. 21.2(a)(2) (2008), require that where
   the protest involves a procurement conducted on the basis of competitive
   proposals under which a debriefing is requested and, when requested, is
   required, these protest grounds can only be raised after the offered
   debriefing.

   10.  While an agency, in an appropriate case, may request under GAO's Bid
   Protest Regulations, 4 C.F.R. sect. 21.3(d) (2008), that a protester
   provide specific relevant documents, of which the agency is aware and does
   not itself possess, this does not allow for "wide-open" document requests
   by an agency of broad categories of documents.

   DECISION

   The Boeing Company protests the award of a contract to Northrop Grumman
   Systems Corporation under request for proposals (RFP) No.
   FA8625-07-R-6470, issued by the Department of the Air Force, for aerial
   refueling tankers.[1] Boeing challenges the Air Force's technical and cost
   evaluations, conduct of discussions, and source selection decision.[2]

   As explained below, we find that the agency's selection of Northrop
   Grumman's proposal as reflecting the best value to the government was
   undermined by a number of prejudicial errors that call into question the
   Air Force's decision that Northrop Grumman's proposal was technically
   acceptable and its judgment concerning the comparative technical
   advantages accorded Northrop Grumman's proposal. In addition, we find a
   number of errors in the agency's cost evaluation that result in Boeing
   displacing Northrop Grumman as the offeror with the lowest evaluated most
   probable life cycle costs to the government. Although we sustain Boeing's
   protest on grounds related to these errors, we also deny many of Boeing's
   challenges to the award.

   Specifically, we sustain the protest, because we find that (1) the Air
   Force did not evaluate the offerors' technical proposals under the key
   system requirements subfactor of the mission capability factor in
   accordance with the weighting established in the RFP's evaluation
   criteria; (2) a key technical discriminator relied upon in the selection
   decision in favor of Northrop Grumman relating to the aerial refueling
   area of the key system requirements subfactor, was contrary to the RFP;
   (3) the Air Force did not reasonably evaluate the capability of Northrop
   Grumman's proposed aircraft to refuel all current Air Force fixed-wing,
   tanker-compatible aircraft using current Air Force procedures, as required
   by the RFP; (4) the Air Force conducted misleading and unequal discussions
   with Boeing with respect to whether it had satisfied an RFP objective
   under the operational utility area of the key system requirements
   subfactor; (5) Northrop Grumman's proposal took exception to a material
   solicitation requirement related to the product support subfactor; (6) the
   Air Force did not reasonably evaluate military construction (MILCON) costs
   associated with the offerors' proposed aircraft consistent with the RFP;
   and (7) the Air Force unreasonably evaluated Boeing's estimated
   non-recurring engineering costs associated with its proposed system
   development and demonstration (SDD).

   BACKGROUND

   Aerial refueling is a key element supporting the effectiveness of the
   Department of Defense's (DoD) air power in military operations and is, as
   such, an important component of national security. See AR, Tab 333,
   Capability Development Document, Dec. 27, 2006, at 2, 7; see also Air
   Force Refueling: The KC-X Aircraft Acquisition Program, Congressional
   Research Service (CRS) Report for Congress, No. RL34398, Feb. 28, 2008, at
   1. Currently, the Air Force uses two types of aircraft for aerial
   refueling: the KC-135, which is considered to be a medium-sized airplane,
   and the larger KC-10. The Air Force's fleet of KC-135 aircraft currently
   has an average age of 46 years and is the oldest combat weapon system in
   the agency's inventory;[3] for the newer KC-10 aircraft, the average age
   is over 20 years. Defense Acquisitions: Air Force Decision to Include a
   Passenger and Cargo Capability in Its Replacement Refueling Aircraft Was
   Made without Required Analyses, GAO-07-367R, Mar. 6, 2007, at 1.

   To begin replacing the aging refueling tanker fleet, the Air Force
   established a three-pronged approach under which it intended to first
   conduct a procurement to replace the older KC-135 tankers, while
   maintaining the remaining KC-135 and KC-10 tankers; the first procurement,
   which is the acquisition protested here, was identified by the Air Force
   as the KC-X procurement or program. See AR, Tab 4, Acquisition Strategy
   Plan Briefing, at 9-10. The Air Force intends to replace the remaining
   KC-135 and KC-10 aircraft in later procurements under programs the agency
   identified as the KC-Y and KC-Z.

   Solicitation

   The RFP, issued January 30, 2007, provided for the award of a contract
   with cost reimbursement and fixed-price contract line items. In this
   regard, offerors were informed that, although the agency would procure up
   to 179 KC-X aircraft over a 15 to 20-year period, the initial contract
   would be for the SDD of the KC-X aircraft and the procurement of up to 80
   aircraft, beginning with the delivery of four SDD aircraft and two low
   rate initial production (LRIP) aircraft.[4] Offerors were also informed
   that the agency contemplated receiving an existing commercial, Federal
   Aviation Administration (FAA) or equivalent certified transport aircraft
   modified to meet the agency's requirements. RFP, Statement of Objectives
   (SOO) for KC-X SDD, at 1.

   A detailed system requirements document (SRD) was provided in the RFP that
   presented the technical performance requirements for the KC-X aircraft. In
   this regard, the SRD stated that

     [t]he primary mission of the KC-X is to provide world-wide, day/night,
     adverse weather aerial refueling . . . on the same sortie to receiver
     capable United States (U.S.), allied, and coalition military aircraft
     (including unoccupied aircraft). [The KC-X aircraft will] provide
     robust, sustained [aerial refueling] capability to support strategic
     operations, global attack, air-bridge, deployment, sustainment,
     employment, redeployment, homeland defense, theater operations, and
     special operations. Secondary missions for KC-X include emergency aerial
     refueling, airlift, communications gateway, aeromedical evacuation (AE),
     forward area refueling point (FARP), combat search and rescue, and
     treaty compliance.

   RFP, SRD sect. 1.2.1. The SRD identified the minimum and desired
   performance/capability requirements for the aircraft. The minimum
   performance capabilities of the aircraft were identified in nine key
   performance parameters (KPP), which the Air Force summarized as follows:

   +------------------------------------------------------------------------+
   |KPP|           Parameter           |        Required Performance        |
   |---+-------------------------------+------------------------------------|
   | 1 |Tanker Air Refueling Capability|Air refueling of all current and    |
   |   |                               |programmed fixed wing receiver      |
   |   |                               |aircraft                            |
   |---+-------------------------------+------------------------------------|
   | 2 |Fuel Offload and Range         |Fuel, offload, range chart          |
   |   |                               |equivalent to KC-135                |
   |---+-------------------------------+------------------------------------|
   | 3 |Communications, Navigation,    |Worldwide flight operations at all  |
   |   |Surveillance/Air Traffic       |times in all civil and military     |
   |   |Management                     |airspace                            |
   |---+-------------------------------+------------------------------------|
   | 4 |Airlift Capability             |Carry passengers, palletized cargo, |
   |   |                               |and/or aeromedical patients on      |
   |   |                               |entire main deck                    |
   |---+-------------------------------+------------------------------------|
   | 5 |Receiver Air Refueling         |Refueled in flight from any boom    |
   |   |Capability                     |equipped tanker aircraft            |
   |---+-------------------------------+------------------------------------|
   | 6 |Force Protection               |Operate in chemical/biological      |
   |   |                               |environments                        |
   |---+-------------------------------+------------------------------------|
   | 7 |Net-Ready                      |Meet enterprise-level joint critical|
   |   |                               |integrated architecture requirements|
   |---+-------------------------------+------------------------------------|
   | 8 |Survivability                  |Operate in hostile environments     |
   |   |                               |(night vision and imaging systems,  |
   |   |                               |electromagnetic pulse, defensive    |
   |   |                               |systems: infrared  detect and       |
   |   |                               |counter, radio frequency detect, no |
   |   |                               |counter)                            |
   |---+-------------------------------+------------------------------------|
   | 9 |Multi-point Refueling          |Multi-point drogue[5] refueling     |
   +------------------------------------------------------------------------+

   AR, Tab 46, Source Selection Evaluation Team (SSET) Final Briefing to
   Source Selection Advisory Council (SSAC) and Source Selection Authority
   (SSA), at 18.

   The RFP provided for award on a "best value" basis and stated the
   following evaluation factors and subfactors:

   +------------------------------------------------------------------------+
   |Mission Capability                                                      |
   |------------------------------------------------------------------------|
   |                    |Key System Requirements                            |
   |                    |---------------------------------------------------|
   |                    |System Integration and Software                    |
   |                    |---------------------------------------------------|
   |                    |Product Support                                    |
   |                    |---------------------------------------------------|
   |                    |Program Management                                 |
   |                    |---------------------------------------------------|
   |                    |Technology Maturity and Demonstration              |
   |------------------------------------------------------------------------|
   |Proposal Risk                                                           |
   |------------------------------------------------------------------------|
   |Past Performance                                                        |
   |------------------------------------------------------------------------|
   |Cost/Price                                                              |
   |------------------------------------------------------------------------|
   |Integrated Fleet Aerial Refueling Assessment (IFARA)                    |
   +------------------------------------------------------------------------+

   Offerors were informed that the mission capability, proposal risk, and
   past performance factors were of equal importance and individually more
   important than the cost/price or IFARA factors, and that the cost/price
   and IFARA factors were of equal importance. The subfactors within the
   mission capability factor were stated to be of descending order of
   importance. RFP sect. M.2.1.

   The RFP stated that the agency, in its evaluation of proposals under the
   mission capability subfactors, would assign one of the color ratings
   identified in the solicitation,[6] and one of the proposal risk ratings
   that were also identified.[7] RFP sections M.2.2, M.2.3. In this regard,
   offerors were informed that proposal risk would only be assessed at the
   mission capability subfactor level and for only the first four subfactors.
   RFP sect. M.2.3.

   With respect to the key system requirements subfactor, the most important
   mission capability subfactor, offerors were informed that the agency would
   assess the offerors' understanding of, and substantiation of their ability
   to meet, the requirements of the SRD (with the exception of the logistics
   requirements that were to be evaluated under the product support
   subfactor). The RFP provided that the offerors' approaches to meeting the
   SRD requirements would be evaluated under the key system requirements
   subfactor in the following five areas: aerial refueling, airlift,
   operational utility, survivability, and "other system requirements." RFP
   sect. M.2.2.1.2.

   In order for a proposal to be found acceptable under this subfactor (and
   overall), an offeror was required to meet the various identified minimum,
   mandatory KPP "thresholds" identified in the SRD for each of the nine
   KPPs. The SRD also identified KPP "objectives" relating to some, but not
   all of, the identified KPP thresholds. In this regard, the RFP stated that

     [a]ll KPP thresholds [relating to the aerial refueling, airlift,
     operational utility, and survivability areas] must be met. Depending on
     substantiating rationale, positive consideration will be provided for
     performance above the stated KPP thresholds up to the KPP objective
     level. No consideration will be provided for exceeding KPP objectives.
     If there is no stated objective and, depending on substantiating
     rationale, positive consideration will be provided when the specified
     capability above the KPP threshold is viewed as advantageous to the
     Government.

   RFP sect. M.2.2.1.1.a.

   Among the minimum requirements identified in the SRD was a KPP No. 1
   threshold that required the offeror's proposed aircraft to be "capable of
   aerial refueling all current [Air Force] tanker compatible fixed wing
   receiver aircraft using current [Air Force] procedures . . . ." RFP, SRD
   sect. 3.2.10.1.1.9. Another minimum requirement was a KPP No. 2 threshold
   that required the offeror's aircraft to be capable of satisfying the fuel
   offload versus unrefueled radius range as depicted in a linear graph
   contained in the RFP; this threshold charted the minimum pounds of fuel an
   aircraft must be capable of offloading to a receiver aircraft at a given
   distance of unrefueled flight by the tanker.[8] See RFP, SRD
   sect. 3.2.1.1.1.1. Also identified under KPP No. 2, as an objective, was
   that the "aircraft should be capable of exceeding the fuel offload versus
   unrefueled radius range as depicted in" this chart. RFP, SRD
   sect. 3.2.1.1.1.2.

   In addition, the SRD identified numerous key system attributes (KSA) for
   the aerial refueling, airlift, operational utility, survivability, and
   "other system requirements" areas, as well as numerous other "non-KPP/KSA
   requirements" for these areas that were desired but not required.[9] The
   RFP provided that these "requirements" did not have to be satisfied by the
   offerors, but were desired and considered part of the offerors' "design
   trade space."[10] RFP sect. M.2.2.1.1.b. With respect to these aspects of
   the evaluation of the key system requirements subfactor, offerors were
   informed that

     [f]or non-KPP requirements, the Government may give consideration for
     alternate proposed solutions or capabilities below the stated SRD
     requirement, depending on substantiating rationale. The Government may
     give additional consideration if the offeror proposes to meet (or exceed
     if there is an objective) the SRD threshold or requirement, depending on
     substantiating rationale.

   RFP sect. M.2.2.1.1.b. The RFP further stated that the Air Force sought an
   affordable KC-X system that not only met all of the KPP threshold
   requirements, but as many KSA and other SRD requirements as possible. RFP,
   SOO for KC-X SDD, at 2.

   Finally, with regard to the overall evaluation of the key system
   requirements subfactor, the RFP stated that "evaluation of the offeror's
   proposed capabilities and approaches against the SRD requirements will be
   made in the following descending order of relative importance: KPPs, KSAs,
   and all other non-KPP/KSA requirements." RFP sect. M.2.2.1.1.c.

   With respect to the aerial refueling area of the key system requirements
   subfactor, offerors were informed that the agency's evaluation would
   include "tanker aerial refueling, receiver aerial refueling, fuel offload
   versus radius range, drogue refueling systems (including simultaneous
   multi-point refueling), the operationally effective size of the boom
   envelope, the aerial refueling operator station and aircraft fuel
   efficiency." RFP sect. M.2.2.1.2.a. With respect to airlift area, the RFP
   provided that the agency's evaluation would include "airlift efficiency,
   cargo, passengers, aero-medical evacuation, ground turn time, and cargo
   bay re-configuration." RFP sect. M.2.2.1.2.b. Offerors were instructed
   with regard to this area to provide an airlift efficiency calculation,
   based upon a calculation procedure stated in the solicitation, that would
   result in a "payload pounds - nautical miles per pound fuel used"
   calculation (in other words, the weight of cargo per pound of fuel
   burned). RFP sect. L.4.2.2.4.1. Under the operational utility area, the
   agency's evaluation would include "aircraft maneuverability, worldwide
   airspace operations, communications/information systems (including
   Net-Ready capability), treaty compliance support, formation flight,
   intercontinental range, 7,000-foot runway operations, bare base airfield
   operations, and growth provisions for upgrades." RFP sect. M.2.2.1.2.c.
   The survivability area evaluation would include "situational awareness,
   defensive systems against threats, chemical/biological capability,
   [electromagnetic pulse] protection, fuel tank fire/explosion protection,
   and night vision capability." RFP sect. M.2.2.1.2.d. The remaining "other
   system requirements" area evaluated SRD requirements were not included in
   any of the other areas. RFP sect. M.2.2.1.2.e.

   Under the system integration and software subfactor, the evaluation was to
   consider the offeror's ability to implement a systems engineering approach
   and software development capability to satisfy the KC-X performance
   requirements, considering a number of listed attributes. RFP sect.
   M.2.2.2.

   Under the product support subfactor, the evaluation was to consider the
   offeror's product support approach that includes logistics planning and
   analysis; interim contractor support; transition to organic two-level
   maintenance support; approach and rationale for proposed operational
   availability, reliability and maintainability and mission capable rates;
   logistics footprint; site activation/beddown; and training. RFP sect.
   M.2.2.3.

   With respect to the program management subfactor, offerors were informed
   that the agency would assess whether "the offeror's proposal demonstrates
   a capability to effectively and efficiently implement and manage the KC-X
   Program." RFP sect. M.2.2.4. Included in this evaluation was whether the
   offeror demonstrated a "sound approach to achieving FAA
   Certification/Validation" and a "feasible, effective, low risk
   manufacturing and quality assurance approach to integrating military
   capability into the commercial baseline aircraft and transition to full
   rate production." RFP sections M.2.2.4.C, M.2.2.4.F.

   With respect to the past performance factor, the RFP informed offerors
   that the agency's performance confidence assessment group (PCAG) would
   conduct an in-depth review and evaluation of all performance data to
   determine how closely the work performed under those efforts related to
   the effort solicited under the RFP. The RFP provided that for this factor
   the agency would assess the degree of confidence that the agency had in an
   offeror's ability to perform the tanker contract, based upon an assessment
   of the offeror's demonstrated record of performance, and focusing on
   performance in five areas: the four mission capability subfactors and the
   cost/price factor.[11] RFP sect. 2.4.1. In this regard, the RFP stated
   that the agency would consider each offeror's, and its major/critical
   subcontractor's, demonstrated record of performance. Offerors were also
   informed that, in assessing an offeror's past performance, the agency
   would consider the relevance of an offeror's (and its subcontractor's,
   joint venture's, and teaming partner's) present and past performance, and
   that "[m]ore recent and more relevant performance by the same
   division/organization may have a greater impact on the performance
   confidence assessment than less recent or less relevant effort." Id.
   sect. M.2.4.5.3. With respect to an offeror's performance problems, the
   RFP stated:

     Where relevant performance records indicate performance problems, the
     Government will consider the number and severity of the problems and the
     appropriateness and effectiveness of any corrective actions taken (not
     just planned or promised). The Government may review more recent
     contracts or performance evaluations to ensure corrective actions have
     been implemented and to evaluate their effectiveness.

   RFP sect. M.2.4.4.

   With respect to the IFARA evaluation factor, the RFP provided that the
   agency would assess the utility and flexibility of a fleet of the
   offeror's proposed aircraft "by evaluating the number of aircraft required
   to fulfill the peak demand of the aerial refueling elements evaluated in
   the 2005 Mobility Capabilities Study."[12] Specifically, offerors were
   informed that the Air Force would analyze offeror-provided data in the
   evaluation scenario "primarily using the Combined Mating and Ranging
   Planning System (CMARPS) modeling and simulation tool" to calculate a
   "fleet effectiveness value," and would report this finding to the source
   selection authority (SSA), along with "any major insights and observations
   gleaned from the evaluation."[13] To calculate the fleet effectiveness
   value, the agency, using the CMARPS modeling tool, would calculate the
   number of KC-135R aircraft and the number of the offeror's proposed
   aircraft needed to satisfy the scenario, and then divide the number of
   KC-135R aircraft required by the number of the offeror's aircraft. The RFP
   stated that, with respect to this ratio, a fleet effectiveness value of
   1.0 would be equal in effectiveness to the KC-135R, while a value in
   excess of 1.0 would be viewed as more advantageous to the agency. RFP
   sect. M.2.6.

   Under the cost/price factor, the RFP provided that offerors' proposed
   costs and prices would be evaluated for realism and reasonableness,
   respectively. RFP sect. M.2.5. Offerors were also informed that the agency
   would calculate a most probable life cycle cost (MPLCC) estimate for each
   offeror, which was described by the solicitation to be "an independent
   government estimate, adjusted for technical, cost, and schedule risk, to
   include all contract, budgetary and other government costs associated with
   all phases of the entire weapon system life cycle (SDD, [Production and
   Deployment], and Operations and Support (O&S))." RFP sect. 2.5.2. The RFP
   provided that, as part of the "other government costs," the agency would
   evaluate anticipated MILCON costs associated with the offerors' proposed
   aircraft. RFP sect. 2.5.2.4. The RFP also provided that the agency would
   assess "technical, cost, and schedule risk for the entire most probable
   life cycle cost estimate based upon the offeror's proposed approach," and
   that the "impact of technical, schedule, and/or cost risk will be
   quantified (dollarized), where applicable, and included in the MPLCC." RFP
   sect. M.2.5.2.5.

   The RFP instructed the offerors to provide detailed cost information
   supported by a basis of estimate. Offerors were informed that the basis of
   estimate must

     completely describe the cost element content . . . philosophy, and
     methodology used to develop the estimate including appropriate
     references to any historical supporting cost date.

   RFP sect. L.6.4.7. The basis of estimate was required to include a
   "narrative with supporting data explaining how the proposed cost estimates
   (SDD, [production and deployment], O&S) were created." RFP sect. L.6.2.
   With respect to proposed O&S costs, which include fuel costs, offerors
   were informed that they should assume a 25-year system life from the date
   each aircraft is delivered and "calculate their O&S costs for 2 years
   beyond the date of their final production delivery"; to support their O&S
   cost projections, offerors were required to provide all "assumptions,
   ground rules, methodology, and supporting data."[14] RFP sections
   L.6.1.1.13, L.6.4.9. In this regard, the offerors were informed that if
   the historical data did not support the proposed prices, the cost
   documentation would be considered adequate only if the agency could
   understand the technical content, estimating methodology, and the
   "build-up" of the offerors' costs. RFP sect. L.6.4.7.

