B-245761, B-245761.2, Jan 31, 1992, 71 Comp.Gen. 203
B-245761,B-245761.2: Jan 31, 1992
Where there was a significant discrepancy between the protester's historical indirect cost rates and its proposed rates. The proposed rates were based upon speculative projections of increased future business and unexplained reductions in overhead expenses. The procuring agency is not bound by the audit recommendations of the Defense Contract Audit Agency since such recommendations are only advisory. Which was questioned during discussions. That was lower than the agency's guideline rate. Which was that used by the offerors and other subcontractors. In a procurement in which technical factors were stated to be more important than cost. Higher priced offer was the most advantageous to the government.
B-245761, B-245761.2, Jan 31, 1992, 71 Comp.Gen. 203
PROCUREMENT - Competitive Negotiation - Best/final offers - Price adjustments 1. The procuring agency, as a part of its cost realism analysis in a negotiated procurement for a cost reimbursement contract, properly adjusted the protester's unsupported assignment in its best and final offer of higher priced personnel from the lowest overhead cost center to the highest overhead cost center, resulting in a significant evaluated increase in the protester's proposed costs, where the agency reasonably concluded that performance of the contract would require the use of the higher priced personnel in the higher overhead cost center. PROCUREMENT - Competitive Negotiation - Offers - Cost realism - Rates - Cost reimbursement contracts 2. The procuring agency, in conducting a cost realism analysis in a negotiated procurement for a cost reimbursement contract, reasonably used the protester's actual indirect cost rates from the prior fiscal year, instead of the protester's significantly lower proposed rates, where there was a significant discrepancy between the protester's historical indirect cost rates and its proposed rates, and the proposed rates were based upon speculative projections of increased future business and unexplained reductions in overhead expenses. PROCUREMENT Competitive Negotiation - Contracting officer duties - Effects - Advisory opinions 3. The procuring agency is not bound by the audit recommendations of the Defense Contract Audit Agency since such recommendations are only advisory. PROCUREMENT - Competitive Negotiation - Offers - Cost realism - Adjustments - Rates 4. The procuring agency's cost realism adjustment of the overhead rates of the protester's proposed subcontractor in a negotiated procurement for a cost reimbursement contract reasonably used the rate associated with on- site performance, where the subcontractor offered, without - explanation, an "off-site" overhead rate for the first time in the solicitation required on-site performance. PROCUREMENT - Competitive Negotiation - Offers - Evaluation - Cost realism - Rates 5. Procuring agency, in conducting a cost realism analysis in a negotiated procurement for a cost reimbursement contract, reasonably utilized the agency's labor escalation rate guideline in adjusting the inadequately supported labor escalation rate of the protester's subcontractor-- which was questioned during discussions-- that was lower than the agency's guideline rate, which was that used by the offerors and other subcontractors. PROCUREMENT - Competitive Negotiation - Contract awards - Administrative discretion - Cost/technical tradeoffs - Technical superiority 6. The procuring agency reasonably determined-- in a procurement in which technical factors were stated to be more important than cost-- that the awardee's higher rated, higher priced offer was the most advantageous to the government, where the agency found that the awardee's technical superiority outweighed the protester's 2 percent evaluated cost advantage, taking into account cost realism. PROCUREMENT - Competitive Negotiation - Offers - Evaluation - Cost/technical tradeoffs - Weighting PROCUREMENT - Competitive Negotiation - Offers - Evaluation errors - Evaluation criteria - Application 7. The agency's consideration of performance efficiency in balancing cost and technical considerations was not the improper application of an unstated evaluation factor but simply a tool in performing the cost/technical tradeoff. PROCUREMENT - Competitive Negotiation - Competitive advantage - Non-prejudicial allegation 8. Protest allegation that awardee received an unfair price advantage, in a negotiated procurement containing incentive awards for meeting negotiated small and small disadvantaged subcontracting plan goals, due to the fact that the procuring agency accepted the awardee's subcontracting plan goals, which were less than the 5 percent goal encouraged by the solicitation, is denied, because the plan, which was negotiable and properly approved under applicable regulations, was otherwise acceptable, and the protester was not prejudiced.
Purvis Systems Incorporated:
Purvis Systems Incorporated protests the award of a contract to Analysis & Technology Incorporated (ATI) under request for proposals (RFP) No. N66604-90-R-1752, issued by the Naval Underwater Systems Center (NUSC), Department of the Navy, New London, Connecticut, for technical and analytical services in support of the agency's performance evaluation of surface ship anti-submarine warfare sonar systems. Purvis objects to NUSC's cost realism evaluation of its proposal, the agency's source selection decision, and the agency's approval of ATI's small and small disadvantaged business subcontracting plan goals.