   Proposals

   The Air Force received proposals from Boeing and Northrop Grumman in
   response to the RFP. Boeing proposed as its KC-X aircraft the KC-767
   Advanced Tanker, a derivative of its commercial 767-200 LRF (long range
   freighter) aircraft.[15] The KC-767 was composed of elements of a number
   of Boeing commercial aircraft, including the 767-200ER, 767-300F, 767-400
   ER, 737, and 777 models. AR, Tab 61, Boeing Initial Technical Proposal,
   Executive Summary, at V1-ES-1. Boeing's proposed production plan for its
   SDD and production KC-X aircraft was to build the 767-200 LRF baseline
   aircraft at the Everett, Washington facility of its commercial division,
   Boeing Commercial Airplanes (BCA), and then fly the aircraft to its
   Wichita, Kansas facility for installation of military equipment and
   software by its military division, Integrated Defense Systems (IDS). Id.
   at V1-ES-2.

   Northrop Grumman proposed the KC-30 aircraft, which was a derivative of
   the Airbus A330-200 commercial aircraft.[16] AR, Tab 140, Northrop Grumman
   Initial Technical Proposal, Executive Summary, at I-1. Northrop Grumman
   proposed a production plan that provided for a number of changed locations
   for the production, assembly, and modification of its SDD and LRIP
   aircraft. For the first SDD aircraft, Northrop Grumman proposed to build
   the commercial A330 aircraft in sections in various European locations,
   then assemble the aircraft in Toulouse, France, add the cargo door in
   Dresden, Germany, and complete militarization of the aircraft in Madrid,
   Spain. For the second and third SDD aircraft, Northrop Grumman proposed
   using its own Melbourne, Florida facility, in place of EADS's Madrid
   facility, to complete militarization. For the last SDD aircraft, Northrop
   Grumman proposed replacing its Melbourne facility with a new facility it
   proposed to build in Mobile, Alabama. For the first LRIP aircraft,
   Northrop Grumman proposed to have the Toulouse facility not only assemble
   the commercial baseline aircraft but also install the cargo door, and the
   Mobile facility would complete the militarization of the aircraft.
   Beginning with the second LRIP aircraft, and thereafter through the
   production phase, Northrop Grumman proposed to build the A330 baseline
   aircraft in sections at various locations in Europe and then ship those
   sections to the Mobile facility, which would assemble the aircraft,
   install the cargo door, and complete militarization of the aircraft. Id.
   at I-6; see also Hearing Testimony (HT) at 1343-52.[17]

   SSET Evaluation

   The proposals were evaluated by the agency's SSET, which initiated
   discussions with the offerors by issuing evaluation notices (EN).[18]
   After evaluating the offerors' EN responses, the SSET provided a
   "mid-term" evaluation briefing to the SSAC and SSA. Because there were
   "concerns regarding how to properly show that all SRD requirements had
   been evaluated," the SSET prepared and provided another briefing to the
   SSA that detailed how each offeror's proposal was evaluated against each
   SRD requirement. COS at 24. Following the SSA's approval of the mid-term
   briefing, the SSET provided mid-term briefings to Boeing and Northrop
   Grumman, at which each offeror was provided with the agency's evaluation
   ratings of their respective proposals.[19] AR, Tabs 129, 130, Boeing's
   Mid-Term Briefings; Tabs, 199, 200, Northrop Grumman's Mid-Term Briefings.

   Following the offerors' mid-term briefings, the SSET provided a
   MPLCC/schedule risk assessment briefing to the SSAC and SSA, and
   subsequently the SSET provided MPLCC/schedule risk assessment briefings to
   the offerors. AR, Tab 133, Boeing's MPLCC/Schedule Risk Assessment
   Briefing; Tab 203, Northrop Grumman's MPLCC/Schedule Risk Assessment
   Briefing.

   Extensive discussions were conducted with each offeror, after which a
   "pre-final proposal revision" briefing was provided to the SSAC and SSA by
   the SSET that presented updated evaluation ratings of Boeing's and
   Northrop Grumman's proposals and discussion responses. Following approval
   of this briefing by the SSA, the SSET again provided to each offeror the
   agency's evaluation ratings of their respective proposals. AR, Tab 135,
   Boeing's Pre-Final Proposal Revision Briefing; Tab 205, Northrop Grumman's
   Pre-Final Proposal Revision Briefing.

   "Final revised proposals" were received from the offerors. Although
   intended by the agency to be the final proposal revisions, shortly after
   receipt of these proposals, the Air Force reopened discussions with the
   offerors in response to the enactment of the National Defense
   Authorization Act for Fiscal Year 2008, Pub. L. No. 110-181, 122 Stat. 3,
   208-12, 222-24 (2008).[20] As a part of these discussions, the Air Force
   provided offerors with additional information concerning the firms'
   respective IFARA evaluations and with a "clarified chart on Airlift
   Efficiency." COS at 25. Subsequently, the agency received the firms' final
   proposal revisions.

   The protester's and awardee's final proposal revisions were evaluated by
   the SSET as follows:

   +------------------------------------------------------------------------+
   |                                      |     Boeing     |Northrop Grumman|
   |--------------------------------------+---------------------------------|
   |Mission Capability/Proposal Risk      |                                 |
   |--------------------------------------+---------------------------------|
   | |Key System Requirements             |    Blue/Low    |    Blue/Low    |
   | |------------------------------------+----------------+----------------|
   | |System Integration/Software         | Green/Moderate | Green/Moderate |
   | |------------------------------------+----------------+----------------|
   | |Product Support                     |    Blue/Low    |    Blue/Low    |
   | |------------------------------------+----------------+----------------|
   | |Program Management                  |   Green/Low    |   Green/Low    |
   | |------------------------------------+----------------+----------------|
   | |Technology Maturity/Demonstration   |     Green      |     Green      |
   |--------------------------------------+----------------+----------------|
   |Past Performance                      |  Satisfactory  |  Satisfactory  |
   |                                      |   Confidence   |                |
   |                                      |                |   Confidence   |
   |--------------------------------------+----------------+----------------|
   |Cost/Price (MPLCC)                    |$108.044 Billion|$108.010 Billion|
   |--------------------------------------+----------------+----------------|
   | |Cost Risk                           |  Moderate/Low  |    Low/Low     |
   | |                                    |                |                |
   | |SDD Phase/Production & Deployment   |                |                |
   | |Phase                               |                |                |
   |--------------------------------------+----------------+----------------|
   |IFARA Fleet Effectiveness Value       |      1.79      |      1.9       |
   +------------------------------------------------------------------------+

   AR, Tab 46, SSET Final Briefing to SSAC and SSA, at 508, 532.

   As indicated by the nearly identical evaluation ratings received by both
   firms' technical proposals and the nearly identical evaluated MPLCCs, the
   competition was very close, and, as evaluated, both firms' proposals were
   found to be advantageous to the government. Ultimately, the SSAC
   concluded, however, that Northrop Grumman's proposal was more advantageous
   to the agency than Boeing's under the mission capability, past
   performance, cost/price, and IFARA factors; the two firms were found to be
   essentially equal under the proposal risk factor. AR, Tab 55, Proposal
   Analysis Report (PAR), at 46-48.

   SSAC's Mission Capability Factor Evaluation

   Northrop Grumman's evaluated advantage under the mission capability factor
   was largely based upon the firm's perceived superiority under the key
   system requirements and program management subfactors; the two firms were
   found essentially equal under the remaining three subfactors. Id. at
   46-47.

   The SSAC assigned both firms' proposals, under the key system requirements
   subfactor (the most important mission capability subfactor), blue, low
   risk ratings, noting:

     Both Offerors proposed to meet all KPP Thresholds. Both Offerors
     proposed capability beyond KPP Thresholds and offered significant trade
     space KSA capability. Additionally, both offered numerous non-KPP/KSA
     trade space capabilities deemed beneficial to the Government.

   Id. at 12. This assessment was documented in the SSAC's PAR, which
   identified evaluated "major discriminators," "discriminators offering less
   benefit" and weaknesses in each offeror's proposal in the aerial
   refueling, airlift, operational utility, survivability, and "other system
   requirements" areas of this subfactor.[21] Id. at 13-28.

   In the aerial refueling area, the SSAC noted "major discriminators" in
   favor of Boeing under several KPP No. 1 objectives, including its
   capability to [Deleted] and [Deleted], and for a "noteworthy non-KPP/KSA
   capability to [Deleted]. Id. at 13.

   The SSAC also noted a number of "major discriminators" in favor of
   Northrop Grumman in the aerial refueling area, including one under the KPP
   No. 2 objective for Northrop Grumman's proposal to exceed the RFP's fuel
   offload versus unrefueled radius range (Boeing's aircraft was also
   evaluated as exceeding this KPP objective but to a lesser degree),[22] and
   for a number of non-KPP/KSA requirements, including the proposal of a
   better aerial refueling efficiency (more pounds of fuel offload per pound
   of fuel used) than Boeing's; a "boom envelope" that was [Deleted] times
   greater than that defined by the Allied Technical Publication (ATP)-56[23]
   (Boeing proposed a boom envelope that was [Deleted] times greater than
   that defined by the publication); and a higher offload and receive fuel
   rate than Boeing. Id. at 13-14.

   In the aerial refueling area, the SSAC also identified five
   "discriminators offering less benefit" for Boeing that were assessed under
   14 different SRD requirements and one such discriminator for Northrop
   Grumman that was assessed under 2 SRD requirements. Id. at 15-16.

   The SSAC found that Boeing's proposal had no weaknesses in the aerial
   refueling area, but identified the following two weaknesses in Northrop
   Grumman's proposal:

     The first weakness is related to the specified lighting around the fuel
     receptacle of the KC-30. The specified lighting for refueling as a
     receiver may provide [Deleted]. The second weakness is related to
     Northrop Grumman's boom approach. The [Deleted].

   Id. at 16. The concern that Northrop Grumman's [Deleted] was assessed
   under a KPP No. 1 threshold; the other weaknesses were assessed under
   non-KPP/KSA requirements. No schedule or cost risk was assigned by the
   SSET or SSAC for either of Northrop Grumman's evaluated weaknesses. See
   AR, Tab 46, SSET Final Briefing to SSAC and SSA, at 196, 198; Tab 55, PAR,
   at 16.

   In the airlift area, the SSAC found that both offerors met all threshold
   requirements for the airlift KPP (there was only one KPP in this area),
   and that both offerors exceeded the threshold requirement for efficiently
   transporting equipment and personnel. AR, Tab 55, PAR, at 16. There were
   no KPP objectives identified by the SRD in the airlift area.

   The SSAC identified one "major discriminator" in favor of Boeing in the
   airlift area: Boeing satisfied the non-KPP/KSA requirement for the
   capability to [Deleted]. Id. at 17. With respect to Northrop Grumman, the
   SSAC identified a number of "major discriminators" in the airlift area.
   That is, with respect to carrying cargo, the SSAC found that Northrop
   Grumman had a better airlift efficiency capability than Boeing, showing an
   improvement of [Deleted] percent over that of the KC-135R, while Boeing's
   airlift efficiency showed only a [Deleted]-percent improvement over the
   KC-135R.[24] The SSAC noted that the KC-30 could carry more 463L
   pallets[25] than Boeing,[26] and that Northrop Grumman offered the
   capability to carry 463L pallets on both the main cargo deck and a lower
   cargo compartment, while Boeing only offered the single cargo deck. The
   SSAC also identified "major discriminators" in Northrop Grumman's proposal
   for passenger carriage ([Deleted] passengers to Boeing's [Deleted]
   passengers) and for aeromedical evacuation capability (Northrop Grumman
   could carry more litters and ambulatory patients). Id. at 18-19.

   Three "discriminators offering less benefit" were identified for Boeing in
   the airlift area and one such discriminator identified for Northrop
   Grumman. No proposal weaknesses were identified for either offeror in the
   airlift area. Id. at 19-20.

   In the operational utility area, the SSAC found that both offerors
   satisfied the three KPP thresholds identified in this area, and partially
   met the one KPP objective identified.[27] The SSAC also found that both
   offerors met the KSA thresholds and objectives in this area. Id. at 20.
   Two "major discriminators" were identified for Boeing in this area:
   (1) [Deleted] and (2) [Deleted]. Id. at 21. Two "major discriminators"
   were also identified for Northrop Grumman: (1) the KC-30 could operate
   from a 7,000-foot runway carrying approximately [Deleted] percent more
   fuel than the KC-767,[28] and (2) the KC-30 provided a ferry range of
   [Deleted] nautical miles as compared to the KC-767's ferry range of
   [Deleted] nautical miles.[29] Id. at 21-22. Numerous "discriminators
   offering less benefit" were identified for both Boeing and Northrop
   Grumman. Among such discriminators identified for Boeing was the KC-767's
   smaller ground footprint, which the SSAC found would enable the KC-767 to
   operate from bare base airfields with confined ramp space.[30] Id.  at 22.
   No proposal weaknesses were identified for either offeror in this area.

   Ultimately, the SSAC concluded, largely based upon Northrop Grumman's
   evaluated advantages in the aerial refueling and airlift areas, that
   Northrop Grumman's proposal was superior to Boeing's under the key system
   requirements subfactor.[31] Specifically, the SSAC noted:

     While [the] KC-767 offers significant capabilities, the overall
     tanker/airlift mission is best supported by the KC-30. [The] KC-30
     solution is superior in the core capabilities of fuel capacity/offload,
     airlift efficiency, and cargo/passenger/aeromedical carriage. These
     advantages in core capabilities outweigh the flexibility advantages of
     the attributes which Boeing offered (e.g. [Deleted], etc.)

   Id. at 29.

   Under the program management subfactor, the SSAC assigned both offerors
   green, low risk ratings, identifying no strengths, deficiencies, or
   uncertainties in either firm's proposal. Id. at 34. Nevertheless, the SSAC
   concluded that Northrop Grumman's program management approach was superior
   to that of Boeing, finding:

     Northrop Grumman's approach of providing four "green" aircraft for use
     early in SDD, by leveraging the existing A330 commercial production
     line, is deemed to be of benefit to the Government by reducing program
     risk. Northrop Grumman's approach adds value for the Government through
     increased confidence in overall program management.

   Id. at 46-47.

   Past Performance Factor Evaluation

   The SSAC found that both offerors had equal confidence ratings in four of
   the five past performance areas; the only difference in ratings was with
   respect to the program management area, under which Northrop Grumman's
   past performance was assessed as "satisfactory confidence" but Boeing's
   proposal was assessed as "little confidence."[32] Id. at 36. Boeing's
   little confidence rating for the program management area was based upon
   the Air Force's assessment of Boeing's past performance of the [Deleted]
   contract with [Deleted], of the [Deleted] contract with the [Deleted], and
   of the [Deleted] with the [Deleted]. The Air Force evaluated as marginal
   Boeing's past performance of these contracts, which were assessed as "very
   relevant." Id. at 37-38.

   IFARA Factor Evaluation

   The SSET also calculated a fleet effectiveness value for each proposed
   aircraft based upon offeror-provided data, which was analyzed under a
   variety of scenarios using the CMARPS modeling and simulation tool.[33] As
   noted above, the fleet effectiveness value reflected the quantity of an
   offeror's aircraft that would be required to perform the scenarios in
   relation to the number of KC-135R aircraft that would have been required.
   See RFP sect. M.2.6. The agency concluded that, whereas [Deleted] KC-135R
   aircraft would be required to perform the identified scenarios, the
   offerors' aircraft could perform the scenarios with fewer aircraft, that
   is, [Deleted] KC-30 aircraft and [Deleted] KC-767 aircraft. AR, Tab 55,
   PAR, at 45. The SSET calculated a fleet effectiveness value of 1.79 for
   the KC-767, and a higher (superior) value of 1.90 for the KC-30. Id. at
   44.

   The SSAC also noted a number of insights and observations concerning the
   IFARA evaluation of the offerors' aircraft. With respect to Boeing's
   proposed aircraft, the agency stated that, as compared to the KC-135R in
   the peak demand scenario:

     [the] KC-767 used [Deleted]% more ramp space (without requiring
     additional bases), burned [Deleted]% more fuel and was able to
     accomplish the scenarios with [Deleted] fewer aircraft when taking the
     aerial refueling receptacle into account. Additional aircraft were
     needed if every runway in the scenario were interdicted to 7,000 feet.
     In the base denial scenarios, when a base was closed, [Deleted]% of the
     Air Tasking Order (ATO) could be completed by basing KC-767s within the
     remaining bases' ramp space. Within the scenarios, [the] KC-767
     offloaded between [Deleted]% and [Deleted]% of its fuel.

   Id. at 45. With respect to Northrop Grumman's aircraft, the agency stated:

     [the] KC-30 used [Deleted]% more ramp space (needing some additional
     bases), burned [Deleted]% more fuel and was able to accomplish the
     scenarios with [Deleted] fewer aircraft when taking the aerial refueling
     receptacle into account. In the base denial sensitivity assessment, in
     some cases when a base was closed, the [Deleted]. [The] KC-30 has
     exceptional short field capability if the runway is interdicted to 7,000
     feet (as noted in Subfactor 1.1). Within the scenarios, [the] KC-30
     offloaded between [Deleted]% and [Deleted]% of its fuel.

   Id.

   Cost/Price Evaluation

   The Air Force calculated a MPLCC for each offeror, which, as noted above,
   was intended to be an independent government estimate of each proposal,
   adjusted for technical, cost and schedule risk and including all contract,
   budgetary and other government costs associated with all phases of the
   aircraft's entire life cycle (SDD, production and deployment, and O&S).
   See RFP sect. 2.5.2; COS at 124.

   With respect to Boeing's proposal, the Air Force made a number of
   adjustments in Boeing's proposed costs in calculating its MPLCC. For
   example, the agency added an additional $[Deleted] million to Boeing's
   proposed costs of $[Deleted] billion for SDD because the agency concluded
   that the firm had not adequately supported its basis of estimate for these
   costs, despite repeated discussions on this issue. Most of this adjustment
   ($[Deleted] million) was associated with a moderate risk rating that was
   assigned to Boeing's cost proposal to account for the agency's concern
   that Boeing had not adequately supported its proposed $[Deleted] billion
   for non-recurring engineering costs that Boeing estimated it would incur
   in the development of its proposed aircraft. As another example, the Air
   Force added $[Deleted] billion to Boeing's proposed costs for the
   production and deployment lots 6 through 13 (the budgetary aircraft)
   because the agency concluded that Boeing had not substantiated an
   approximately [Deleted]-percent decrease in proposed costs for these lots
   following the fixed-price production lots (lots 1 through 5). The Air
   Force also upwardly adjusted Boeing's MPLCC by $[Deleted] billion for
   "other government costs," the bulk of which ($[Deleted] billion) reflected
   additional O&S repair costs because the Air Force did not accept Boeing's
   estimating methodology of these costs. The agency also added additional
   costs to Boeing's MPLCC to account for the agency's estimated MILCON costs
   of $[Deleted] billion. AR, Tab 46, SSET Final Briefing to SSAC and SSA, at
   451-76; Tab 55, PAR, at 40-42.

   The Air Force also made a number of adjustments in Northrop Grumman's
   proposed costs, including upwardly adjusting the proposed SDD costs by
   $[Deleted] million and the firm's estimated costs for lots 6 through 13
   (budgetary aircraft) by $[Deleted] million. In addition, the Air Force
   added additional costs to Northrop Grumman's MPLCC to account for the
   agency's estimated MILCON costs of $[Deleted] billion. AR, Tab 46, SSET
   Final Briefing to SSAC and SSA, at 479-502; Tab 55, PAR, at 42-43.

   The Air Force calculated a MPLCC for Boeing of $108.044 billion and a
   MPLCC for Northrop Grumman of $108.010 billion.

   In comparing the firms' evaluated costs, the SSAC noted that Northrop
   Grumman had a lower evaluated MPLCC, but that the firms' evaluated MPLCCs
   were within $34 million of each other (approximately a .03-percent
   difference). The SSAC noted, however, that Boeing's slightly higher
   evaluated MPLCC was "driven" primarily by the firm's much higher SDD
   costs, "which reflected Boeing's more complex design, development, and
   integration activities." AR, Tab 55, PAR, at 43. In addition, the SSAC
   accepted the SSET's evaluation that Boeing's proposal presented a moderate
   cost risk for SDD. Northrop Grumman's proposal was assessed as a low cost
   risk for SDD costs. The SSAC viewed this difference in cost risk for the
   SDD phase to be the discriminator under this factor. Id. at 44.

   SSAC Recommendation

   Ultimately, the SSAC recommended to the SSA that [the SSA] select Northrop
   Grumman's proposal for award, because the SSAC concluded that Northrop
   Grumman's proposal was more advantageous under the mission capability,
   past performance, cost/price, and IFARA evaluation factors. With respect
   to cost/price, the SSAC noted that, although the difference between the
   two proposals' MPLCC was "negligible," Northrop Grumman's risk rating
   under this factor (low risk) for the SDD phase was lower than that
   assigned to Boeing's proposal (moderate cost/price risk) for the SDD
   phase. Id. at 46-48.

   Selection Decision

   As noted above, the SSA was presented with the SSET's evaluation results
   in a number of briefings at various stages in the procurement. In
   addition, the SSA was briefed by the SSAC with respect to that council's
   recommendation for award and was presented with the SSAC's detailed PAR,
   which documented the SSAC's weighing of the offerors' respective strengths
   and weaknesses and the SSAC's award recommendation.

   The SSA agreed with the SSAC's recommendation that Northrop Grumman's
   proposal reflected the best value to the agency, and [the SSA] identified
   Northrop Grumman's evaluated superiority under the mission capability,
   past performance, cost/price, and IFARA factors as supporting this
   conclusion; [the SSA] also concluded that neither offeror had an advantage
   under the proposal risk factor. With respect to the mission capability
   factor, the SSA emphasized that Northrop Grumman's evaluated superiority
   in the aerial refueling and airlift areas of the key system requirements
   subfactor were key factors in [the SSA's] decision.[34] AR, Tab 54, Source
   Selection Decision, at 9. Although not key to [the SSA's] determination
   that Northrop Grumman's proposal was more advantageous than Boeing's under
   the key system requirements subfactor, the SSA noted Boeing's evaluated
   superiority in the survivability area; [the SSA] also noted that neither
   offeror had an advantage in the operational utility area. Id. at 8-9.

   With respect to the aerial refueling area, the SSA noted that Northrop
   Grumman exceeded the KPP objective for fuel offload capability for the
   unrefueled radius range to a greater degree than did Boeing; this, the SSA
   found, demonstrated that a "single KC-30 can refuel more receivers or
   provide more fuel per receiver than a single KC-767." AR, Tab 54, Source
   Selection Decision, at 5-6. In addition, the SSA noted that Northrop
   Grumman offered a larger boom envelope than Boeing, and proposed a
   superior fuel offload and receive rate than Boeing. Id. at 6. Although
   Northrop Grumman had weaknesses in the aerial refueling area, and Boeing
   did not, the SSA agreed with the SSAC that the weaknesses (associated with
   receiver lighting and the firm's boom design) would have no impact on
   program cost and schedule. Id. at 6-7.

   With respect to the airlift area, the SSA noted Northrop Grumman's
   superior airlift efficiency, dual cargo deck configuration, and ability to
   carry more passengers and aeromedical litters and patients. The SSA
   concluded that the KC-30's airlift capability was "compelling to my
   decision." Id. at 7.

   In sum, the SSA selected Northrop Grumman's proposal for award, finding

     Northrop Grumman's proposal was better than Boeing's proposal in four of
     the five factors evaluated and equal in one. Northrop Grumman's offer
     was clearly superior to that of Boeing's for two areas within KC-X's
     Mission Capability factor: aerial refueling and airlift. Additionally,
     Northrop Grumman's KC-30's superior aerial refueling capability enables
     it to execute the IFARA scenario described in the RFP with [Deleted]
     fewer aircraft than Boeing's KC-767 -- an efficiency of significant
     value to the Government. I am confident that Northrop Grumman will
     deliver within the cost, schedule, and performance requirements of the
     contract because of their past performance and the lower risk of their
     cost/price proposal.

   Id. at 19.

   Award was made to Northrop Grumman on February 29, 2008, and following
   receipt of a required debriefing,[35] Boeing protested to our Office on
   March 11.

   DISCUSSION

   In its protest, which was supplemented numerous times as evaluation
   documents were provided during the development of the case,[36] Boeing
   challenges the Air Force's evaluation of technical and cost proposals,
   conduct of discussions, and source selection decision. In this regard, the
   protester identifies what it alleges are prejudicial errors under each of
   the RFP's evaluation factors and subfactors, and contends that, if the
   proposals had been evaluated in accordance with the RFP, its proposal
   would have been selected for award. As discussed below, we find a number
   of significant errors in the Air Force's evaluation under the key system
   requirements and product support subfactors of the mission capability
   evaluation factor and in its cost evaluation, and that the agency
   conducted misleading and unequal discussions with Boeing.[37]

   Document Production

   During the development of the record, Boeing requested that the Air Force
   provide various documents pursuant to our Bid Protest Regulations, 4
   C.F.R. sect. 21.3(g). We granted Boeing's requests in part where we were
   persuaded that the requested documents were relevant to the protest
   issues.

   The Air Force also requested that Boeing produce certain broad categories
   of documents bearing upon, among other things, Boeing's interpretation of
   the solicitation and several of its protest allegations. Boeing objected
   to that request, asserting that the documents sought were not relevant.
   The agency responded that its request was reasonable and limited, and
   sought relevant documents, which would be "necessary to allow GAO to
   perform a complete and accurate review of the issues in Boeing's
   protests." Air Force's Response to Boeing's Objection to Air Force's
   Document Production Request (Apr. 11, 2008) at 1.

   Our Bid Protest Regulations provide, in pertinent part, that "[i]n
   appropriate cases, the contracting agency may request that the protester
   produce relevant documents, or portions of documents, that are not in the
   agency's possession." 4 C.F.R. sect. 21.3(d).

   We denied the Air Force's request, because our regulations do not provide
   for broad agency request for documents whose existence and relevance are
   not at all apparent.[38] Rather, our regulations are intended to permit a
   contracting agency, in an appropriate case, to request a specific relevant
   document or documents, of which the agency is aware and does not itself
   possess. See 60 Fed. Reg. 40737, 40738 (wherein, in establishing this
   rule, we indicated that our regulations were not intended to allow
   "wide-open" document requests of protesters); see also Bid Protests at
   GAO: A Descriptive Guide, 8^th ed. 2006, at 28, in which our Office
   described the purpose for our "reverse discovery" rule as follows:

     Occasionally, the agency may be aware of the existence of relevant
     documents that only the protester possesses. In appropriate cases, the
     agency may request that the protester produce those documents.[39]

   Dismissal Requests

   Prior to the submission of the agency's report, the Air Force and Northrop
   Grumman requested that we summarily dismiss a substantial portion of
   Boeing's protest as untimely. The agency and intervenor argued that some
   of Boeing's protest grounds were untimely challenges to alleged, apparent
   solicitation improprieties. They also argued that some of Boeing's
   challenges to the agency's evaluation of proposals were untimely because
   Boeing was allegedly aware of the bases of these protest grounds during
   the competition, but did not protest until after award and the firm's
   receipt of a post-award debriefing.

   Our Bid Protest Regulations contain strict rules for the timely submission
   of protests. These timeliness rules reflect the dual requirements of
   giving parties a fair opportunity to present their cases and resolving
   protests expeditiously without disrupting or delaying the procurement
   process. Peacock, Myers & Adams, B-279327, Mar. 24, 1998, 98-1 CPD para.
   94 at 3-4; Professional Rehab. Consultants, Inc., B-275871, Feb. 28, 1997,
   97-1 CPD para. 94 at 2. Under these rules, a protest based on alleged
   improprieties in a solicitation that are apparent prior to closing time
   for receipt of proposals must be filed before that time. 4 C.F.R. sect.
   21.2(a)(1). Protests based on other than alleged improprieties in a
   solicitation must be filed not later than 10 days after the protester knew
   or should have known of the basis for protest, whichever is earlier. 4
   C.F.R. sect. 21.2(a)(2). Our regulations provide an exception to this
   general 10-day rule for a protest that challenges "a procurement conducted
   on the basis of competitive proposals under which a debriefing is
   requested and, when requested, is required." Id. In such cases, as here,
   with respect to any protest basis which is known or should have been known
   either before or as a result of the requested and required debriefing, the
   protest cannot be filed before the debriefing date offered, but must be
   filed not later than 10 days after the date on which the debriefing is
   held. Id.; see Bristol-Myers Squibb Co., B-281681.12, B-281681.13,
   Dec. 16, 1999, 2000 CPD para. 23 at 4.

   We did not, and do not now, agree with the Air Force and Northrop Grumman
   that Boeing's protest is a challenge to the ground rules established by
   the RFP for this procurement. We find that Boeing, rather than objecting
   to any of the RFP's requirements or evaluation criteria, is instead
   protesting that the Air Force failed to reasonably evaluate proposals in
   accordance with the RFP's identified requirements and evaluation
   criteria.[40] We also do not agree with the agency and intervenor that,
   because Boeing was informed during the competition of the agency's view of
   the merits of its proposal and/or how the proposals were being evaluated,
   Boeing was required to protest the agency's evaluation or evaluation
   methodology prior to award and to the protester's receipt of its required
   debriefing. Even where the protester is apprised of agency evaluation
   judgments with which it disagrees or where it believes the evaluation is
   inconsistent with the solicitation's evaluation scheme, our Bid Protest
   Regulations require that these protest grounds be filed after the receipt
   of the required debriefing.[41] See 4 C.F.R. sect. 21.2(a)(2); see also 61
   Fed. Reg. 39039, 39040 (July 26, 1996) ("to address concerns regarding
   strategic or defensive protests, and to encourage early and meaningful
   debriefings," a protester shall not file an initial protest prior to its
   required debriefing); Rhonda Podojil--Agency Tender Official, B-311310,
   May 9, 2008, 2008 CPD para. 94 at 3 (application of debriefing exception
   to A-76 competitions conducted on the basis of competitive proposals).

   Key System Requirements Subfactor Evaluation

   Boeing complains that the Air Force failed to evaluate the firms'
   proposals under the key system requirements subfactor--the most important
   subfactor of the mission capability factor--in accordance with the RFP's
   identified evaluation scheme. As noted above, under this subfactor, the
   agency was to assess the offerors' understanding of, and ability to meet,
   the various SRD requirements. Boeing argues that the Air Force did not
   reasonably consider the weighting assigned to the various SRD requirements
   by the RFP in making its source selection,[42] even though they involve
   the "major discriminators" referenced in justifying the award to Northrop
   Grumman.[43] Boeing also asserts that the evaluation did not account for
   the fact that the RFP specifically requested offerors to propose as many
   of these "trade space" requirements as possible. In this regard, Boeing
   complains that the agency assigned no credit for the fact that Boeing's
   aircraft satisfied significantly more trade space SRD requirements than
   did Northrop Grumman's under the key system requirements subfactor. See
   Boeing's Post-Hearing Comments at 18.

   The Air Force and Northrop Grumman deny that the agency failed to evaluate
   the firms' proposals in accordance with the solicitation criteria. They
   contend that the SSET performed an elaborate evaluation, "identifying
   specifically how Boeing and [Northrop Grumman] met or exceeded KPP
   thresholds, and how each traded, partially met or met desired requirements
   (trade space)." See, e.g., Air Force's Memorandum of Law at 62. The Air
   Force notes that the SSET identified potential strengths, which the SSAC
   categorized, as relevant here, as "major discriminators" or
   "discriminators offering less benefit." Id. The agency argues:

     Because Boeing and [Northrop Grumman] offered significant trade space
     and the benefits for each [SRD] reference line capability were not of
     equal value, the SSAC gave positive consideration for additional
     capability beyond the applicable threshold based upon the magnitude of
     benefit to the Air Force. The offerors' approaches, their relative
     benefits, advantages, and operational contributions for the five areas
     within [the] Key System Requirements [subfactor] were evaluated by the
     SSAC for accomplishing the comparative analysis. The SSAC deliberated
     extensively, using expert technical, engineering, and operational
     judgment to carefully evaluate the capabilities offered, consistent with
     the RFP Measures of Merit and the priorities of KPP, KSA, and
     non-KPP/KSAs. Both offerors proposed highly capable solutions to the
     requirements that offered tremendous benefit above current Air Force
     tanker capability.

   Id. at 63-64.

   An agency is obligated to conduct an evaluation consistent with the
   evaluation scheme set forth in the RFP. FAR sect. 15.305(a); Serco, Inc.,
   B-298266, Aug. 9, 2006, 2006 CPD para. 120 at 8. We recognize that
   proposal evaluation judgments are by their nature often subjective;
   nevertheless, the exercise of these judgments in the evaluation of
   proposals must be reasonable and must bear a rational relationship to the
   announced criteria upon which competing offers are to be selected. Systems
   Research and Applications Corp.; Booz Allen Hamilton, Inc., B-299818 et
   al., Sept. 6, 2007, 2008 CPD para. 28 at 11. In order for our Office to
   perform a meaningful review, the record must contain adequate
   documentation showing the bases for the evaluation conclusions and source
   selection decision. Southwest Marine, Inc.; American Sys. Eng'g Corp.,
   B-265865.3, B-265865.4, Jan. 23, 1996, 96-1 CPD para. 56 at 10.

   Here, we agree that the SSET's evaluation identified and documented the
   SRD requirements under which the firms' evaluated strengths and weaknesses
   were assessed. Nevertheless, the record does not establish that the SSAC
   and SSA, in considering those strengths and weaknesses, applied the
   relative weights identified in the RFP for the various SRD requirements
   (under which the KPPs were most important). Moreover, the record does not
   show any consideration by the SSAC or SSA of the fact that Boeing's
   proposal was evaluated as satisfying significantly more SRD requirements
   than Northrop Grumman's.

   For example, the record shows that most of Boeing's evaluated "major
   discriminators" in the aerial refueling area were assessed under KPP
   requirements, and conversely most of Northrop Grumman's evaluated "major
   discriminators" were assessed under less important non-KPP/KSA
   "requirements." Specifically, the SSAC identified as "major
   discriminators" the following requirements that Boeing's aircraft
   satisfied but Northrop Grumman's aircraft did not: (1) the capability to
   [Deleted] (a KPP No. 1 objective); (2) the capability, [Deleted] (another
   KPP No. 1 objective); (3) the capability to [Deleted] (another KPP No. 1
   objective); and (4) the capability to [Deleted] (a "noteworthy"
   non-KPP/KSA requirement).[44] See AR, Tab 55, PAR, at 13. The SSAC
   identified as a "major discriminator" for Northrop Grumman that firm's
   satisfaction of one KPP objective (KPP No. 2 objective for exceeding the
   fuel offload unrefueled range), where Boeing also satisfied this objective
   but to a lesser degree. The Air Force also identified as "major
   discriminators" for Northrop Grumman under this area the firm's better air
   refueling efficiency, larger boom envelope, and better offload and receive
   rates, all of which were non-KPP/KSA requirements.

   Although the record thus evidences that most of Boeing's evaluated "major
   discriminators" were assessed under KPP requirements, and conversely most
   of Northrop Grumman's evaluated "major discriminators" were assessed under
   less important non-KPP/KSA requirements,[45] we have found no document in
   the contemporaneous evaluation record that shows that the SSAC or SSA gave
   any meaningful consideration to the weights that were to be assigned to
   the various KPP, KSA, and other requirements. That is, the SSAC's briefing
   slides to the SSA and its PAR do no more than identify the SRD
   requirements for which the evaluated discriminators were assessed, but do
   not evidence any consideration of the descending order of importance
   assigned to these various SRD requirements.

   In its briefing to the SSA, the SSAC merely reports each of the firms'
   "advantages" without any analysis of whether or not Boeing's "advantages"
   (which as indicated above are mostly derived from KPP objectives) were
   entitled to greater weight than Northrop Grumman's advantages (which are
   mostly derived from less important non-KPP/KSA requirements). See, e.g.,
   AR, Tab 55, SSAC Recommendation Briefing to SSA, at 6-7 (aerial refueling
   discriminators). Similarly, in the PAR, the SSAC duly reports the relative
   order of importance that was to be assigned to the KPP, KSA and other
   requirements, see AR, Tab 55, PAR, at 4, but there is no suggestion that
   the assigned weights to these requirements were applied in any of the
   SSAC's comparative analyses of the firms' evaluated discriminators. See,
   e.g., id. at 13-14 (aerial refueling discriminators). Thus, although it is
   true that the SSAC reported in the PAR that it considered the "priorities
   of KPP, KSA, and non-KPP/KSAs," see id. at 29, the record does not provide
   any evidence of such a weighing.

   Similarly, in [the SSA's] selection decision, the SSA reports that the
   evaluation of the "offerors' proposed capabilities and approaches against
   the SRD requirements were made in the following order of importance: KPPs,
   KSAs, and all other non-KPP/KSA requirements." See AR, Tab 54, Source
   Selection Decision, at 5. Despite this reported recognition of the varying
   weights assigned to the different SRD requirements, the SSA's decision
   document does not evidence any consideration of the fact that Boeing's
   assessed "major discriminators" were derived from requirements that were
   identified as being more important than most of the requirements from
   which Northrop Grumman's discriminators were derived. See id. at 5-7.

   We agree with the Air Force that it is permissible to identify relative
   strengths found under less important evaluation factors to be
   discriminators for selection purposes, where there are lesser relative
   differences favoring another proposal under more important evaluation
   factors. However, we find no evidence in this record that any such
   analysis, which considered the relative weight of the KPPs, KSAs and
   non-KPP/KSA requirements, was performed here.

   The Air Force also identified more "discriminators offering less benefit"
   for Boeing's proposal than for Northrop Grumman's proposal in the aerial
   refueling area. Specifically, the SSAC identified five such discriminators
   for Boeing that were assessed under 13 different SRD requirements, and
   only one such discriminator for Northrop Grumman that was assessed under 2
   SRD requirements. As noted above, the RFP requested that offerors satisfy
   as many of the "trade space" SRD requirements "as possible." See RFP, SOO
   for KC-X SDD, at 2. Despite having solicited proposals that satisfy as
   many SRD requirements as possible, there is no evidence in the record
   showing that either the SSAC or the SSA accounted for the fact that
   Boeing's proposal was evaluated as satisfying significantly more SRD
   requirements in the aerial refueling area than did Northrop Grumman's
   proposal. In fact, in deciding that Northrop Grumman had a significant
   advantage in the aerial refueling area, the SSA did not even discuss the
   fact that Boeing had more "discriminators offering less benefit" than did
   Northrop Grumman, much less that Boeing satisfied far more SRD
   requirements than did Northrop Grumman in this area.

   As noted by the Air Force, the assignment of adjectival ratings and the
   source selection should generally not be based upon a simple count of
   strengths and weaknesses, but upon a qualitative assessment of the
   proposals. See Kellogg Brown & Root Servs., Inc., B-298694.7, June 22,
   2007, 2007 CPD para. 124 at 5. Such a qualitative assessment must be
   consistent with the evaluation scheme, however. Here, although the RFP
   expressly encouraged offerors to satisfy as many of the "trade space" SRD
   requirements "as possible," see RFP, SOO for KC-X SDD, at 2, the record
   shows no evidence that the Air Force gave any consideration to Boeing's
   offer to satisfy significantly more trade space SRD requirements. This, in
   our view, is not a matter of simply counting strengths and weaknesses, but
   of evaluating the advantages and disadvantages of competing proposals
   consistent with the RFP's evaluation scheme. See, e.g., Systems Research
   and Applications Corp.; Booz Allen Hamilton, Inc., supra, at 14.

   In short, our review of the record indicates that, as illustrated by the
   aerial refueling area examples above, the Air Force failed to evaluate
   proposals in accordance with the RFP's evaluation criteria.[46] That is,
   the record evidences that the Air Force failed to assess the relative
   merits of the offerors' proposals based upon the importance assigned to
   the various SRD requirements by the RFP or to account for the fact that
   Boeing proposed to satisfy far more SRD requirements than did Northrop
   Grumman.

   Fuel Offload versus Unrefueled Radius Range KPP Objective

   Boeing protests that one of the key discriminators relied upon by the SSA
   in [the SSA's] selection decision was contrary to the RFP's evaluation
   criteria. This contention concerns the significant discriminator assessed
   by the Air Force under the aerial refueling area of the key system
   requirements subfactor. The assessed significant discriminator reflects
   the conclusion that Northrop Grumman's proposed aircraft exceeded to a
   greater degree than Boeing's aircraft a KPP objective to exceed the RFP's
   identified fuel offload to the receiver aircraft versus the unrefueled
   radius range of the tanker. The SSA noted in this regard that Northrop
   Grumman's aircraft exceeded the threshold by [Deleted] percent at 1,000
   nautical miles and by [Deleted] percent at 2,000 nautical miles, whereas
   Boeing's aircraft exceeded the threshold by [Deleted] percent at
   1,000 nautical miles and by [Deleted] percent at 2,000 nautical miles. AR,
   Tab 54, Source Selection Decision, at 5. This was a key reason supporting
   the SSA's determination that Northrop Grumman's proposed aircraft was more
   advantageous than Boeing's aircraft in the aerial refueling area and was
   superior overall to Boeing's. See id. at 6-7, 9, 19.

   The RFP informed offerors that the agency would evaluate the offerors'
   approach to meeting the SRD requirements related to aerial refueling,
   which would include the fuel offload versus radius range. RFP
   sect. M.2.2.1.2.a. With respect to fuel offload versus unrefueled range,
   the RFP identified as a KPP threshold (a mandatory minimum requirement)
   the range that offerors must satisfy to be found acceptable. See RFP, SRD
   sect. 3.2.1.1.1.1. The RFP also identified as a KPP objective that the
   offerors' "aircraft should be capable of exceeding" the threshold. See
   RFP, SRD sect. 3.2.1.1.1.2. Finally, the RFP specifically informed
   offerors that "[n]o consideration will be provided for exceeding KPP
   objectives." RFP sect. M.2.2.1.1.a.

   Boeing argues that section M.2.2.1.1.a. of RFP unambiguously prohibited
   crediting Northrop Grumman for exceeding the fuel offload versus
   unrefueled range objective to a greater extent than Boeing. Boeing asserts
   that this limitation on providing credit for exceeding KPP objectives
   "played an important role in shaping . . . how offerors designed and
   selected the aircraft that was proposed to meet the stated SRD
   requirements," see Protester's Comments at 14, and states that, had Boeing
   known of the agency's desire for a larger aircraft which can carry more
   fuel, it likely would have offered the agency an aircraft based upon the
   777 aircraft platform.[47] See Protest at 2, 40.