We deny the protests. /1/
The RFP, issued September 11, 1990, contemplated the award of a 5 year, cost-plus-fixed-fee, indefinite delivery, indefinite quantity contract for technical and engineering services in support of the agency's performance evaluation of its surface ship anti-submarine warfare sonar systems. Specific contract services include the design and planning of system tests and exercises; the design and development of test equipment; and the collection and analysis of test data. A total estimated level of effort was stated, along with an estimate of the composition of the total level of effort by labor category.
The RFP stated that award would be made to the responsible offeror whose conforming offer was determined to be the most advantageous to the government, cost and other factors considered. The RFP listed the following technical evaluation factors:
Technical approach was stated to be the most important factor with personnel being the next most important factor followed by the other three factors, which were equally weighted. The RFP also provided that technical considerations were more important than cost and that the degree of cost's importance would increase in relation to the degree of evaluated equality of offerors' technical capabilities. Offerors were also informed that a cost realism analysis would be performed.
By the November 5, 1990, closing date, NUSC received three offers, including the offers of Purvis and ATI. Only Purvis and ATI were determined susceptible of being made technically acceptable. /2/
A cost realism analysis of the initial offers was conducted by the contracting officer and an NUSC cost/price analyst. The agency questioned Purvis's proposed overhead and general and administrative (G&A) rates. this regard, Purvis proposed to perform this RFP work using several of its cost centers. Purvis utilizes several cost centers-- each with a separate overhead rate-- in its accounting system for the purposes of accumulating and allocating costs to be charged to appropriate cost objectives, e.g., contracts. Purvis's indirect cost rates were questioned because they were lower than the firm's historical rates the agency obtained from the Defense Contract Audit Agency (DCAA). /3/ The agency also questioned direct labor costs, facilities capital cost of money, and subcontractor costs.
Only Purvis and ATI were determined to be in the competitive range, and written discussions were conducted with each firm. NUSC requested support for each item of Purvis's proposed costs that had been questioned in the initial cost evaluation. Purvis was also informed that its "proposed overhead and G&A rates are not in accordance with current DCAA recommendations."
On April 3, 1991, NUSC received revised proposals from Purvis and ATI, responding to the written discussions. Purvis, in response to NUSC's questions regarding its assertedly unrealistically low indirect cost rates, reduced its indirect cost rates from those originally proposed and stated that its new rates reflected adjustments Purvis had calculated, assuming award under the RFP and other future business. Purvis also stated that its proposed rates had been submitted to DCAA as part of a forward pricing rate submission. /4/ A copy of the DCAA forward pricing rate was included in Purvis's revised proposal.
NUSC reevaluated Purvis's and ATI's offers including the technical proposal revisions. While both firms' offers were evaluated as technically acceptable, ATI's proposal was determined to be technically superior to Purvis's. In this regard, the agency used point scores as a guideline in assessing the relative merits of the proposals. ATI received 90.2 points while Purvis scored 85 points. ATI's proposal was rated technically superior to Purvis's under the most important evaluation factor-- technical approach-- and was rated better under the other factors, except for the facilities evaluation factor where the proposals were rated as equal. NUSC determined that ATI's evaluated strengths would "result in greater efficiency during contract performance."
Best and final offers (BAFO) were received from Purvis and ATI on May 7. The technical evaluation did not change. Another cost realism analysis was conducted because NUSC was concerned that both offerors had significantly reduced their proposed costs from those offered in their initial proposals. Purvis's BAFO cost proposal was also different from the revised cost proposal that it had submitted on April 3.
ATI's cost difference was determined to be attributable to the firm's change in its accounting system, as documented in its cost accounting standards (CAS) disclosure statement, which had been audited and approved by its cognizant administrative contracting officer (ACO). /5/ Purvis's cost differences were attributable to the firm's significant reduction in its projected indirect cost rates (overhead and G&A). In addition, Purvis proposed indirect cost rates that steadily declined for each year of contract performance, based upon Purvis's postulated future business. /6/ Purvis also allocated some of its higher salaried personnel from its cost center with its highest overhead rate to the cost center with its lowest overhead rate. Also, one of Purvis's proposed subcontractors reduced its proposed overhead rates by changing its proposed place of performance from "on-site" to "off-site."
On May 13, NUSC contacted DCAA and obtained actual 1991 fiscal year indirect rates for Purvis. These rates were significantly higher than the rates projected by Purvis in its BAFO. NUSC also contacted Purvis to determine why the firm had moved personnel from one cost center to another in its BAFO cost proposal. Purvis responded that in accordance with its established estimating system policy it utilized personnel in the cost center in which the employee would perform work. On July 26, NUSC received an unsolicited letter from Purvis stating that DCAA had informed Purvis, in an exit interview, that DCAA's audit recommendation would be that DCAA had "no exception to the forward pricing rates ... proposed by Purvis."