   The Air Force and Northrop Grumman respond that the agency "appropriately
   found [Northrop Grumman's] superior ability to offload fuel at radius to
   be a major discriminator and of operational benefit to the Air Force." Air
   Force's Memorandum of Law at 70; see Northrop Grumman's Comments at 18-19.
   In this regard, the agency and intervenor argue, despite the plain
   solicitation language cited above by the protester, that the RFP, read as
   a whole, indicated to offerors that the agency would consider, and award
   credit for, the amount by which offerors proposed to exceed the fuel
   offload versus unrefueled radius range chart identified in the KPP. In
   this regard, the Air Force and Northrop Grumman argue that this KPP
   objective did not identify an objective level, and therefore this
   particular objective was "unbounded," such that unlimited credit could be
   provided for exceeding this KPP objective. See, e.g., Air Force's Request
   for Partial Dismissal at 19; Northrop Grumman's Post-Hearing Comments
   at 102. The Air Force argues that:

     [t]he RFP made clear that the Air Force desired maximum fuel offload at
     radius because it described the objective in qualitative rather than
     quantitative terms, thereby placing both offerors on notice that the
     extent to which each offeror's proposed solution exceeded the threshold
     could become a potential discriminator between the offerors.

   Air Force's Memorandum of Law at 70. The agency also argues that, reading
   this KPP objective to exceed the fuel offload versus radius range
   threshold, see RFP, SRD sect. 3.2.1.1.1.2, in conjunction with the
   non-KPP/KSA trade space requirement that the aircraft "should operate with
   maximum fuel efficiency," see RFP, SRD sect. 3.2.1.1.1.3, offerors should
   have known that the agency would be giving credit under this KPP objective
   for the degree to which the offerors would exceed the charted KPP
   threshold with no upward limits. See Air Force's Request for Partial
   Dismissal at 17. Northrop Grumman contends that Boeing's reading of this
   provision is inconsistent with the general nature of what the Air Force
   sought, which Northrop Grumman argues was "a greater refueling capacity,
   including the possibility of reducing the number of airplanes required to
   complete a mission." Northrop Grumman's Comments at 27.

   Where, as here, a dispute exists as to the actual meaning of a particular
   solicitation provision, our Office will resolve the matter by reading the
   solicitation as a whole and in a manner that gives effect to all its
   provisions; to be reasonable, an interpretation of a solicitation must be
   consistent with such a reading. Stabro Labs., Inc., B-256921, Aug. 8,
   1994, 94-2 CPD para. 66 at 4.

   We find from our review of the solicitation that the offerors were
   unambiguously informed that their proposals would not receive additional
   consideration or credit for exceeding a KPP objective. This is true
   whether we look to the express provision itself, the meaning of which is
   plain, or whether we view this restriction within the context of the whole
   solicitation. The only reasonable interpretation of the KPP objective here
   is that an offeror would be credited for meeting the fuel offload versus
   unrefueled radius range objective if its aircraft exceeded the charted KPP
   threshold, and that no additional credit would be provided for exceeding
   the charted threshold amount to a greater degree than other proposed
   aircraft.

   Contrary to the Air Force's and Northrop Grumman's positions that this KPP
   objective was "unbounded" because no finite number or level is stated as
   part of the objective, the plain language of section M.2.2.1.2.a. of the
   RFP unequivocally prohibited any consideration for exceeding the stated
   KPP objective and the RFP did not suggest that the stated objective must
   be finite or be at an objective level in order for this section to be
   applicable. To read this provision as suggested by the intervenor and
   agency would render meaningless section M.2.2.1.2.a, and be inconsistent
   with identification of an objective for this KPP threshold. See Brown &
   Root, Inc. and Perini Corp., a joint venture, B-270505.2, B-270505.3,
   Sept. 12, 1996, 96-2 CPD para. 143 at 8 (a solicitation should be
   reasonably read to give effect to all of its provisions). We do not find
   such a reading reasonable.

   The Air Force, as the drafter of the RFP, could have provided for
   unbounded consideration of the degree to which offerors exceeded the fuel
   offload versus unrefueled range, but did not. In fact, the last sentence
   in section M.2.2.1.1.a. states that "[i]f there is no objective and,
   depending on substantiating rationale, positive consideration will be
   provided when the specified capability above the KPP threshold is viewed
   as advantageous to the Government." Thus, according to the RFP,
   "unbounded" credit could be given for exceeding the KPP where no KPP
   objective is stated (depending on the substantiating rationale and when
   advantageous to the government).[48] Indeed, the solicitation contained a
   number of KPP thresholds that did not have corresponding KPP objectives,
   see, e.g., RFP, SRD sect. 3.2.1.6.1. (KPP No. 4, Airlift Capability);
   sect. 3.2.8 (KPP No. 8, Survivability), but that is not the case with
   respect to this KPP threshold.

   We also note that the RFP elsewhere specifically informed offerors of
   other objectives for which their proposals could receive additional
   consideration for exceeding objectives; that is, with respect to non-KPP
   requirements, the RFP stated that the agency may give "additional
   consideration if the offeror proposes to meet (or exceed if there is an
   objective) the SRD threshold or requirement, depending on the
   substantiating rationale." See RFP sect. M.2.2.1.1.b. In addition,
   offerors were informed with regard to certain non-KPP objectives that they
   should try to exceed the requirement by as much as possible. See, e.g.,
   RFP, SRD sect. 3.2.10.1.5.2.2 ("The boom envelope should exceed the ATP-56
   envelope as much as possible (OBJECTIVE).")

   We also agree with Boeing that the RFP, read as whole, established a
   complex set of trade-offs for offerors to consider in determining what
   aircraft to propose to the agency, and we do not agree that "common sense"
   mandates that "unbounded" refueling capabilities were being sought by the
   RFP. [49] Although it is apparent that a larger aircraft could provide
   greater refueling capabilities, there could be associated disadvantages
   with respect to costs and space constraints. Thus, given that the RFP did
   not establish any size requirements or limitations upon the aircraft that
   could be proposed, the restriction on credit for exceeding this KPP
   objective provided offerors with a key consideration in determining what
   sort of aircraft to offer, as well as how to best structure their
   proposals.

   As indicated above, the Air Force and Northrop Grumman argued that
   Boeing's protest of the agency's evaluation of the firms' proposal under
   this KPP objective is untimely because it is actually a challenge to the
   terms of the solicitation.[50] They base this argument upon their
   contention that Boeing learned of the agency's interpretation from the
   agency's briefings during the competition. However, we agree with Boeing's
   contention that the agency's briefings supported Boeing's understanding
   that no credit would be given for exceeding this KPP objective. For
   example, in Boeing's mid-term briefing, the Air Force reported to Boeing
   with regard to the aerial refueling area of the key system requirements
   that, although its aircraft exceeded the fuel offload versus unrefueled
   range and the agency identified by how much Boeing's aircraft exceeded the
   range, its proposal was evaluated to have "met" the objective.[51] See AR,
   Tab 129, Mid-term Briefing to Boeing, at 26. Similarly, in its pre-Final
   Proposal Revision Briefing, Boeing was informed that its offer to exceed
   the KPP threshold for this requirement was evaluated as having "met" the
   objective.[52] See AR, Tab 135, Pre-Final Proposal Revision Briefing to
   Boeing, at 30. Based on our review of the record, Boeing was not informed
   in its briefings of the SSA's and SSAC's interpretation that the RFP
   allowed "unbounded" credit to be given for exceeding the fuel offload
   versus unrefueled radius range KPP objective, and only became aware of the
   agency's interpretation from the redacted source selection decision that
   was provided to Boeing at its post-award required debriefing.[53]

   In sum, we find that a key discriminator relied upon by the SSA in making
   [the SSA's] selection decision--that is, the assessment related to the KPP
   objective to exceed the fuel offload versus unrefueled range--was not
   consistent with the RFP. It is a fundamental principle of competitive
   procurements that competitors be treated fairly, and fairness in
   competitions for federal procurements is largely defined by an evaluation
   that is reasonable and consistent with the terms of the solicitation. For
   that reason, agencies are required to identify the bases upon which
   offerors' proposals will be evaluated and to evaluate offers in accordance
   with the stated evaluation criteria. See Competition in Contracting Act of
   1984, 10 U.S.C. sect. 2305(a)(2)(A), (b)(1) (2000); FAR sections
   15.304(d), 15.305(a); Sikorsky Aircraft Co.; Lockheed Martin Sys.
   Integration-Owego, B-299145 et al., Feb. 26, 2007, 2007 CPD para. 45 at 4.
   The Air Force did not fulfill this fundamental obligation here.

   KC-30 Overrun and Breakaway Capability

   Boeing also complains that the Air Force did not reasonably assess the
   capability of Northrop Grumman's proposed aircraft to refuel all current
   Air Force fixed-wing tanker-compatible aircraft using current Air Force
   procedures, as required by a KPP No. 1 threshold under the aerial
   refueling area of the key system requirements subfactor.[54] See RFP, SRD
   sect. 3.2.10.1.1.9. Specifically, Boeing notes that current Air Force
   refueling procedures require that the tanker aircraft be capable of
   "overrun" and "breakaway" procedures when necessary, which would require
   the tanker aircraft to fly faster than the receiver aircraft or quickly
   accelerate during refueling.[55] Boeing's Second Supplemental Protest at
   29. Boeing contends that the Air Force unreasonably determined that
   Northrop Grumman's proposed aircraft would meet these requirements.

   With regard to the overrun issue, the record shows that Northrop Grumman
   was twice informed by the Air Force during discussions that the firm's
   initially identified maximum operational airspeed of [Deleted] Knots
   Indicated Air Speed (KIAS) would not be sufficient under current Air Force
   overrun procedures to achieve required overrun speeds of [Deleted] KIAS
   for various fighter aircraft, including the [Deleted], or [Deleted] KIAS
   for the [Deleted].[56] See AR, Tab 184, EN NPG-MC1-003, at 2;
   EN NPG-MC1-003a, at 1-2. Ultimately, Northrop Grumman informed the Air
   Force that a [Deleted] limited the aircraft's operational speed, but that
   Northrop Grumman proposed to include a [Deleted] to achieve the necessary
   overrun speed.[57] See id., Northrop Grumman Response to EN NPG-MC1-003a,
   at 2-7. The Air Force accepted Northrop Grumman's proposed solution as
   satisfying this KPP threshold. HT at 628.

   Boeing complains that Northrop Grumman's proposed solution of [Deleted] to
   achieve overrun speed requires [Deleted], which is not consistent with the
   Air Force's current procedures as is required by the KPP.[58] See Boeing's
   Second Supplemental Protest at 29-32; Boeing's Comments at 64. Boeing also
   argues that the agency did not note that Northrop Grumman qualified its
   promise to increase its maximum operational airspeed in its EN response.
   Specifically, Boeing points out that Northrop Grumman stated that,
   [Deleted], the KC-30 had a maximum airspeed of [Deleted] KIAS, and not the
   [Deleted] KIAS evaluated by the Air Force. See AR, Tab 184, Northrop
   Grumman Response to EN NPG-MC1-003a, at 9.

   At the hearing that our Office conducted in this protest, the Air Force
   produced its SSET mission capability factor team chief to testify
   regarding the agency's evaluation of the capability of Northrop Grumman's
   aircraft to satisfy this KPP threshold.[59] This witness, in response to
   direct examination, stated that the SSET found that [Deleted] would allow
   the KC-30 to achieve the necessary airspeed to perform the required
   overrun and breakaway procedures. Specifically, he testified that the SSET
   was convinced that, by [Deleted], the KC-30 could achieve an operational
   airspeed of [Deleted] KIAS, because Northrop Grumman had informed the
   agency in its EN response that the commercial A330 aircraft had a maximum
   "dive velocity"[60] of 365 KIAS and had been flight tested to a dive
   velocity of [Deleted] KIAS, and that analysis had been done showing that
   the A330 could even achieve a dive velocity of [Deleted] KIAS. HT at
   626-27. The mission capability factor team chief testified that the SSET
   evaluated Northrop Grumman's response to indicate that the [Deleted], see
   HT at 637-38, and that in any event Air Force current procedures did not
   require the use of the [Deleted] during aerial refueling operations. HT
   at 638-39.

   From this record, we cannot conclude that the Air Force reasonably
   evaluated the capability of Northrop Grumman's proposed aircraft to
   satisfy the KPP threshold requirement to refuel all current Air Force
   fixed-wing tanker-compatible aircraft using current Air Force procedures.
   The contemporaneous record, as explained by the hearing testimony, does
   not establish that the Air Force understood Northrop Grumman's response in
   discussions concerning its ability to satisfy the solicitation
   requirements, nor does it demonstrate that the agency had a reasonable
   basis upon which to accept Northrop Grumman's promises of compliance.

   First, we agree with Boeing that the SSET erred in concluding that the
   [Deleted] in tanker refueling operations was not a current Air Force
   procedure.[61] See HT at 638, 735; Air Force's Post-Hearing Comments at
   19. As noted above, the contemporaneous evaluation record shows that the
   agency interpreted the solicitation requirement to comply with "current
   [Air Force] procedures" to mean compliance with the procedures set forth
   in the agency's flight manuals for the KC-135 and KC-10 tanker aircraft,
   and expressly informed Northrop Grumman during discussions that the flight
   manuals for the KC-135 and KC-10 established the current Air Force
   procedures for refueling operations. See AR, Tab 184, EN NPG-MC1-003a, at
   1, wherein the agency stated "[a]erial refueling procedures were contained
   in T.O. 1-1C-1-3 and 1-1C-1-33 for the KC-135 and KC-10 respectively when
   the RFP was released."[62] These manuals show that current Air Force
   procedures provide that tanker pilots [Deleted] in refueling operations.
   For example, the KC-135 manual under Section IV, Air Refueling Procedures,
   warns tanker pilots that they "must be prepared to assume aircraft control
   [Deleted]," and under Section V, Emergency Air Refueling Procedures,
   instructs tanker pilots that in a breakaway situation, if a climb is
   required, they must "[Deleted]." See AR, Tab 289, Flight Manual KC-135
   (Tanker) Flight Crew Air Refueling Procedures, Supp. III, T.O. 1-1C-1-3,
   Jan. 1, 1987, as revised Sept. 1, 2004, at [Deleted]. Similarly, the KC-10
   flight manual provides under Section III, Air Refueling Procedures, that
   the "[Deleted]." Id., Flight Manual, KC-10A Aircraft, Flight Crew Tanker
   Air Refueling Procedures, USAF Series, T.O. 1-1C-1-33, Sept. 1, 2002, as
   revised Jan. 31, 2005, at [Deleted]. In this regard, Boeing provided the
   statement of a retired Air Force pilot, who had extensive experience as
   both a KC-10 and KC-135 tanker pilot and had operated each aircraft as
   both a tanker and a receiver in refueling missions; this individual
   stated:

     Refueling is more demanding and difficult for both tanker and receiver
     aircraft if the tanker [Deleted]. For the tanker pilot, [Deleted]. For
     the receiver pilot, [Deleted]. Due to these realities, existing
     refueling guidelines dictate that [Deleted] should be used for refueling
     under normal circumstances. [Citations omitted.] Beginning aerial
     refueling [Deleted] should it become necessary, violates this policy. As
     previously noted, [Deleted].

   Boeing's Comments, attach. 14, Declaration of Retired Air Force Pilot, at
   3-4. Although the Air Force and Northrop Grumman generally disagree with
   Boeing's consultant that the Air Force's current procedures provide for
   the [Deleted], neither the agency or intervenor have directed our
   attention to anything in the KC-135 or KC-10 flight manuals or to any
   other source that would establish that Boeing's view, which appears to be
   reasonable on its face, is in error.

   We also find unsupported the agency's conclusion that Northrop Grumman's
   proposed solution of [Deleted] did not also involve [Deleted]. In its EN
   response, Northrop Grumman informed the Air Force that 330 KIAS was the
   normal design maximum operating velocity of the commercial A330 aircraft,
   and that "selection of a [maximum operating velocity] drives overall
   design characteristics of the aircraft, specifically aerodynamic and
   structural design limits, handling quality definition, and thrust." See
   AR, Tab 184, Northrop Grumman Response to EN NPG-MC1-003a, at 2. Northrop
   Grumman explained that its [Deleted] limited the aircraft to its maximum
   operating velocity, but that the firm could [Deleted] to exceed the
   maximum operating velocity. The awardee then stated "three cases . . . to
   illustrate the performance of the KC-30 with and without [Deleted]." Id.
   at 3. The three cases that Northrop Grumman identified and separately
   described were (1) KC-30 [Deleted]; (2) KC-30 [Deleted]; and (3) KC-30
   [Deleted], which indicated that the KC-30 could only meet the overrun
   requirement under the third case where both the [Deleted]. Id. at 3-6.

   The SSET read, as described by the testimony of its mission capability
   factor team chief, Northrop Grumman's EN response to describe a "fourth
   case" (although not identified as such) under the "third case" heading,
   but located at the end of that section, where, the agency contends, the
   KC-30's [Deleted] but the [Deleted]. See HT at 664. However, we are unable
   to accept such a reading of Northrop Grumman's EN response. It ignores the
   logical structure of Northrop Grumman's response to the agency, which only
   identified and described three cases. Moreover, nowhere in its response to
   the agency's EN does Northrop Grumman suggest a "fourth case" where the
   [Deleted]; rather, the only reference to both the [Deleted] in the third
   case expressly states that the [Deleted] ("Case 3: KC-30 [Deleted]").[63]
   See AR, Tab 184, Northrop Grumman Response to EN NPG-MC1-003a, at 6. In
   any event, given the uncertainty surrounding the agency's interpretation
   of Northrop Grumman's solution to a matter the agency believed could
   render the firm's proposal unacceptable, see HT at 625, 649, this is
   something the agency should have continued to clarify and resolve during
   discussions with the firm.[64]

   Even apart from the agency's apparent misreading of Northrop Grumman's EN
   response and disregard of the current Air Force procedure to [Deleted],
   the record does not establish that the agency had a reasonable basis for
   concluding that Northrop Grumman's proposed solution would allow its
   aircraft to obtain the requisite overrun airspeeds to satisfy this KPP
   threshold. The witness that the Air Force produced to support its
   arguments on this point testified that the SSET had concluded that the
   KC-30 had the "inherent capability" of reaching airspeeds greater than
   [Deleted] KIAS (the aircraft's certified maximum operational airspeed)
   based upon the far greater airspeed ([Deleted] KIAS) identified by the
   firm for its certified dive velocity.[65] See HT at 624-28; Air Force's
   Post-Hearing Comments at 17-18. In this regard, the SSET apparently
   believed that simply [Deleted] would enable the aircraft to achieve its
   indicated dive velocity airspeed as its operational airspeed.

   Although the SSET mission capability factor team chief repeatedly
   testified that the dive speed indicated that the aircraft would have the
   structural ability to fly at the dive speed limitation, see, e.g., HT at
   674, he also admitted under cross examination that he did not know what
   the relationship was between maximum operating airspeed and design dive
   speed:

   Q: What's your understanding of what the general margin is between maximum
   operational velocity and dive velocity?

   A: I'm not aware.

   Q: Was there somebody on your team that was advising you about what the
   general margin is or difference is between maximum operational velocity
   and dive velocity?

   A: There could have been. We had advisors for handling qualities.

   Q: I know you had advisors. I'm asking you, were there any advisors who
   actually helped you with understanding the difference between dive
   velocity and maximum operational velocity?

   A: They did not help me, no.

   Q: Did they help the team?

   A: Not that I'm aware of.

   HT at 669-70. The SSET mission capability factor team chief's (and
   presumably the SSET's) lack of knowledge concerning the relationship
   between maximum operating airspeed and design dive airspeed[66] is
   particularly troubling given the definition of maximum operating limit
   speed in FAA's regulations:

     The maximum operating limit speed . . . is a speed that may not be
     deliberately exceeded in any regime of flight (climb, cruise, or
     descent), unless a higher speed is authorized for flight test or pilot
     training operations. [The maximum operating limit speed] must be
     established so that it is not greater than the design cruising speed
     . . . and so that it is sufficiently below [dive speed and velocity] to
     make it highly improbable that the latter speeds will be inadvertently
     exceeded in operations.

   14 C.F.R. sect. 25.1505.

   In sum, despite having identified, as an issue for the hearing, the
   capability of Northrop Grumman's proposed aircraft to satisfy the airspeed
   requirements of this KPP threshold, we have been presented with no
   testimony or documented analysis that explains why simply [Deleted] on the
   KC-30 would ensure that the proposed aircraft would achieve required
   overrun airspeeds that are in excess of its FAA certified maximum
   airspeed.[67] Furthermore, neither the Air Force nor Northrop Grumman has
   directed us to any documentation establishing that the agency analyzed
   what would be entailed in designing the KC-30 to exceed the certified
   maximum operational airspeed limit.[68] Given Northrop Grumman's
   recognition in its EN response that selection of the maximum operational
   airspeed limit "drives overall design characteristics of the aircraft,
   specifically the aerodynamic and structural design limits, handling
   quality definition, and thrust," see AR, Tab 184, Northrop Grumman
   Response to EN NPG-MC1-003a, at 2, it would seem apparent that some design
   and FAA re-certification efforts could be necessary.

   Boeing also complains that the Air Force did not reasonably evaluate the
   capability of Northrop Grumman's aircraft to initiate emergency breakaway
   procedures when refueling the [Deleted].[69] Current Air Force procedures,
   as reflected by the KC-135 flight manual, specifies that the tanker will
   refuel the [Deleted] at an airspeed of [Deleted] KIAS, see AR, Tab 289,
   Flight Manual KC-135 (Tanker) Flight Crew Air Refueling Procedures, Supp.
   III, T.O. 1-1C-1-3, Jan. 1, 1987, as revised Sept. 1, 2004, at [Deleted],
   and Northrop Grumman's EN response indicates that the KC-30's airspeed is
   limited to [Deleted] KIAS with the aircraft's [Deleted]. See AR, Tab 184,
   Northrop Grumman Response to EN NPG-MC1-003a, at 9. Boeing contends,
   citing the statement of its former tanker/receiver pilot consultant, that
   there is insufficient margin between airspeed at which [Deleted] are
   refueled and the KC-30's operational airspeed limit during refueling (a
   [Deleted]-KIAS margin) to allow for emergency breakaway maneuvers. See
   Boeing's Comments, attach. 14, Declaration of Retired Air Force Pilot, at
   3-4.

   As was true with respect to whether the KC-30 can satisfy the current Air
   Force procedures with respect to overrun airspeed, there is no
   documentation in the record setting forth an analysis of whether Northrop
   Grumman's proposed aircraft has sufficient operational airspeed when
   refueling the [Deleted] to initiate an emergency breakaway procedure. The
   agency's counsel provided a proffer at the hearing that the SSET's
   analysis of whether the KC-30 was capable of performing a breakaway
   maneuver with the [Deleted] was contained in the SSET's Final Evaluation
   Summary Report for Northrop Grumman. See HT at 784; see AR, Tab 215,
   Evaluation Summary Report for Northrop Grumman, at 3. Neither the page
   referenced by agency counsel or any other part of that document contains
   any analysis of whether Northrop Grumman's proposed aircraft can perform a
   breakaway procedure while refueling the [Deleted]; rather, the page
   referenced by agency counsel merely states that "[t]he Offeror has
   substantiated the ability to deliver a KC-X aircraft that meets (minimum
   requirement) all KPP thresholds associated with aerial refueling," and
   provides no reasons or analysis supporting this conclusion. AR, Tab 215,
   Evaluation Summary Report for Northrop Grumman, at 3.

   Although the SSET mission capability factor team chief was examined
   extensively about the SSET's consideration of the KC-30's ability to
   perform breakaway procedures, he recalled little about the SSET's
   discussions in this regard. His testimony does indicate, however, that the
   SSET accepted that the KC-30's maximum operational airspeed when refueling
   ([Deleted]) was [Deleted] KIAS, and that the SSET apparently believed
   that, to initiate the emergency breakaway procedure, with Northrop
   Grumman's proposed aircraft, the tanker would have to start accelerating
   and [Deleted] simultaneously. See HT at 706. During cross examination, the
   SSET mission capability factor team chief admitted that he did not know
   how long it would take [Deleted] Northrop Grumman's proposed [Deleted] or
   what the procedure was for [Deleted], nor was he aware of whether this was
   ever analyzed by the agency in its evaluation.[70] HT at 685-87, 707.

   In sum, we conclude that the record does not demonstrate that the agency
   reasonably determined that Northrop Grumman's proposed aircraft would be
   able to refuel all current Air Force fixed-wing tanker-compatible receiver
   aircraft in accordance with current Air Force procedures as was required
   by this KPP No. 1 threshold.

   Operational Utility Area

   Boeing also complains that the Air Force unreasonably evaluated the firms'
   proposals in the operational utility area under the key system
   requirements subfactor. The RFP provided that evaluation of this area
   would consist of an assessment of the offeror's approach to meeting (or
   exceeding, where appropriate) SRD requirements, "including the following:
   aircraft maneuverability, worldwide airspace operations,
   communication/information systems (including Net-Ready capability), treaty
   compliance support, formation flight, intercontinental range, 7,000-foot
   runway operations, bare base airfield operations, and growth provisions
   for upgrades." RFP sect. M.2.2.1.2.c. Boeing contends that its proposal
   should have been found technically superior to Northrop Grumman's in this
   area, and not essentially equal, as the SSA and SSAC concluded. See AR,
   Tab 54, Source Selection Decision, at 8.

   As noted above, the SSET found that both offerors satisfied the three KPP
   thresholds identified in this area, and partially met the one KPP
   objective identified; the SSET also found that both offerors met all
   associated KSA thresholds and objectives. See AR, Tab 55, PAR, at 20-21.
   The SSAC also identified two "major discriminators" in each of the firms'
   respective proposals; the discriminators for Boeing were the firm's
   (1) [Deleted] and (2) [Deleted], and for Northrop Grumman were (1) the
   ability of the KC-30 to operate from a 7,000-foot runway carrying more
   fuel than the KC-767, and (2) the KC-30's longer ferry range compared to
   the KC-767's. Id. Boeing's two "major discriminators" were assessed under
   17 different SRD requirements, while Northrop Grumman's two discriminators
   were assessed under only 2 SRD requirements. The SSAC also identified a
   number of "discriminators offering less benefit" for each firm: six such
   discriminators for Boeing assessed under 19 SRD requirements, and five
   such discriminators for Northrop Grumman assessed under 6 SRD
   requirements. Id. at 22-24.

   Here, too, as we described above with respect to the aerial refueling
   area, the record does not evidence that the SSAC and SSA, in determining
   that the firms' proposals were essentially equal in the operational
   utility area, gave any consideration to the fact that Boeing's proposal
   was evaluated as satisfying more SRD requirements than Northrop Grumman's
   in this area, as was sought by the RFP. Given this failure by the SSAC and
   SSA to address Boeing's apparent advantage in meeting more SRD
   requirements than Northrop Grumman, we conclude that the agency's
   evaluation and selection decision was unreasonable in this regard.

   Boeing also complains that the agency conducted misleading discussions
   with Boeing with respect to whether Boeing had fully satisfied the KPP No.
   7 objective, Net-Ready Capability. RFP, SRD sect. 3.2.4.1.1; app. A,
   Net-Ready Capability KPP, at 3. The KPP No. 7 objective provides that the
   offeror's "system should be capable of accomplishing all operational
   activities identified in Table 5." RFP, SRD, app. A, at 4. Table 5 of the
   appendix identified a number of information exchange requirements. Id.
   at 15-25.

   Specifically, Boeing complains that at its mid-term briefing it was
   informed of an uncertainty regarding the firm's net ready capability, see
   AR, Tab 129, Mid-Term Briefing to Boeing, at 77, and that ultimately the
   firm responded to an EN concerning the firm's System Requirements Matrix
   and System Specification with respect to complying with the SRD
   requirements for KPP No. 7. See AR, Tab 210, Boeing Response to EN
   BOE-MC1-041. Boeing believed that its EN response charted how its proposal
   met the KPP No. 7 thresholds and objective in total, see Boeing's Comments
   at 29, and during the firm's Pre-Final Proposal Revision Briefing the Air
   Force informed Boeing that the firm "met" both the KPP thresholds and the
   objective requirements for KPP No. 7. See AR, Tab 135, Boeing's Pre-Final
   Proposal Revision Briefing, at 57. Accordingly, Boeing made no further
   revisions to its proposal in this area. Boeing's Second Supplemental
   Protest at 53. The Air Force, however, changed its evaluation rating of
   this aspect of Boeing's proposal to "partially met" the KPP objective (the
   same rating that Northrop Grumman received) without further notice to
   Boeing.[71] Boeing contends that the Air Force's misleading discussions
   prevented the firm from addressing the agency's concerns with respect to
   this objective.

   The Air Force does not dispute that it informed Boeing during discussions
   that the firm had satisfied all of the thresholds and the objective under
   KPP No. 7, but contends that at the time it later determined that Boeing
   had not fully satisfied this objective, discussions had already been
   closed. See Second Supplemental COS at 77. The agency argues that, in any
   event, it was under no obligation to inform Boeing of the changed
   evaluation rating associated with this objective because the objective
   "constituted trade space," the absence of which would not be a deficiency
   or weakness. Agency Memorandum of Law at 131.

   We do not agree with the Air Force that the agency was permitted, after
   informing Boeing that its proposal fully met this objective, to change
   this evaluation conclusion without affording Boeing the opportunity to
   satisfy this requirement. It is a fundamental precept of negotiated
   procurements that discussions, when conducted, must be meaningful,
   equitable, and not misleading. See 10 U.S.C. sect. 2305(b)(4)(A)(i); AT&T
   Corp., B-299542.3, B-299542.4, Nov. 16, 2007, 2008 CPD para. 65 at 6.
   Here, by informing Boeing prior to the submission of the firm's final
   proposal revision that it satisfied all aspects of KPP No. 7, the Air
   Force deprived the firm of the opportunity to further address these
   particular requirements. See AT&T Corp., supra, at 12; see also Bank of
   Am., B-287608, B-287608.2, July 26, 2001, 2001 CPD para. 137 at 13.

   In contrast, the Air Force informed Northrop Grumman prior to the
   submission of that firm's final proposal revision that it had only
   partially met this KPP objective, which permitted that firm the
   opportunity to further address the KPP objective requirements. See AR, Tab
   205, Northrop Grumman's Pre-Final Proposal Revision Briefing, at 61.
   Moreover, Boeing submitted its final submission addressing this KPP
   objective several months prior to the pre-FPR briefing, and, as indicated
   above, the agency actually reopened discussions on other subjects after
   submission of the FPRs and obtained revised FPRs. Boeing's Protest at 66;
   Boeing's Second Supplemental Protest at 53. In short, the Air Force misled
   Boeing when the agency advised the firm that it met this objective, but
   later determined that Boeing did not fully meet this objective, and did
   not reopen discussions with Boeing on this issue. The Air Force also
   treated the firms unequally when it provided Northrop Grumman, but not
   Boeing, with continued discussions on this same objective. It is axiomatic
   that procuring agencies may not conduct discussions in a manner that
   favors one offeror over another. See FAR sect. 15.306(e)((1); Chemonics
   Int'l, Inc., B-282555, July 23, 1999, 99-2 CPD para. 61 at 8-9.

   We also find a reasonable possibility that Boeing was prejudiced by the
   Air Force's misleading and unequal discussions, given the greater weight
   that KPPs were supposed to receive in the agency's evaluation. In this
   regard, if Boeing had been evaluated as fully satisfying this KPP
   objective, which was the only KPP objective in the operations utility
   area, it could well have been considered to be superior in this area to
   Northrop Grumman, which was evaluated as only partially satisfying this
   KPP objective.

   Other Key System Requirements Subfactor Issues

   Boeing also protests the Air Force's conclusion in the aerial refueling
   area that Northrop Grumman's proposed larger boom envelope (relative to
   that offered by Boeing) offered a meaningful benefit to the Air Force. See
   AR, Tab 55, PAR, at 14. From our review of the record, including hearing
   testimony on this issue, we do not find a basis to object to the Air
   Force's judgment that Northrop Grumman had offered a larger boom envelope
   and that this offer provided a measurable benefit.[72]

   Boeing also challenges the Air Force's evaluation judgment in the airlift
   area that Northrop Grumman's proposed aircraft offered superior cargo,
   passenger, and aeromedical evacuation capability than did Boeing's
   aircraft. From our review of the record, including the hearing testimony,
   we see no basis to conclude that the Air Force's evaluation that Northrop
   Grumman's aircraft was more advantageous in the airlift area is
   unreasonable.

   Product Support Subfactor Evaluation

   Boeing also complains that the Air Force misevaluated Northrop Grumman's
   proposal under the product support subfactor. This subfactor required the
   agency to evaluate the "offeror's proposed product support approach for an
   efficient, effective and comprehensive support program for the service
   life of the KC-X fleet." RFP sect. M.2.2.3. Specifically, Boeing contends
   that the Air Force improperly ignored Northrop Grumman's refusal to commit
   to providing the required support necessary to allow the agency to achieve
   initial organic depot-level maintenance capability within the time
   required by the RFP, namely, within 2 years after delivery of the first
   full-rate production aircraft.[73] Boeing's Post-Hearing Comments at 84.
   The Air Force evaluated Boeing's and Northrop Grumman's proposals to be
   essentially equal under the product support subfactor.[74] See AR, Tab 54,
   Source Selection Decision Document, at 10; Tab 55, PAR, at 34.

   Offerors were informed that the long-term support concept for the KC-X
   program was for two levels of organic maintenance: organization level and
   depot level, and that a program objective was a product support approach
   that effectively addressed all the integrated support elements, including
   "[t]imely, cost effective transition to organic support." RFP, SOO for
   KC-X SDD, at 1-2. One of the specific minimum program tasks required by
   the SOO with regard to "logistics" was for the contractor to

     [p]lan for and support the Government to achieve an initial organic
     [depot]-level maintenance capability in accordance with the [Source of
     Supply Assignment Process] for core-designated workloads, at a minimum,
     within two years after delivery of the first full-rate production
     aircraft.

   Id. at 14; see also RFP, SOO for KC-X LRIP and Full-Rate Production, at
   1.[75] The RFP instructed offerors to ensure that their proposed
   contractual statements of work (SOW) would "conform to the Government's
   SOO" and that "[t[he proposed SOWs shall define the tasks required for the
   KC-X program, ensuring all minimum requirements of the Government provided
   SOOs and preliminary [work breakdown structure] have been addressed." See
   RFP sections L.2.1, L.8.3.7.2.

   The Air Force recognized in its evaluation that, although Northrop Grumman
   promised to provide the necessary planning and support for the agency to
   achieve an initial depot-level maintenance capability, the firm did not
   commit to providing this required support within 2 years after delivery of
   the first full-rate production aircraft, as required by the RFP. Thus, at
   the mid-term briefing, Northrop Grumman was informed that the timing of
   the firm's proposed depot level maintenance support was "unclear," see AR,
   Tab 199, Northrop Grumman's Mid-Term Briefing, at 134, and then again at
   the pre-final proposal briefing, Northrop Grumman was informed that the
   agency had assigned it a weakness for its failure "to include the time
   frame for initial organic depot standup in Offeror's Production SOW (SOO
   states within two years after delivery of the first full-rate production
   aircraft)."[76] See AR, Tab 205, Northrop Grumman's Pre-Final Proposal
   Revision Briefing, at 141. Northrop Grumman did not resolve its failure to
   commit to the 2-year timeframe for this product support requirement during
   the procurement. In the firm's final proposal revision, Northrop Grumman
   stated in one place that resolution of this "timing issue will be
   determined in coordination with the Government at contract award" and, in
   another place, that action to "resolve government identified weaknesses"
   would occur "after contract award." See AR, Tab 187, Northrop Grumman's
   Final Proposal Revision, KC-X Program Summary Document, at 2-3.

   In its final evaluation, the SSET evaluated Northrop Grumman's refusal to
   commit to providing these product support services within the 2-year
   timeframe as a weakness. AR, Tab 46, SSET Final Briefing to SSAC and SSA,
   at 360, 362. The SSAC concluded that this was an "administrative
   documentation oversight" because Northrop Grumman had promised to provide
   the required services and its "cost/schedule documentation is consistent
   with standing up depot capability within two years of delivery of the
   first full-rate production aircraft." AR, Tab 55, PAR, at 34. The SSA
   concurred with the SSAC that this was "merely an administrative
   oversight." AR, Tab 54, Source Selection Decision, at 10.

   We agree with Boeing that Northrop Grumman's refusal to commit to the
   required 2-year timeframe within which to provide these depot-level
   maintenance planning and support services cannot be reasonably viewed as
   an administrative or documentation oversight. As noted above, Northrop
   Grumman was clearly informed several times by the Air Force of the
   agency's concern that the firm had not committed to the required
   timeframe, and Northrop Grumman responded that it was not resolving this
   failure before award. Although throughout the protest and during the
   hearing, the agency steadfastly asserted that Northrop Grumman's failure
   to so commit was an "oversight,"[77] see, e.g., Air Force's Memorandum of
   Law at 151-53, in its post-hearing rebuttal comments, the agency admitted
   for the first time that Northrop Grumman's "omission" appeared to be a
   conscious decision. See Air Force's Post-Hearing Rebuttal Comments at 9.
   Northrop Grumman also finally admits in its rebuttal comments that its
   decision to not commit to the 2-year timeframe was "intentional."[78]
   Northrop Grumman's Post-Hearing Rebuttal Comments at 29 n.13.

   The Air Force and Northrop Grumman argue, however, that, apart from
   Northrop Grumman's refusal to commit to the 2-year timeframe, Northrop
   Grumman committed generally and specifically to performing the planning
   and support services solicited by the RFP in its proposal and proposal
   revisions, and that the firm would otherwise be obligated to perform the
   required services under whatever schedule the agency chooses. See, e.g.,
   Air Force's Post-Hearing Rebuttal Comments at 11; Northrop Grumman's
   Post-Hearing Rebuttal Comments at 29. The parties disagree as to whether
   Northrop Grumman's proposal demonstrates the ability to provide the
   required services within 2 years of delivery of the first full-rate
   production aircraft, and based on our review of Northrop Grumman's
   proposal and revisions, we find that it is far from clear whether or not
   Northrop Grumman's proposed schedule establishes that it would perform
   these services within the 2-year time frame.

   Whether or not Northrop Grumman's proposed schedule accommodates providing
   these product-support services within the 2-year timeframe misses the
   point, however. By explicitly refusing to contractually commit to the
   2-year timeframe for providing these services in the SOW as it was
   repeatedly requested to do, we think that Northrop Grumman has taken
   exception to this solicitation requirement. See C-Cubed Corp., B-272525,
   Oct. 21, 1996, 96-2 CPD para. 150 at 3. It is a fundamental principle in a
   negotiated procurement that a proposal that fails to conform to a material
   solicitation requirement is technically unacceptable and cannot form the
   basis for award. See TYBRIN Corp., B-298364.6; B-298364.7, Mar. 13, 2007,
   2007 CPD para. 51 at 5.

   The Air Force and Northrop Grumman also argue that the 2-year requirement
   is not a material solicitation provision. However, their arguments in this
   regard are belied by the agency's contemporaneous actions during the
   procurement and the testimony of the SSET product support subfactor team
   chief. As noted above, the agency repeatedly raised this matter with
   Northrop Grumman during discussions in an unsuccessful effort to have the
   firm commit to this solicitation requirement, and Northrop Grumman just as
   steadfastly refused to commit. Moreover, the SSET product support
   subfactor team chief identified the purpose or intent of this particular
   SOO requirement as follows: "It was a binding function to bind it to a
   specific time line," see HT at 1216, and that this 2-year requirement was
   "an important requirement." HT at 1245. We find, from our review of the
   record, that the requirement to plan for and support the agency's
   achieving an initial organic depot-level maintenance capability within 2
   years after delivery of the first full-rate production aircraft was a
   material requirement.

   In sum, the Air Force improperly accepted Northrop Grumman's proposal,
   where that proposal clearly took exception to a material solicitation
   requirement.[79]

   System Integration and Software Subfactor Evaluation

   Boeing also complains that, although both firms were evaluated as
   acceptable but with a moderate risk under the system integration and
   software subfactor, the Air Force should have viewed Northrop Grumman's
   proposal as riskier than Boeing's. See Boeing's Comments at 100-01. The
   Air Force states that it viewed both firms' offers of substantial software
   reuse to be risky, and this, with other weaknesses the agency noted in
   each firm's proposal under this subfactor, resulted in an assignment of a
   moderate risk. We see no basis in this record to object to the agency's
   evaluation under this subfactor.

   Program Management Subfactor Evaluation

   Boeing also complains that the Air Force did not reasonably assess
   schedule or cost risks under the program management subfactor with respect
   to Northrop Grumman's proposed changes during contract performance in its
   production approach and production lines. See Boeing's Comments at 75-96.
   The Air Force contends that Northrop Grumman agreed to appropriate
   mitigation measures that supported the agency's conclusion that the firm
   presented low cost or schedule risk under the program management
   subfactor. From our review of the record, including hearing testimony on
   this issue, we do not find a basis to object to the Air Force's evaluation
   of Northrop Grumman's proposal under this subfactor.

   Past Performance Factor Evaluation

   Boeing also challenges the Air Force's evaluation of Boeing's and Northrop
   Grumman's past performance, arguing that the agency's assessment of the
   relevance of contracts to be considered was unreasonable, that the agency
   treated the offerors disparately, and that the past performance evaluation
   judgments were not adequately documented. See Boeing's Comments at 148. We
   find from our review of the record no basis to object to the Air Force's
   past performance evaluation, under which both firms' past performance
   received a satisfactory confidence rating. We also find no basis to
   question the SSA's judgment that, despite the equal confidence ratings
   that the firms received under this factor overall, Northrop Grumman's
   higher "satisfactory confidence" rating, as compared to Boeing's "little
   confidence" rating, under the program management area was a reasonable
   discriminator. The Air Force evaluated Boeing's past performance as
   marginal in this area based on the agency's judgments as to Boeing's
   program management performance under the [Deleted] contract, the [Deleted]
   contract, and the [Deleted] contract. We have no basis, on this record, to
   find the Air Force's judgment unreasonable.

   IFARA Factor Evaluation

   Boeing also challenges the Air Force's evaluation of the firms' proposals
   under the IFARA evaluation factor. Boeing complains that the Air Force
   unreasonably concluded that Northrop Grumman's proposed aircraft was
   superior to Boeing's under this factor based only upon the fleet
   effectiveness value and without considering evaluated major insights and
   observations, which Boeing asserts favored its proposal. See Boeing's
   Comments at 146. Our review of the record discloses that the SSAC and SSA
   did consider the agency's evaluated insights and observations in their
   evaluation of the firms' proposals under this factor, and therefore find
   no basis to object to the agency's evaluation.