After the foregoing, NUSC, in its BAFO cost realism evaluation, determined that Purvis had not justified its movement of personnel from the cost center in which they were originally proposed. NUSC also found that Purvis's actual 1991 fiscal year indirect rates were more accurate indicators of the rates the agency would have to pay under the contract than the rates proposed by Purvis. NUSC also adjusted Purvis's subcontractor's costs based on its review of the subcontractor's overhead and labor escalation rates. The agency's cost adjustments resulted in the following probable cost determination: /7/
On August 8, 1991, the contracting officer determined that ATI's proposal was the most advantageous to the government, considering its technical superiority, which outweighed the $216,000 realistic cost advantage offered by Purvis. On August 13, the NUSC contract review board concurred with the award decision, subject to documentation of Purvis's historical indirect cost rates.
On August 14, 1991, Purvis's historical rates for fiscal years 1987 through 1991 were obtained from DCAA, and these rates showed that Purvis's indirect costs rates had been fairly steadily increasing. Despite the disparity between Purvis's recent indirect cost rates and those offered in its proposed forward pricing rate, DCAA indicated to NUSC that it was likely that DCAA would approve Purvis's forward pricing rates. NUSC again reviewed Purvis's indirect cost rates and determined that Purvis's proposed indirect cost rates were inconsistent with its historical trends, and that Purvis's most current 1991 rates were the most realistic assessment of its costs.
Award was made to ATI on September 12, and this protest followed Purvis's debriefing.
Purvis first protests that NUSC's cost realism adjustments of its proposed costs are unreasonable. Specifically, Purvis objects to (1) the agency's "reassigning" of its personnel to cost centers other than those cost centers proposed by Purvis in its BAFO, (2) NUSC's use of its fiscal year 1991 indirect cost rates in lieu of Purvis's proposed rates, and (3) the adjustments of its subcontractor's overhead and labor escalation rates.
When an agency evaluates proposals for the award of a cost reimbursement contract, an offeror's proposed estimated costs of contract performance and proposed fees are not considered controlling, since an offeror's estimated costs may not provide valid indications of the final actual costs that the government is required, within certain limits, to pay. See FAR Sec. 15.605(d); Amtec Corp., B-240647, Dec. 12, 1990, 90-2 CPD Para. 482. Consequently, a cost realism analysis must be performed by the agency to determine the extent to which an offeror's proposed costs represent what the contract should cost, assuming reasonable economy and efficiency. CACI, Inc.-- Federal, 64 Comp.Gen. 71 (1984), 84-2 CPD Para. 542. Because the contracting agency is in the best position to make this cost realism determination, our review is limited to determining whether the agency's cost realism analysis is reasonably based and not arbitrary. General Research Corp., 70 Comp.Gen. 279 (1991), 91-1 CPD Para. 183; Grey Advertising, Inc., 55 Comp.Gen. 1111 (1976), 76-1 CPD Para. 325.
NUSC states that it did not accept Purvis's "reassignment" of personnel in its BAFO from the cost center with the highest overhead rate to the cost center with the lowest overhead rate because Purvis did not provide an adequate explanation for this reassignment. The agency questioned Purvis regarding the cost centers, and Purvis explained that the higher overhead cost center is essentially a laboratory/manufacturing facility with responsibility for hardware and responsible for technical services. Purvis contends that it assigned personnel based on its determination of what facilities were necessary to perform the contract. NUSC concluded that the reassignment was unreasonable because, in order to adequately perform the RFP-required design and development of test equipment services, the reassigned personnel would need to use the facilities of the higher overhead rate cost center.
Purvis contends that contract is basically for data collection and analysis, and that performance would not require significant use of the laboratory/manufacturing facilities.
Also, Purvis argues that its offer reflected the necessity of some use of the equipment in the high overhead rate cost center "but rationally allocated the time of personnel to that cost center according to the need."
While it is true that the RFP essentially seeks data collection and analysis, the RFP also requires design and development services, which will require the use of the facilities in Purvis's higher overhead cost center. Purvis, in its initial proposal and BAFO, listed the facilities of the higher overhead cost center as available for contract performance and therefore received its highest technical score for the facilities evaluation factor. /8/ Contrary to the protester's arguments, Purvis's BAFO does not explain or justify the reallocation of personnel between the cost centers. Nor does the BAFO address how Purvis intends to satisfy the required design and development of test equipment contract tasks without the reassigned senior engineers or technicians. It is an offeror's obligation, when introducing changes in its BAFO, to demonstrate how its revised offer will satisfy the government's requirements since the agency is not required to reopen discussions to afford an offeror an opportunity to demonstrate compliance. See Loral Elecs. Sys., B-224540, Feb. 10, 1987, 87-1 CPD Para. 143 (insufficiently explained cost center revisions in BAFO justified downgrading of BAFO from initial proposal rating).