   Evaluation of MILCON Costs

   Boeing also complains that the Air Force did not reasonably evaluate the
   firms' cost/price proposals in accordance with the RFP. As noted above,
   the solicitation provided that the Air Force would calculate an MPLCC
   estimate for each offeror, which reflected the agency's independent
   estimate of all contract, budgetary, and other government costs associated
   with all phases of the aircraft's life cycle from SDD through production
   and deployment and O&S MILCON costs were specifically identified as a
   cost that the agency would evaluate in calculating the firms' MPLCCs. See
   RFP sect. M.2.5.2. Boeing contends that the Air Force's evaluation of
   MILCON costs greatly understated the difference between the firms' MILCON
   costs and that Northrop Grumman's much larger and heavier aircraft would
   have correspondently higher MILCON costs. See Boeing's Comments at 110-18;
   Boeing's Post-Hearing Comments at 117-18.

   The Air Force disputes Boeing's complaint, contending that it reasonably
   assessed the likely life cycle costs associated with each firm's proposed
   aircraft. In this regard, the agency states that, because it did not know
   at which bases ("beddown sites") the new KC-X aircraft would be assigned,
   it conducted site surveys at four airbases ([Deleted] Air Force Base
   (AFB), [Deleted] AFB, [Deleted] AFB, and [Deleted] AFB) to determine what
   military construction would be required at those bases for the offerors'
   proposed aircraft. The agency then extrapolated those results to six other
   airbases to calculate the agency's MILCON costs for the offerors. Air
   Force's Memorandum of Law at 221-22; Air Force's Post-Hearing Comments
   at 120-22. As indicated above, the agency added $[Deleted] billion in
   MILCON costs to Boeing's MPLCC and $[Deleted] billion in MILCON costs to
   Northrop Grumman's MPLCC. AR, Tab 55, PAR, at 40-43.

   An agency's life cycle cost evaluation, like other cost analyses, requires
   the exercise of informed judgment concerning the extent to which proposed
   costs or prices represent a reasonable estimation of future costs. Our
   review of the agency's cost analysis is limited to the determination of
   whether the evaluation was reasonable and consistent with the terms of the
   RFP. See Cessna Aircraft Co., B-261953.5, Feb. 5, 1996, 96-1 CPD para. 132
   at 21. The agency's analysis need not achieve scientific certainty;
   rather, the methodology employed must be reasonably adequate to provide
   some measure of confidence that the agency's conclusions about the most
   probable costs under an offeror's proposal are realistic in view of other
   cost information reasonably available to the agency at the time of its
   evaluation. See Information Ventures, Inc., B-297276.2 et al., Mar. 1,
   2006, 2006 CPD para. 45 at 7.

   As a threshold matter, the Air Force admits that in "defending this
   protest" it discovered five errors in its assessment of MILCON costs,
   which, when corrected, would result in Boeing displacing Northrop Grumman
   as the offeror with the lowest evaluated MPLCC. Specifically, the Air
   Force states that it underestimated Northrop Grumman's MILCON costs by
   $122.5 million, and overestimated Boeing's costs by $3.3 million. After
   correction of these $125.8 million in errors, Boeing's MPLCC would be
   $108.041 billion and Northrop Grumman's would be $108.133 billion.[80] Air
   Force's Memorandum of Law at 201-02.

   Here, the record shows that the agency's MILCON cost evaluation was
   otherwise flawed. In this regard, the RFP contemplated that the agency's
   MILCON cost evaluation would be based upon "the offeror's proposed KC-X
   aircraft solution," see RFP sect. M.2.5.2.4, which is consistent with the
   rule that an agency must consider an offeror's proposed approach in
   estimating the likely costs associated with that offeror's proposal. See
   Hughes STX Corp., B-278466, Feb. 2, 1998, 98-1 CPD para. 52 at 8. The
   record shows, however, that the agency's evaluation of MILCON costs was
   based upon site surveys that were conducted prior to the receipt of
   proposals in response to the RFP. HT at 472-73, 1293; Air Force's
   Post-Hearing Comments at 120. Admittedly, the agency's site surveys were
   based upon the size and dimensions of the A330-200 and 767-200, the
   commercial aircraft from which the offerors' proposed KC-X aircraft were
   derived. See, e.g., AR, Tab 297, Site Survey Report for [Deleted] Air
   Force Base, at 3. However, it is equally clear that the Air Force could
   not and did not evaluate MILCON costs associated with some aspects of the
   offerors' proposed aircraft because the site surveys were conducted before
   the receipt of proposals, and no further evaluation of the additional
   MILCON costs for the improvements/changes necessary to support each of
   these particular aircraft was performed after the proposals were received.

   For example, although the Air Force recognizes that there will be a "need
   for seat storage" associated with the KC-X aircraft, the survey teams were
   unable to assess the likely MILCON costs associated with this need
   because, at the time of the surveys, the agency did not know the number of
   seats associated with the firms' respective aircraft.[81] See Air Force's
   Post-Hearing Comments at 127. Accordingly, at [Deleted] AFB, the team
   assumed that the offerors' aircraft had seating capacities similar to that
   of the KC-10 and, on this basis, concluded that the facilities at
   [Deleted] AFB were adequate. HT at 497. The KC-10, however, has only 75
   seats, which is far less than the [Deleted] seats carried by the KC-30 and
   less than the [Deleted] seats carried by the KC-767. Similarly, at
   [Deleted] AFB, the survey team assigned no MILCON costs associated with
   seat storage because it determined, without any actual knowledge of the
   number of seats the proposed aircraft would carry, that there would be
   adequate storage available. Air Force's Post-Hearing Comments at 128. At
   [Deleted] AFB, the survey team concluded that there would be insufficient
   storage space to accommodate the seats and that an additional storage
   facility would need to be constructed; the cost of this facility
   ($[Deleted] million) was estimated to be the same for both offerors
   because the team did not know how many seats the aircraft carried and
   therefore "assigned a seat requirement the same for both aircraft." HT
   at 499-500.

   As another example, the survey team at [Deleted] AFB noted that the
   battery shop at the base may not have enough capability to service the
   batteries for the KC-X aircraft, if the new aircraft used different
   batteries from the other aircraft (the KC-135 and C-17 aircraft) at the
   base. AR, Tab 297, Site Survey Report for [Deleted] AFB, at 13. The team
   assigned no cost for this concern:

     Battery concern was noted because we did not know what the requirements
     were for the two different batteries, since we did not know the battery
     type on the A330, does that mean you only need to be separated by
     certain amount of spaces in the facility? Could you put up a wall? Would
     you actually need a whole new facility? So we didn't have enough detail
     to know if we needed to build anything or if there was going to be no
     cost.

   HT at 506. Other hearing testimony indicated that Boeing's proposed
   aircraft uses the same batteries as the [Deleted] aircraft, but that
   Northrop Grumman's aircraft may not. HT at 546-47; see also Boeing's
   Comments at 116.

   Also unexplained in the contemporaneous record is the agency's failure to
   consider in its evaluation of MILCON costs the offerors' own estimates of
   likely MILCON at Fairchild AFB that were included in their proposals.
   Specifically, the RFP instructed offerors, as part of its response to the
   product support subfactor, to

     describe the offeror's approach to meet the government's 2-level
     maintenance requirements. This proposal shall lay out:

                                    * * * *

     KC-X facilities, infrastructure requirements and design criteria.

     Facilities required to support the first operational bed down location
     at (assume Fairchild AFB, WA), including requirements for space,
     utilities or special requirements (such as clean rooms, special storage,
     etc.) with sufficient detail to assess installation capabilities to
     support the KC-X. The offeror shall describe facilities recommended to
     support the KC-X aircraft. The offeror shall, at a minimum, address the
     square footage for parking, maintenance facilities, infrastructure
     (e.g., power requirements, compressed air, office requirements,
     storage), personnel, and support equipment required to operate two
     squadrons of 16 aircraft for Main Operating Base (MOB) 1 and MOB 2. MOB
     3, MOB 4 and MOB 5 will be determined at a later date.

   RFP sections L.4.2.4.4, L.4.2.4.4.5, L.4.2.4.4.6.

   Northrop Grumman informed the Air Force in its proposal that based upon a

     [Deleted].

   AR, Tab 167, Northrop Grumman's Pre-Final Proposal Revision, vol. II,
   Mission Capability/Proposal Risk, Book 2, at II-SF3-48. Northrop Grumman
   also informed the agency that, among other changes that would be needed,
   [Deleted] in an identified building on Fairchild AFB would require
   "[Deleted]." Id. at II-SF3-49.

   The Air Force argues that it was reasonable to ignore the offerors' views
   as to the sufficiency of the facilities at Fairchild AFB with respect to
   their proposed aircraft because this information was requested in the
   solicitation instructions for the product support subfactor, and the
   offerors were not informed that this information would be used in the
   agency's evaluation of MPLCCs.[82] See Air Force's Post-Hearing Rebuttal
   Comments at 22-23. The agency does not explain, however, for what purpose
   this information was requested if not to aid in its evaluation of the
   facilities that would be needed to support the KC-X aircraft at Fairchild
   AFB. Given that both offerors responded to this solicitation instruction,
   it is apparent that neither offeror was confused as to the purpose of this
   instruction, which plainly sought the offerors' views as to whether the
   facilities at Fairchild AFB were adequate for their respective aircraft.
   In short, we find no reasonable basis to ignore the information that both
   offerors provided with respect to the adequacy of, or need for changes to,
   facilities with respect to their proposed aircraft.

   We also find, as described below, that the record does not otherwise
   demonstrate the reasonableness of the Air Force's notional (hypothetical)
   methodology for assessing likely MILCON costs. A notional beddown plan was
   developed because the agency did not know where the KC-X aircraft would be
   assigned. Under this scheme, the KC-X aircraft would be assigned in
   specified numbers to a test base (Air Force Materiel Command), a training
   base (Air Education & Training Command), three major operating bases (Air
   Mobility Command) within the continental United States (CONUS), four air
   reserve command (ARC) bases, and two major operating bases outside the
   continental United States (OCONUS).[83] See AR, Tab 309, Notional KC-X
   Beddown Plan Memorandum, June 29, 2007. As noted above, to assess the
   MILCON costs associated with each offeror's aircraft, the agency conducted
   site surveys at [Deleted] AFB, [Deleted] AFB, and [Deleted] AFB (major
   operating bases) and at [Deleted] AFB (a training base). See Air Force's
   Memorandum of Law at 221-22. The agency then extrapolated the results of
   its [Deleted] AFB survey to six other bases (four unspecified air reserve
   command bases and two unspecified OCONUS major operating bases) to
   calculate the agency's MILCON costs for the offerors. With respect to the
   two OCONUS airbases, the agency added a 10-percent premium to the
   extrapolated costs. Air Force's Post-Hearing Comments at 122. The sole
   reason identified by the Air Force for selecting [Deleted] AFB as the base
   from which it would extrapolate costs to the four ARC airbases and two
   OCONUS major operating bases was that a roughly comparable number of
   aircraft would be assigned at each of these bases. See HT at 63,
   1299-1300; Air Force's Post-Hearing Comments at 122.

   Where, as here, anticipated requirements cannot be reasonably ascertained,
   an agency may establish a reasonable hypothetical, or notional, plan to
   provide for a common basis for evaluating costs. See, e.g., PWC Logistics
   Servs., Inc., B-299820, B-299820.3, Aug. 14, 2007, 2007 CPD para. 162 at
   11-15 Aalco Forwarding, Inc., et al., B-277241.15, Mar. 11, 1998, 98-1 CPD
   para. 87 at 11. But that said, we are unable to conclude on this record
   that the agency's extrapolation of the [Deleted] AFB MILCON costs to the
   ARC airbases provided a reasonable basis to evaluate these costs. In this
   regard, Boeing argues that [Deleted] AFB, as a former Strategic Air
   Command, bomber base, has "a great deal more infrastructure" than do ARC
   airbases and thus cannot be used as a reasonable forecast of potential
   MILCON costs, such as for pavement improvements for runways, ramps, and
   parking aprons, at other bases. See Boeing's Post-Hearing Comments at
   136-41.

   Although the Air Force dismisses Boeing's argument as being speculative
   and argues that many ARC airbases have substantial infrastructure, see Air
   Force's Post-Hearing Rebuttal Comments at 28, the agency has not produced
   any explanation for selecting [Deleted] AFB other than its similar
   squadron size, nor presented any evidence, either through testimony or by
   reference to documentation in the record, showing why the infrastructure
   at [Deleted] AFB would be similar enough to ARC airbases to find that the
   costs evaluated for [Deleted] AFB are a reasonable representation of the
   MILCON costs to be expected for ARC airbases.

   Similarly, no evidence has been presented by the Air Force to explain why
   the application of a 10-percent premium to the costs assessed for
   [Deleted] AFB provides a reasonable estimate of MILCON costs for the
   OCONUS major operating bases. In this regard, Boeing contends, with no
   rebuttal, that the OCONUS airbases would have different issues associated
   with MILCON costs, such as compliance with foreign labor laws and foreign
   exchange rates, and that overseas bases would have to accommodate parking
   for all of the assigned KC-X aircraft, as opposed to the 75 percent of
   assigned aircraft that was done for CONUS airbases, such as [Deleted] AFB.
   See Boeing's Post-Hearing Comments at 137. The only evidence supporting
   the 10-percent factor is the testimony of the agency's SSET cost/price
   factor team chief that it was "based on estimator judgment" of one of the
   cost/price factor evaluators. HT at 209. No contemporaneous written
   description or explanation of that judgment has been provided for the
   record, however.

   In sum, we do not find reasonable support in the record for the agency's
   evaluation of the MILCON costs.

   Evaluation of Boeing's Non-recurring Engineering Costs

   Boeing also protests the Air Force's MPLCC adjustment for Boeing's
   estimated non-recurring engineering costs in the SDD phase of the
   contract. The Air Force added $[Deleted] million to the MPLCC beyond the
   $[Deleted] billion for non-recurring engineering that Boeing estimated for
   the SDD phase. Boeing states that its proposal approach is to acquire the
   baseline 767-200 LRF aircraft (which Boeing asserts and the Air Force
   concedes is a commercial item) from its commercial division, BCA, under a
   fixed-price subcontract, and that its estimated non-recurring engineering
   costs are included in the subcontract's fixed price.[84] Boeing argues
   that the agency unreasonably did not accept Boeing's commercial data in
   support of its estimated non-recurring engineering costs, and that it was
   improper to add costs to its MPLCC, given that the non-recurring
   engineering costs are part of a fixed-price subcontract for a commercial
   item. See Boeing's Comments at 118-22.

   The Air Force responds that, despite repeated discussions with Boeing
   regarding the firm's need to substantiate its estimated non-recurring
   engineering costs, see, e.g., AR, Tab 116, EN BOE-CP-001, EN BOE-CP-023,
   Boeing did not adequately support its estimated non-recurring engineering
   costs, and that the agency therefore concluded that there was a moderate
   risk associated with Boeing's non-recurring engineering cost estimate,
   although the agency did not determine that Boeing's estimated
   non-recurring engineering costs were unrealistic. COS at 136-37; HT at
   111-12.

   The agency also decided that it was necessary to upwardly adjust the MPLCC
   by $[Deleted] million to reflect this risk. To calculate this amount, the
   agency used a "Monte Carlo" analysis,[85] and concluded that Boeing was
   likely to incur a 36-percent cost growth with respect to its non-recurring
   engineering costs during the SDD phase of the procurement, which the
   agency then adjusted to account for a cost sharing provision that Boeing
   had proposed. COS at 139-40; HT at 221. The agency also states that as a
   "crosscheck" it looked at Boeing's P-8A Poseidon Multi-Mission Maritime
   Aircraft contract with the Navy, under which the Air Force contends that
   Boeing had an overall [Deleted]-percent cost growth, and that this
   compared favorably with the overall 36-percent cost growth it forecast
   using its Monte Carlo model. Air Force's Memorandum of Law at 205-06.

   We find reasonable the agency's assignment of a moderate risk to Boeing's
   proposal because of its failure to adequately substantiate its SDD
   non-recurring engineering costs. As noted above, the RFP placed upon the
   offerors the responsibility for substantiating their cost estimates. See,
   e.g., RFP sections L.6.1.2, 6.4.7. Here, the Air Force found, reasonably
   we conclude, that despite repeated requests during discussions Boeing
   failed to substantiate its SDD non-recurring engineering cost estimate. In
   this regard, we disagree with Boeing that, even if its purchase of the
   baseline aircraft from its commercial division, is considered to be the
   purchase of a commercial item, this prohibited the Air Force from
   requesting substantiating cost information from Boeing. Although FAR
   sect. 15.403-1(b)(3) provides that a contracting officer should not
   request the submission of certified cost or pricing data when a commercial
   item is being procured, this does not limit the right of the agency to
   request other cost information to determine price reasonableness or
   realism. See FAR sect. 15.403-3(c). We also note that it is not clear that
   the subcontract between Boeing and its commercial division is a
   fixed-price subcontract, as Boeing asserts, given Boeing's response in
   discussions that indicated that the price would not be fixed until the
   aircraft's configuration specifications were established, which had not
   yet happened. See AR, Tab 119, Boeing Response to EN BOE-K-015, at 2-3;
   Tab 259, Subcontract between Boeing and BCA.

   Nevertheless, as discussed below, we conclude for a somewhat different
   reason that the Air Force's MPLCC adjustment of Boeing estimated
   non-recurring engineering costs for SDD was unreasonable. When an agency
   evaluates proposals for the award of a cost-reimbursement contract, an
   offeror's proposed estimated cost of contract performance is not
   considered controlling since, regardless of the costs proposed by an
   offeror, the government is bound to pay the contractor its actual and
   allowable costs. Earl Indus., LLC, B-309996, B-309996.4, Nov. 5, 2007,
   2007 CPD para. 203 at 8. As a result, a cost realism analysis is required
   to determine the extent to which an offeror's proposed costs represent the
   offeror's likely costs in performing the contract under the offeror's
   technical approach, assuming reasonable economy and efficiency. See FAR
   sections 15.305(a)(1), 15.404-1(d)(1). A cost realism analysis involves
   independently reviewing and evaluating specific elements of each offeror's
   cost estimate to determine whether the estimated proposed cost elements
   are realistic for the work to be performed, reflect a clear understanding
   of the requirements, and are consistent with the unique methods of
   performance and materials described in the offeror's proposal. FAR
   sect. 15.404-1(d)(1); Advanced Commc'n Sys., Inc., B-283650 et al., Dec.
   16, 1999, 2000 CPD para. 3 at 5. Based on the results of the cost realism
   analysis, an offeror's proposed costs should be adjusted "to realistic
   levels based on the results of the cost realism analysis." FAR
   sect. 15.404-1(d)(2)(ii).

   Here, the record shows that the Air Force made no determination that
   Boeing's estimated $[Deleted] billion for SDD non-recurring engineering
   costs was unrealistic. See Air Force's Post-Hearing Comments at 90-91. In
   this regard, the SSET cost/price factor team chief testified under cross
   examination as follows:

   Q: Yes. You're supposed to look at whether what -- what Boeing proposed
   for the [non-recurring engineering], for the fixed price [non-recurring
   engineering] was realistic for the work to be performed. . . .

   A: No.

                                    * * * *

   Q: But when you made your adjustment, for example, one of the things that
   I would expect you would do is you would try to make an adjustment to make
   it, looking at the third item there, consistent with unique methods of
   performance and materials described in the offeror's technical proposal.
   Did you make any adjustments consistent with the unique methods of
   performance and materials described in Boeing's proposal when you adjusted
   upward using this Rand study?

   A: We added cost risk.

   HT at 111-12.

   The Air Force and Northrop Grumman argue that section M.2.5.2.5 of the RFP
   provided for the quantification of "pure cost risk," and for including
   that quantified dollar amount in the agency's evaluated MPLCC. Air Force's
   Post-Hearing Comments at 87; Northrop Grumman's Post-Hearing Comments at
   27-28. We disagree. This solicitation section states in its entirety:

     Risk Adjustments. The Government will assess the technical, cost, and
     schedule risk for the entire most probable life cycle cost estimate
     based upon the offeror's proposed approach. The Government will perform
     a Schedule Risk Assessment (SRA) and quantify the schedule risk
     accordingly. The Government will also assess risks associated with
     technical content as identified in the evaluation of the Mission
     Capability factor/subfactors 1 through 4, and other pure cost risks as
     identified during the cost evaluation. The impact of technical,
     schedule, and/or cost risk will be quantified (dollarized), where
     applicable, and included in the MPLCC. Additionally, the Government
     reserves the right to adjust budgetary estimates for technical, cost,
     and schedule risk.

   RFP sect. M.2.5.2.5.

   We do not agree that this section allows the agency to upwardly adjust the
   cost element of an offeror's "probable" costs of performance where the
   agency does not conclude that the proposed cost element is unrealistic or
   not probable. Rather, we find that this section allows the agency to
   assess the risk associated with an offeror's probable costs and, "where
   applicable," to quantify that risk and add the quantified amount in the
   agency's evaluated MPLCC for an offeror.[86] The increase to the MPLCC is
   "applicable" where the agency concludes that the higher number is more
   probable or more realistic than the lower one. The Air Force's and
   Northrop Grumman's reading is also inconsistent with other sections of the
   RFP that provided that the Air Force would assess the realism of offerors'
   proposed costs in accordance with FAR sect. 15.404-1 and that the agency's
   evaluated MPLCCs would be the agency's estimates of the probable or likely
   life cycle costs associated with the offerors' aircraft. See RFP sections
   M.2.5.1.1, M.2.5.2. Such a reading is also inconsistent with FAR sect.
   15.404-1, which, as noted above, provides for adjusting an offeror's
   proposed costs "to realistic levels based on the results of the cost
   realism analysis." See FAR sect. 15.404-1(d)(2)(ii).

   Moreover, even assuming a cost realism adjustment would have been proper
   in this case, we do not find reasonable the agency's use here of its Monte
   Carlo simulation model. Although we have recognized that a Monte Carlo
   model can be a useful evaluation tool, see TRW, Inc., supra, at 5, the
   validity of a Monte Carlo simulation, like all cost estimation models,
   depends upon the quality of the data used in the simulation or model. See
   Cost Assessment Guide: Best Practices for Estimating and Managing Program
   Costs, supra, at 144. Here, the cost evaluators used three inputs, "best
   case, worst case, and most likely case," in the Monte Carlo simulation to
   provide for a triangular distribution. HT at 29. Those three inputs were:
   (1) no cost growth (the best case); (2) 28-percent cost growth, which was
   derived from a GAO report, AR, Tab 281, Defense Acquisitions: Major Weapon
   Systems Continue to Experience Cost and Schedule Problems under DoD's
   Revise Policy, GAO-06-368, April 2006, (the most likely case); and (3)
   58-percent cost growth, which was derived from a Rand Corporation study,
   AR, Tab 282, Historical Cost Growth of Completed Weapon System Programs,
   (RAND 2006), (the worst case). See COS at 139. These reports, however, are
   discussing weapon systems and cost growth at an overall program level, and
   the reported cost growth would likely be attributable to a number of
   factors, including program changes and delays. In any event, we fail to
   see how overall program cost growth is a reliable predictor of anticipated
   growth in a single cost element, such as non-recurring engineering costs,
   nor has the Air Force or Northrop Grumman provided any explanation as to
   why that should be so.[87]

   Cost Evaluation Errors Conclusion

   In sum, we find that the Air Force unreasonably evaluated the MILCON costs
   associated with the firms' proposed aircraft and unreasonably adjusted
   Boeing's estimated non-recurring engineering costs, without finding those
   costs to be unrealistic. The correction of these errors in the Air Force's
   cost evaluation result in Boeing's MPLCC being lower than that of Northrop
   Grumman's.[88]

   Other Cost Issues

   Boeing also challenges a number of other aspects of the Air Force's
   evaluation of its cost proposal, including the Air Force's addition of
   $[Deleted] billion to Boeing's proposed costs for budgetary aircraft (lots
   6 through 13) and the addition of $[Deleted] billion to reflect additional
   O&S repair costs. In addition, Boeing challenges a number of aspects of
   the Air Force's evaluation of Northrop Grumman's proposed costs, including
   that the agency did not evaluate the fuel costs associated with that
   firm's larger and heavier aircraft and the costs of upgrades (such as the
   [Deleted]) that may be added to Northrop Grumman's aircraft in the future.
   We find no basis from our review of the record to object to the agency's
   evaluation of these other aspects of the Air Force's evaluation of
   costs.[89]

   CONCLUSION AND RECOMMENDATION

   This decision should not be read to reflect a view as to the merits of the
   firms' respective aircraft. Judgments about which offeror will most
   successfully meet governmental needs are largely reserved for the
   procuring agencies, subject only to such statutory and regulatory
   requirements as full and open competition and fairness to potential
   offerors. Foundation Health Fed. Servs., Inc.; QualMed, Inc., B-254397.4
   et al., Dec. 20, 1993, 94-1 CPD para. 3 at 43. Here, we find, as described
   above, a number of errors in the Air Force's conduct of this procurement,
   including the failure to evaluate proposals in accordance with the RFP
   criteria and requirements and to conduct discussions in a fair and equal
   manner. But for these errors, we believe that Boeing would have had a
   substantial chance of being selected for award.[90] Accordingly, we
   sustain Boeing's protest of the Air Force's award of a contract to
   Northrop Grumman for the aerial refueling tankers.

   The protest is sustained.

   We recommend that the Air Force reopen discussions with the offerors,
   obtain revised proposals, re-evaluate the revised proposals, and make a
   new source selection decision, consistent with this decision. If the Air
   Force believes that the RFP, as reasonably interpreted, does not
   adequately state its needs, the agency should amend the solicitation prior
   to conducting further discussions with the offerors. If Boeing's proposal
   is selected for award, the Air Force should terminate the contract awarded
   to Northrop Grumman. We also recommend that Boeing be reimbursed the
   reasonable costs of filing and pursuing the protest, including reasonable
   attorneys' fees. 4 C.F.R. sect. 21.8(d)(1). Boeing should submit its claim
   for costs, detailing and certifying the time expended and costs incurred,
   with the contracting agency within 60 days after receipt of this decision.
   4 C.F.R. sect. 21.8(f)(1).

   Gary L. Kepplinger
   General Counsel

   ------------------------

   [1] This acquisition has been identified as a Major Defense Acquisition
   Program. See Agency Report (AR), Tab 8, Acquisition Strategy Report, at 1.

   [2] The record in this case, which the agency largely provided
   electronically to GAO and the private parties, is voluminous and complex,
   and some of the record is classified. Although we considered the
   classified information, it is not discussed in this decision.

   [3] The Air Force acquired 732 KC-135A aircraft between 1957 and 1965. In
   the 1980s, a number of KC-135A aircraft were upgraded to the KC-135E
   aircraft, and later other KC-135A aircraft were upgraded to the KC-135R
   aircraft. Currently, the Air Force has 85 KC-135E aircraft and 418 KC-135R
   aircraft in its fleet. The agency also has 75 newer KC-10A aircraft in its
   fleet. See Air Force Refueling: The KC-X Aircraft Acquisition Program, CRS
   Report for Congress, at 4-5.

   [4] LRIP is defined as:

     The first effort of the Production and Deployment (P&D) phase. The
     purpose of this effort is to establish an initial production base for
     the system, permit an orderly ramp-up sufficient to lead to a smooth
     transition to Full Rate Production (FRP), and to provide production
     representative articles for Initial Operational Test and Evaluation
     (IOT&E) and full-up live fire testing. This effort concludes with a Full
     Rate Production Decision Review (FRPDR) to authorize the Full Rate
     Production and Deployment (FRP&D) effort.

   Glossary of Defense Acquisition Acronyms & Terms, Defense Acquisition
   University, 12^th ed., July 2005, at B-96-97.

   [5] A drogue is a small windsock placed at the end of a flexible hose that
   trails from a tanker aircraft in flight in order to stabilize the hose and
   to provide a funnel for the receiver aircraft, which inserts a probe into
   the hose to receive fuel. See Aerial Refueling Methods: Flying Boom versus
   Hose-and-Drogue, CRS Report for Congress, No. RL32910, June 5, 2006, at 1.

   [6] For example, a "blue" rating reflected an exceptional proposal that

     [e]xceeds specified minimum performance or capability requirements in a
     way beneficial to the Government; proposal must have one or more
     strengths and no deficiencies to receive a blue.

   A "green" rating reflected an acceptable proposal that

     [m]eets specified minimum performance or capability requirements
     delineated in the [RFP]; proposal rated green must have no deficiencies
     but may have one or more strengths.

   RFP sect. M.2.2.

   [7] For example, a "low" risk rating reflected a proposal that

     [h]as little potential to cause disruption of schedule, increased cost
     or degradation of performance. Normal contractor effort and normal
     Government monitoring will likely be able to overcome any difficulties.

   A "moderate" risk rating reflected a proposal that

     [c]an potentially cause disruption of schedule, increased cost, or
     degradation of performance. Special contractor emphasis and close
     Government monitoring will likely be able to overcome difficulties.

   RFP sect. M.2.3.

   [8] For example, the graph indicated an aircraft must be capable of
   offloading 117,000 pounds of fuel at a radius of 500 nautical miles and
   94,000 pounds at a radius of 1,000 nautical miles. RFP, SRD
   sect. 3.2.1.1.1.1, Figure 3-1, Fuel Offload vs. Radius Range.

   [9] Although identified as "requirements" by the RFP, these non-KPP
   "requirements" were not mandatory, but reflect features and performance of
   the aircraft that the agency desired. There were thresholds and objectives
   identified for some of the KSAs and the other SRD requirements.

   [10] The Air Force described "trade space" as follows:

     [the RFP] also provided the offerors considerable "trade space," meaning
     some performance parameters of the tanker were required, while others
     were not. The optional capabilities or attributes could be traded away
     for better or different performance in other areas depending on the
     offeror's unique approach. . . . Essentially, this asked the offerors to
     tender their best proposals, and encouraged them to be creative in doing
     so. With such a structure, the RFP harnessed the power of the commercial
     marketplace competition to drive innovation as well as efficiency.

   Air Force's Memorandum of Law at 5.

   [11] The RFP provided that the PCAG would assign a confidence rating of
   high confidence, significant confidence, satisfactory confidence, unknown
   confidence, little confidence, or no confidence.

   [12] The 2005 Mobility Capabilities Study assessed the mobility
   capabilities of DoD against the backdrop of a revised National Security
   Strategy; the study was intended to support decisions on future strategic
   airlift, aerial refueling, aircraft, and sealift procurements needed to
   meet varying military requirements. See Defense Transportation: Study
   Limitations Raise Questions about the Adequacy and Completeness of the
   Mobility Capabilities Study and Report, GAO-06-938, at 6.

   [13] CMARPS is a system that is comprised of the Contingency Mating and
   Ranging Program, Tanker Mating and Ranging Program, and Graphically
   Supported Interactive Control System user interface. Using inputs, such as
   aircraft performance characteristics, and assumptions and ground rules
   (such as the maximum number of a particular aircraft that could be located
   at a particular base given "ramp geometrics and aircraft dimensions" and
   pavement strengths of ramps and runways), the agency would conduct
   simulations, or "runs," where a proposed tanker fleet attempts to satisfy
   tanker demand; the results of these simulations are intended to reflect
   the effectiveness of those runs. See Fourth Supplemental Contracting
   Officer's Statement (COS) at 9-12.

   [14] With regard to fuel costs, offerors were requested to provide a
   fuel-consumption "sample calculation" for an average mission-ready KC-X,
   including fuel, crew, and mission equipment on board, in gallons per hour
   per primary aircraft assigned multiplied by the number of flying hours in
   a given fiscal year. Offerors were required to document the source of the
   input data and rationale. RFP sect. L, attach. 15, KC-X O&S Data Form, at
   7.

   [15] Boeing stated in its proposal that the "767-200LRF is a new minor
   model (a family of variants as defined by the FAA such as 767-200,
   767-300F, or 767-400ER) that includes design features that satisfy KC-X
   requirements." AR, Tab 61, Boeing Executive Summary, at V1-ES-1.

   [16] Airbus is a division of European Aeronautic Defence & Space Company
   (EADS), Northrop Grumman's principal subcontractor for this procurement.
   After award, the Air Force changed the designation of Northrop Grumman's
   aircraft to the KC-45; throughout this decision, however, we refer to
   Northrop Grumman's aircraft by the firm's KC-30 designation.

   [17] Although not requested by the parties, we conducted a hearing to
   receive testimony from a number of Air Force witnesses to complete and
   explain the record. In this regard, we provided a detailed description of
   the hearing issues to the parties in a pre-hearing conference and in a
   written Confirmation of Hearing notice. We also expressly informed the
   parties that identification of some of the protest issues as hearing
   issues did not indicate GAO's views as to the merits of any issue in the
   case. The Air Force was informed that it was responsible for identifying
   and producing those witnesses who could knowledgeably testify with respect
   to the identified issues. Although invited to do so, neither Boeing nor
   Northrop Grumman offered any witnesses. At the conclusion of the hearing,
   the parties were informed that they could address any aspect of the
   protest in their post-hearing comments and rebuttal comments. HT at 1524.

   [18] The Air Force conducted numerous rounds of written and oral
   discussions with the firms; in total, Boeing received 271 ENs, and
   Northrop Grumman received 295 ENs. AR, Tab 46, SSET Final Briefing to SSAC
   and SSA, at 9-10.

   [19] Limited information was provided to the offerors in the mid-term
   briefing with respect to the agency's schedule risk assessment and its
   impact on the offeror's MPLCC. COS at 24.

   [20] Discussions were conducted with the offerors to address any possible
   impact on their proposals from section 804 of the National Defense
   Authorization Act for Fiscal Year 2008 (related to Buy American Act
   requirements with respect to specialty metals) and section 815 of the Act
   (related to treatment of major defense acquisition program systems,
   components, and spare parts as commercial items). Air Force's Memorandum
   of Law at 20 n.6.

   [21] A "major discriminator" was defined to be

     an offered feature evaluated as a strength that provided extensive
     capability and a substantial difference in magnitude of benefit to the
     Air Force, when compared to the other Offeror.

   A "discriminator offering less benefit" was defined to be

     an offered feature evaluated as a strength that provided some capability
     and some difference in benefit to the Air Force when compared to the
     other Offeror.

   AR, Tab 55, PAR, at 12.

   [22] The SSAC reported that the KC-30 met the objective by offering a fuel
   offload versus unrefueled range capability of [Deleted] pounds at 1,000
   nautical miles and [Deleted] pounds at 2,000 nautical miles, which
   exceeded the threshold by [Deleted] percent at 1,000 nautical miles and by
   [Deleted] percent at 2,000 nautical miles. The KC-767 was also found to
   meet the objective by offering a capability of [Deleted] pounds at
   1,000 nautical miles and [Deleted] pounds at 2,000 nautical miles, which
   exceeded the threshold by [Deleted] percent at 1,000 nautical miles and
   [Deleted] percent at 2,000 nautical miles. AR, Tab 55, PAR, at 13-14.

   [23] The ATP is an aerial refueling publication issued by the North
   Atlantic Treaty Organization.

   [24] Airlift efficiency was calculated using the following formula:
   (pounds of payload) x (nautical miles)/(pounds of fuel). The SSET
   performed this calculation at various distances for the offerors to derive
   a payload-range curve to provide for a comparative analysis. AR, Tab 55,
   PAR, at 17.

   [25] The 463L pallet is the standard air cargo pallet used by the Air
   Force and within the defense transportation system.

   [26] The SSAC noted, however, that the KC-30's total weight carriage
   capability on the main cargo deck was not substantially greater than that
   of the KC-767. AR, Tab 55, PAR, at 17.

   [27] The KPP No. 7 objective, the only objective under this KPP, provides
   that the offeror's "system should be capable of accomplishing all
   operational activities identified in Table 5." RFP app. A, Net-Ready Key
   Performance Parameter for the KC-X SRD, Feb. 23, 2007, at 3. Table 5 of
   the appendix identified a number of information exchange requirements. Id.
   at 15-25.

   [28] The capability to operate from a 7,000-foot runaway at sea level at
   the aircraft's maximum gross weight was a non-KPP/KSA trade space
   requirement, see RFP, SRD sect. 3.2.1.1.4.2, which both Boeing and
   Northrop Grumman satisfied. AR, Tab 55, PAR, at 21.

   [29] An unrefueled ferry range of 9,500 nautical miles starting at maximum
   takeoff gross weight and using a maximum range flight profile was
   identified as a non-KPP/KSA trade space requirement. RFP, SRD sect.
   3.2.1.1.1.4.

   [30] The SRD provided that the "KC-X shall be capable of supporting aerial
   refueling operations from bare base airfields with confined ramp space."
   RFP, SRD sect. 3.2.10.1.1.3.

   [31] The SSAC did recognize that Boeing's proposal was more advantageous
   than Northrop Grumman's in the survivability area. AR, Tab 55, PAR, at 29.

   [32] A "satisfactory confidence" rating was assigned where

     [b]ased on the offeror's performance record, the government has
     confidence the offeror will successfully perform the required effort.
     Normal contractor emphasis should preclude any problems.

   A "little confidence" rating was assigned where

     [b]ased on the offeror's performance record, substantial doubt exists
     that the offeror will successfully perform the required effort.

   RFP sect. M.2.4.1.

   [33] Much of the information detailing the agency's evaluation under the
   IFARA factor is classified.

   [34] The SSA concluded that the offerors' proposals were essentially equal
   under the remaining four mission capability subfactors. Regarding the
   evaluation of Northrop Grumman's proposal under the product support
   subfactor, although the Air Force found that Northrop Grumman had failed
   to specifically commit to providing planning and support for the "initial
   organic D-level [depot-level] maintenance capability" within 2 years
   following delivery of the first full-rate production aircraft as required
   by the RFP, see RFP, SOO for KSC-X SDD, at 14, the SSA agreed with the
   SSAC that this was "merely an administrative oversight." AR, Tab 54,
   Source Selection Decision, at 10. (The SSAC termed Northrop Grumman's
   failure in this regard to be "an administrative documentation oversight."
   AR, Tab 55, PAR, at 34.)

   [35] Where, as here, a procurement is conducted on the basis of
   competitive proposals, "an unsuccessful offeror, upon written request
   received by the agency within 3 days after the date on which the
   unsuccessful offeror receives the notification of the contract award,
   shall be debriefed and furnished the basis for the selection decision and
   award." 10 U.S.C. sect. 2305(b)(5)(A) (2000); Federal Acquisition
   Regulation (FAR) sect. 15.506(a)(1).

   [36] Although our regulations allow a procuring agency 30 days to provide
   relevant documents to the protester, see 4 C.F.R. sect. 21.3(d) (2008),
   the Air Force provided many relevant, core documents to Boeing and
   Northrop Grumman within days of the filing of the initial protest. The Air
   Force, however, continued to produce relevant documents even after the
   filing of its agency report and up to the date of the hearing conducted in
   this protest, which resulted in Boeing filing a series of supplemental
   protests.

   [37] Although we have not sustained all of Boeing's protest allegations,
   nor do we address them all in this decision, we considered them all, which
   required substantial development of the issues during the protest.

   [38] Our document production rules are much narrower than other federal
   discovery rules, such as the Federal Rules of Civil Procedure (FRCP),
   which permits litigants to seek the existence of documents that are
   reasonably calculated to lead to the discovery of admissible evidence.
   See, e.g., FRCP Rule 26(b)(1). In contrast, our regulations provide for
   the production of relevant documents. See 4 C.F.R. sect. 21.3(d).

   [39] Although we denied the Air Force's request for documents from Boeing,
   we also informed the agency that if, during the further development of the
   case, the agency became aware of a specific relevant document, or
   documents, that only the protester possesses, the agency was permitted to
   request that document or documents. No such request was made.

   [40] We will more fully address below certain of the agency's and
   intervenor's dismissal requests, such as the arguments concerning the
   evaluation of Northrop Grumman's proposal with respect to the fuel offload
   versus unrefueled radius range.

   [41] To require an offeror to file a protest each time during a
   procurement that it is advised of an evaluation judgment with which it
   disagrees or believes is inconsistent with the RFP would not be consistent
   with the regulatory requirement that such protests can only be filed after
   a required debriefing. The objective of this regulation is to avoid the
   filing of "defensive protests" out of fear that our Office may dismiss the
   protests as untimely and the associated potential to unnecessarily disrupt
   procurements.

   [42] As noted above, the RFP provided that KPP requirements were more
   important than KSA requirements, which were in turn more important than
   non-KPP/KSA requirements.

   [43] As discussed above, the RFP indicated that KPP thresholds were
   minimum, mandatory requirements that must be satisfied and that the
   remaining "requirements," including KPP objectives and KSA thresholds and
   objectives, were desired functions or characteristics that the firms could
   choose to offer.

   [44] The PAR combined the (1) and (2) major discriminators into one major
   discriminator, although they actually concern separate KPP No. 1
   objectives.

   [45] Although the Air Force credited Northrop Grumman with exceeding to a
   greater degree than Boeing the KPP No. 2 objective related to fuel offload
   versus unrefueled range, we find, as discussed below, that this was
   inappropriate, given that the RFP provided that no additional credit would
   be provided for exceeding KPP objectives. The evaluation record thus shows
   that, instead of this being a discriminator, Northrop Grumman and Boeing
   should have received equal credit for satisfying this KPP objective.

   [46] While we here illustrate this issue in our discussion of the aerial
   refueling area, as further illustrated below, the agency's failure to
   account for the relative weights given the various SRD requirements or
   consider the RFP's request that offerors propose to satisfy as many of the
   "trade space" requirements as possible permeates the evaluation of the key
   system requirements subfactor.

   [47] The Air Force recognizes that Boeing could have proposed an aerial
   refueling tanker based upon the larger 777 platform. See, e.g., Air
   Force's Memorandum of Law at 84 n.30.

   [48] Thus, the Air Force's and Northrop Grumman's interpretation of this
   objective as unbounded would render the last sentence of section
   M.2.2.1.1.a meaningless, given that that sentence addresses the situation
   where unbounded credit will be given for exceeding a KPP threshold.

   [49] The Air Force and Northrop Grumman argue that Boeing's interpretation
   is unreasonable because this would mean that meeting and exceeding the KPP
   threshold relating to fuel offload versus unrefueled radius range would
   necessarily approximate one another because both would be bounded by a
   single line on a chart, which would render the establishment of an
   objective meaningless. We do not agree. As Boeing notes, the establishment
   of KPP objectives was expressly for the purpose of limiting the KPP trade
   space available to offerors, and we find from our review of the entire
   record that Boeing had a reasonable basis for believing that this
   limitation--that no credit would be given for exceeding the KPP threshold
   amount--was "consistent with real-world tanker operations." See Boeing's
   Comments at 16.

   [50] Both the Air Force and Northrop Grumman cite our decision in PM
   Servs. Co., B-310762, Feb. 4, 2008, 2008 CPD para. 42, for the proposition
   that Boeing, having learned how its proposal was being evaluated with
   respect to this KPP objective, was required to protest before the next
   closing time for receipt of proposals. In that case, we therefore
   concluded, unlike here, that the protester was informed during discussions
   of the agency's interpretation of a solicitation provision that was
   otherwise clear on its face, and that the protester's later post-award
   challenge of this provision was an untimely protest of an apparent
   solicitation impropriety. Id. at 3.

   [51] The Air Force's mid-term briefings to Boeing and Northrop Grumman
   stated that "[o]nly SRD KPP Threshold requirements must be met --
   strengths may be awarded for greater capability, but not beyond Objective
   levels (if an Objective is stated)." See AR, Tab 129, Mid-term Briefing to
   Boeing, at 17; Tab 199, Mid-Term Briefing for Northrop Grumman, at 17. The
   parties' pre-final proposal revision briefings did not include this
   language.

   [52] Although the pre-award briefings provided to Boeing identified
   "benefits" associated with the firm's offer to exceed the fuel offload
   versus unrefueled radius range threshold, Boeing could not know until it
   received the source selection decision that the agency was actually
   providing additional credit for the degree to which the offerors were
   exceeding the fuel offload versus unrefueled radius range.

   [53] Even were we to consider the limitation on consideration above the
   KPP objective for the fuel offload versus unrefueled range requirement to
   be a latent ambiguity, Boeing's protest would still be timely. See Vitro
   Servs. Corp., B-233040, Feb. 9, 1989, 89-1 CPD para. 136 at 3 n.1 (protest
   filed within 10 days of the date the protester learned of an agency's
   interpretation of a latent solicitation ambiguity is timely).

   [54] The capability of Northrop Grumman's proposed aircraft to satisfy
   this KPP threshold concerns a matter of technical acceptability; stated
   differently, if Northrop Grumman could not establish the capability of its
   aircraft to refuel all current fixed-wing tanker-compatible fixed wing
   aircraft using current Air Force procedures, its proposal could not be
   accepted. HT at 625, 649.

   [55] In aerial refueling operations, tankers maneuver to a rendezvous
   point and establish an orbit pattern at a constant airspeed to await
   receiver aircraft. See, e.g., AR, Tab 289, Flight Manual, KC-10A Aircraft,
   Flight Crew Tanker Air Refueling Procedures, USAF Series, Technical Order
   (T.O.) 1-1C-1-33, Sept. 1, 2002, as revised Jan. 31, 2005, at 2-2, 2-15.
   If a receiver aircraft overruns the tanker during the final phase of
   rendezvous, the tanker and receiver pilots are directed to adjust to
   specified overrun speeds, and after overtaking the receiver aircraft, the
   tanker will decelerate to a refueling airspeed. Id. at 2-16. A breakaway
   maneuver is an emergency procedure that is done when any tanker or
   receiver aircraft crewmember perceives an unsafe condition that requires
   immediate separation of the aircraft. See id. at 6-2; see also HT at 619.
   In such a situation, the tanker pilot is directed to accelerate in level
   flight to achieve separation, or, if required, to accelerate and climb
   (during which the tanker pilot is directed to "not allow the airspeed to
   decrease below that indicated at the start of climb.") See, e.g., AR, Tab
   289, Flight Manual, KC-10A Aircraft, Flight Crew Tanker Air Refueling
   Procedures, USAF Series, T.O. 1-1C-1-33, Sept. 1, 2002, as revised Jan.
   31, 2005, at 6-2.

   [56] In the first EN to Northrop Grumman addressing that firm's aircraft
   overrun capability, the Air Force identified [Deleted] KIAS, as the
   required overrun speed for the [Deleted]. See AR, Tab 184, EN NPG-MC1-003,
   at 2. In the second EN to Northrop Grumman, the agency corrected this to
   [Deleted] KIAS, see id., EN NPG-MC1-003a, at 1, which reflects the overrun
   speed identified for the [Deleted] in the KC-135 flight manual. See Tab
   289, Flight Manual KC-135 (Tanker) Flight Crew Air Refueling Procedures,
   Supp. III, T.O. 1-1C-1-3, Jan. 1, 1987, at [Deleted].

   [57] Initially, Northrop Grumman informed the Air Force that the agency
   should change its current overrun procedures. See AR, Tab 184, Northrop
   Grumman Response to EN NPG-MC1-003, at 1-3. Thereafter, Northrop Grumman
   asserted that there was nothing in the RFP requirements that established
   airspeed limitations for specific aircraft in overrun situations. Id.,
   Northrop Grumman Response to EN NPG-MC1-003a, at 1-2. As noted by the Air
   Force in the second EN provided to Northrop Grumman on this issue, the
   agency's current procedures are established by its flight manuals for the
   KC-135 and KC-10 that provide operational airspeed and overrun airspeed
   requirements specific for each receiver aircraft type. See id.,
   EN NPG-MC1-003a, at 1; e.g., Tab 289, Flight Manual KC-135 (Tanker) Flight
   Crew Air Refueling Procedures, Supp. III, T.O. 1-1C-1-3, Jan. 1, 1987, at
   [Deleted]; see also HT at 622.

   [58] The Air Force and Northrop Grumman contend that Boeing's
   contention--that Northrop Grumman's [Deleted] causes [Deleted]--was
   untimely raised in Boeing's comments and must be dismissed. Air Force's
   Post-Hearing Comments at 18; Northrop Grumman's Post-Hearing Comments at
   141. We disagree. In its second supplemental protest (filed within 10 days
   of receipt of the first production of documents), Boeing specifically
   challenged the Air Force's evaluation of Northrop Grumman's proposal to
   [Deleted], arguing that the "Air Force never considered the feasibility of
   this extreme measure or its implications on the KC-30's ability to carry
   out the refueling mission." See Boeing's Second Supplemental Protest
   at 29-31. The arguments concerning the [Deleted] are thus within the scope
   of Boeing's timely protest.

   [59] As noted above, our Office requested that the Air Force provide
   knowledgeable witnesses who could testify with respect to the previously
   identified hearing issues.

   [60] FAA's regulations provide that the design dive speed of an aircraft
   be established so that the design cruise speed is no greater than 0.8
   times the design dive speed. See 14 C.F.R. sect. 25.335(b) (2008).

   [61] The record indicates that the evaluators did not consider during the
   procurement whether the [Deleted] during aerial refueling operations was a
   current Air Force procedure and how this may affect Northrop Grumman's
   proposed solution to satisfying the overrun speed requirements. Rather,
   these issues apparently were only considered in response to Boeing's
   protest allegations. See HT at 711.

   [62] The flight manuals for the KC-135 and KC-10 both state that the
   "[p]rocedures in this manual are mandatory and must be performed in the
   prescribed manner except where deviations are required in the interest of
   safety of flight." AR, Tab 289, Flight Manual KC-135 (Tanker) Flight Crew
   Air Refueling Procedures, Supp. III, T.O. 1-1C-1-3, Jan. 1, 1987, as
   revised Sept. 1, 2004, at i; Tab 289, Flight Manual, KC-10A Aircraft,
   Flight Crew Tanker Air Refueling Procedures, USAF Series, T.O. 1-1C-1-33,
   Sept. 1, 2002, as revised Jan. 31, 2005, at ii.

   [63] In its post-hearing comments, Northrop Grumman argues that [Deleted].
   See Northrop Grumman's Post-Hearing Comments at 141-43. We provide little
   weight to this post-procurement description of Northrop Grumman's proposed
   design, given that this argument seems inconsistent with Northrop
   Grumman's EN response and is not supported by statements of consultants or
   other knowledgeable sources, and it represents information that was not
   presented to the agency for its consideration during the procurement.

   [64] In this regard, in response to cross examination, the SSET mission
   capability team chief testified that, although Northrop Grumman in its EN
   response was not "very good at articulating what they were doing at the
   end there, okay," the evaluation team did not ask Northrop Grumman to
   clarify what it was proposing in its EN response. See HT at 664.

   [65] Northrop Grumman provided to the Air Force with its EN response a FAA
   Type Certificate for the Airbus A330-200 and A330-300 series aircraft,
   which identified the maximum operating limit airspeed as 330 KIAS and the
   design diving speed as 365 KIAS. AR, Tab 184, Northrop Grumman Response to
   EN NPG-MC1-003a, attach., FAA Type Certificate Data Sheet No. A46NM, Rev.
   10, Mar. 19, 2007, at II-J-72. In November 2007, the FAA-type certificate
   for the A330 aircraft was revised, but stated the same maximum operating
   limit and dive speeds. See Boeing's Airspeed Hearing exh. 13, FAA Type
   Certificate No. A46NM, Rev. 11, Nov. 13, 2007, at 12.

   [66] We have been presented with no testimony, statements or documentation
   from any member of the SSET professing to understand the relationship of
   maximum operational airspeed and dive velocity or airspeed, or to
   otherwise support the agency's conclusion that the A330's certified design
   dive velocity indicated that the aircraft was capable of achieving that
   speed as a maximum operational airspeed.

   [67] The SSET mission capability factor team chief also testified that
   Northrop Grumman's response indicated to the SSET that the KC-30 could
   achieve [Deleted] KIAS with both the [Deleted]. HT at 636. In this regard,
   Northrop Grumman's EN response contained a "Note" under case two
   ([Deleted]), which stated that the [Deleted]. AR, Tab 184, Northrop
   Grumman Response to EN NPG-MC1-003a, at 6. It is unexplained in what
   situation this occurs, given that the [Deleted] is supposed to [Deleted].
   Moreover, neither the Air Force nor Northrop Grumman has identified any
   evidence in the contemporaneous record of the agency's consideration of
   this note.

   [68] Although the Air Force argues that it considered whether there was
   any schedule or cost risk associated with the proposed changes to Northrop
   Grumman's aircraft to satisfy the airspeed requirements, see Air Force's
   Post-Hearing Comments, at 18, we have not been directed to documentation
   in the record establishing that such an analysis was performed. Instead,
   the Air Force relies upon the testimony of the SSET mission capability
   factor team chief that the SSET concluded that Northrop Grumman had
   provided "associated costs and schedule impact" for the firm's proposed
   approach to satisfying the airspeed requirements. See HT at 629. However,
   he was unable to point to anything in the record to support his testimony,
   except his statement that a structural engineer on the SSET reviewed
   Northrop Grumman's EN response and determined that any required changes to
   the proposed aircraft could be accomplished within Northrop Grumman's
   proposed schedule. HT at 721. The totality of that review by the
   structural engineer, however, was apparently captured in an e-mail sent
   during the evaluation. See HT at 757-59, 783 (proffer by Air Force
   counsel). This e-mail does not establish that the structural engineer
   validated the capability of Northrop Grumman's aircraft to satisfy the
   overrun airspeed requirements or that changes in the aircraft's maximum
   operational airspeed could be achieved within the offeror's proposed
   schedule or cost. Rather, the e-mail states that the effect of higher
   airspeed on the integrity of KC-30 aircraft structure has not been
   analyzed and that "[b]ottom line, these [Deleted] are major concerns that
   must be addressed by Analysis for sure and Flight Test if warranted." AR,
   Tab 332, E-mail 32002 re: EN NPG-MC1-003a, Jan. 25, 2008.

   [69] The Air Force did not contemporaneously express any concern to
   Northrop Grumman with respect to its aircraft's ability to achieve
   breakaway speeds. HT at 619.

   [70] Boeing argues, citing the procedures identified in Northrop Grumman's
   final proposal, that [Deleted] Northrop Grumman's [Deleted] could take as
   long as [Deleted], which would require the tanker pilot to either
   accelerate beyond [Deleted] KIAS before [Deleted] or delay initiating the
   breakaway until after [Deleted]; Boeing contends that the Air Force did
   not assess these concerns. See Boeing's Post-Hearing Comments at 68,
   citing, AR, Tab 187, Northrop Grumman's Final Proposal Revision, vol. II,
   Mission Capability/Proposal Risk, Jan. 4, 2008, at II-SF116-16a.

   [71] It is unclear from the record when the Air Force changed its
   evaluation of this KPP objective.

   [72] As set forth above, the agency also identified a weakness for
   Northrop Grumman in the aerial refueling area related to the firm's boom
   approach. Because the record did not contain any documentation explaining
   why the Air Force's evaluated concern with Northrop Grumman's proposed
   boom design represents a low risk as to schedule or cost, we also
   identified this as an area in which hearing testimony would be required to
   "explain why evaluated weaknesses in Northrop Grumman's boom have low
   schedule or cost risk." See GAO Confirmation of Hearing, Apr. 29, 2008,
   at 3. The Air Force produced its SSET team chief to address this issue,
   and, although he clearly articulated the SSET's evaluated concerns with
   regard to Northrop Grumman's boom design, his testimony regarding any
   schedule and/or cost risk associated with these concerns was conclusory.
   See, e.g., HT at 1009-13, 1016-17, 1022. Although the record, including
   the SSET team chief's testimony, indicates that some analyses of the
   impact of these evaluated concerns may have been performed, little detail
   has been provided. In this regard, we have been provided with no other
   testimony or statements from SSET members or citation to documentation in
   the record that would otherwise support the agency's judgment that there
   is little schedule or cost risk associated with these evaluated concerns.
   Given our recommendation below that the Air Force obtain and re-evaluate
   revised proposals, we think that this is also a matter that the agency
   should consider further.

   [73] "Organic" maintenance refers to maintenance that the agency does for
   itself as opposed to maintenance provided by the contractor. See HT at
   1215.

   [74] Unlike Northrop Grumman, Boeing committed to providing the required
   planning and support services within the specified 2-year timeframe. HT at
   1221.

   [75] The agency's product support subfactor team chief testified regarding
   this requirement:

     the idea behind that is to support the government in standing up this
     capability, so their approach would have to include the planning and
     support, the planning part being those type of actionable steps that
     support the type of things they would support us within that time
     constraint.

   HT at 1215.

   [76] The SSET product support subfactor team chief stated that the
   pre-final proposal revision briefing slide erroneously did not also refer
   to the SDD SOW, in addition to the production SOW, and that both SOWs
   would be implicated by Northrop Grumman's failure to commit to providing
   these services within the required 2-year timeframe. See HT at 1266-67.

   [77] Similarly, the SSET's product support subfactor team chief doggedly
   insisted that Northrop Grumman's failure to agree to perform the required
   services within the specified time frame was merely an oversight, even
   where he admitted under cross examination that "Northrop [Grumman] didn't
   forget about this issue," that Northrop Grumman's "[final proposal
   revision] was not silent on the issue," and that in fact "Northrop Grumman
   did consider the issue; they just decided not to address it in their
   [final proposal revision]." See HT at 1274-76.

   [78] Northrop Grumman does not explain why it made the "intentional"
   choice not to specifically include the 2-year requirement in the
   contractual SOW, even though it was repeatedly requested to do so by the
   Air Force.

   [79] In any event, the SSAC's and SSA's judgment that the firms' proposals
   were essentially equal under the product support subfactor is undermined
   by their erroneous conclusion that Northrop Grumman's failure to commit to
   the 2-year timeframe was an oversight.

   [80] Thus, the Air Force essentially concedes that the conclusion in the
   source selection decision that Boeing's evaluated MPLCC was more than
   Northrop Grumman's was in error. [In preparing the public version of the
   protected decision, as the Air Force correctly points out, one of the five
   acknowledged errors was actually with respect to repair costs. The
   magnitude of these five errors remains unchanged.]

   [81] The KC-30 is capable of carrying [Deleted] passengers, while the
   KC-767 can carry [Deleted] passengers. AR, Tab 55, PAR, at 18-19.

   [82] The Air Force also suggests that the RFP "directed that the MILCON
   portion of the MPLCC would be estimated entirely by the Air Force, with
   inputs from both Air Mobility Command (AMC) and Air Force Materiel Command
   (AFMC)." Air Force's Post-Hearing Comments at 119; see Air Force's
   Memorandum of Law at 221. This argument is based upon section M.2.5.2.4 of
   the RFP, which informed offerors that the agency's assessment of MPLCC
   would include evaluating MILCON costs and further informed offerors:
   "Note: Air Mobility Command and Air Force Materiel Command are estimating
   MILCON." This section does not, however, inform offerors that the Air
   Force would not consider their proposals in preparing this estimate.

   [83] The test base was not included in the agency's MILCON cost
   evaluation.

   [84] The SDD aircraft are provided to the Air Force under a cost
   reimbursement line item.

   [85] A Monte Carlo simulation is a cost risk analysis model that is
   generally used for quantifying the lowest and highest possible costs of
   weapons systems, based upon estimated costs of various components. See
   TRW, Inc., B-234558, June 21, 1989, 89-1 CPD para. 584 at 3 n.1. Developed
   in 1946 by a mathematician who pondered the probabilities associated with
   winning a card game of solitaire, a Monte Carlo simulation is used to
   approximate the probability outcomes of multiple trials by generating
   random numbers. In determining the uncertainty associated with a program's
   point estimate, a Monte Carlo simulation randomly generates values for
   uncertain variables over and over to simulate a model. Cost Assessment
   Guide: Best Practices for Estimating and Managing Program Costs,
   GAO-07-1134SP, July 2007, at 154.

   [86] We have in a number of decisions explained the relationship between
   probable cost adjustments and proposal risk, but in no case have we found
   that an agency's adjustment of an offeror's proposed costs of performing a
   contract should be based only upon risk, and not upon a reasoned
   assessment of the realism of the proposed costs being adjusted. Thus, for
   example, we have recognized that an agency is not required to upwardly
   adjust an offeror's proposed costs which the agency found realistic, even
   where the agency also assessed some risk with regard to those costs. See,
   e.g., ITT Indus., Inc., B-294389 et al., Oct. 20, 2004, 2004 CPD para. 222
   at 15-16; Vinnell Corp., B-270793, B-270793.2, Apr. 24, 1996, 96-1 CPD
   para. 271 at 6. Conversely, an agency may both make cost realism
   adjustments and assign proposal risk, where "the cost adjustments are
   necessary to reflect the offeror's probable costs of performance based on
   its proposal," and that there continued to be proposal risk despite the
   cost adjustment. See Raytheon Co., B-291449, Jan. 7, 2003, 2003 CPD
   para. 54 at 16 n.12.

   [87] Similarly, we do not see any validity to using the overall cost
   growth associated with the Boeing's Multi-Mission Maritime Aircraft
   contract with the Navy to forecast cost growth associated with Boeing's
   SDD non-recurring engineering costs. In addition, Boeing asserts, without
   rebuttal, that the cost growth under that contract was due to reasons
   unrelated to non-recurring engineering costs. See Boeing's Protest at 85.

   [88] The Air Force argues that Boeing is not prejudiced by these errors
   because the SSA in [the SSA's] selection decision stated that [the SSA]
   would have selected Northrop Grumman's proposal for award "even if
   Boeing's proposed cost/price had not been adjusted upward by the
   Government and Boeing's cost/price risk rating for SDD had been rated as
   LOW." AR, Tab 54, Source Selection Decision, at 19-20. We disagree. As
   concluded above, the Air Force erred in the evaluation of technical
   proposals and the conduct of discussions and this statement by the SSA
   does not address any of those errors. In any event, this statement by the
   SSA, which is unsupported by specific analysis, would not seem to reflect
   the reasoned consideration of cost or price to the government that a
   selection official is required to provide in performing a trade-off
   analysis. See, e.g., Shumaker Trucking and Excavating Contractors, Inc.,
   B-290732, Sept. 25, 2002, 2002 CPD 169 at 6.

   [89] The Air Force's evaluation of the fuel costs associated with the
   firms' proposed aircraft has been the subject of much argument and hearing
   testimony, and the record indicates that the agency did not do much more
   than an assessment that the offerors' own proposed fuel burn rates
   (gallons of fuel burned per hour) was reasonable. The record also shows,
   however, that even a small increase in the amount of fuel that is burned
   per hour by a particular aircraft would have a dramatic impact on the
   overall fuel costs (for example, Boeing notes that even a
   [Deleted]-percent increase in the amount of fuel per hour that is burned
   by the KC-30 would result in a $[Deleted] million increase in Northrop
   Grumman's life cycle costs for fuel, see Boeing's Post-Hearing Comments,
   at 139). Given our recommendation below that the Air Force reevaluate
   proposals and obtain revised proposals, this is another matter that the
   agency may wish to review to ascertain whether a more detailed analysis of
   the fuel costs is appropriate.

   [90] Our Office will not sustain a protest unless the protester
   demonstrates a reasonable possibility of prejudice, that is, unless the
   protester demonstrates that, but for the agency's actions, it would have
   had a substantial chance of receiving the award. See McDonald Bradley,
   B-270126, Feb. 8, 1996, 96-1 CPD para. 54 at 3; see Statistica, Inc. v.
   Christopher, 103 F.3d 1577, 1581 (Fed.Cir. 1996).