From our review of Purvis's initial, revised, and BAFO proposals, we do not find unreasonable the agency's conclusion that Purvis, in order to perform the contract work, would be required to have its reassigned personnel use the facilities of the higher overhead cost center-- as Purvis had originally proposed in its initial proposal. /9/ The record reasonably shows that Purvis's unsupported shifting of personnel in its BAFO was merely intended to reduce proposed costs, a reduction that the Navy reasonably determined may not be realized in the actual performance of the contract. Accordingly, we think the agency reasonably adjusted Purvis's proposed BAFO costs by placing the reassigned high salaried personnel in the cost center where they were originally proposed.
Purvis also objects to NUSC's cost realism adjustment of its proposed indirect cost rates. Purvis argues that the use of its higher 1991 fiscal year indirect cost rates, rather than its estimated future indirect cost rates, was unreasonable, because its estimated rates are based upon its projected future business, including award under the RFP, and upon reduced overhead expenses, such as rent and depreciation. Purvis contends that its forward pricing rate submission, as provided to DCAA and NUSC, substantiates the reasonableness of its much lower estimated indirect cost rates. Purvis also argues that DCAA approved these rates and NUSC unreasonably ignored DCAA's "approval."
During discussions, Purvis was informed of NUSC's concerns regarding the significant discrepancy between the firm's proposed indirect cost rates and its historical rates. Purvis failed, however, to demonstrate that its estimated rates were more accurate indicators than its actual rates of the indirect costs that the government would be required to pay under the contract. /10/ Rather, the information available to the agency, and in this record, shows that Purvis's significantly lower proposed indirect cost rates were contingent upon acquiring substantial future government contract business, including the RFP work, in addition to retaining all of its current business with the government.
Purvis's projections of future work, other than the RFP work, were not supported by any evidence to indicate the basis for the projections, i.e., current solicitations under which Purvis was competing or future solicitations under which Purvis intended to compete. Moreover, the record indicates that Purvis was not retaining all of its current government business but that this business was actually declining. addition, Purvis's documentation of its expected overhead costs was based upon a substantial reduction in its rent and upon no future depreciation expenses; however, these reductions in its overhead expenses were unexplained and unsupported.
We think the agency reasonably concluded that Purvis's projections of its future business and overhead expenses were too speculative to support the proposed indirect cost rates, particularly in light of the historical data that showed Purvis's indirect cost rates had been steadily increasing since its 1987 fiscal year. /11/ Even though NUSC expressed its concerns about the proposed rates to Purvis during discussions, Purvis lowered its overhead still further in its BAFO and, as discussed above, did so without adequate justification. Under the circumstances, NUSC could reasonably rely upon the prior year's actual rates as the best information available to the agency to determine the indirect cost rates that it could reasonably be required to pay under the contract. See Marine Design Techs., Inc., B-221897, May 29, 1986, 86-1 CPD Para. 502.
Purvis argues that DCAA approved its estimated indirect cost rates, and therefore NUSC should have accepted DCAA's determination. /12/ disagree. Even assuming DCAA had "approved" Purvis's indirect cost rates, which is not clear from this record, contracting officers are not bound by DCAA's audit recommendations since such recommendations are only advisory. OAO Corp., B-228599.2, July 13, 1988, 88-2 CPD Para. 42. Indeed, a contracting officer may not blindly rely on DCAA audit recommendations, since the contracting officer is solely responsible for the cost realism determination. See American Mgmt. Sys., Inc.; Dept. of the Army-- Recon., 70 Comp.Gen. 510 (1991), 91-1 CPD Para. 492. Here, as noted above, NUSC's contracting officer considered all the information available, including that from DCAA and NUSC's cost/price analyst, in reasonably determining that Purvis's actual indirect cost rates were more accurate indicators of the cost the government could reasonably be required to pay under the contract. Marine Design Techs., Inc., supra.
The record also supports the agency's adjustment of Purvis's subcontractor's overhead. Purvis's subcontractor, for the first time in its BAFO and without explanation, lowered its proposed overhead rates by changing its proposed place of performance from "on-site" to "off-site." Since the RFP required all performance to be on the contractor's site, NUSC used the overhead rate associated with the subcontractor's on-site performance. Purvis now argues that its subcontractor's intended off-site performance was to be at Purvis's facility. Since this was not explained in Purvis's BAFO, we think the agency reasonably assumed that the subcontractor improperly intended to perform at a facility that was not on the contractor's site and disregarded the unsubstantiated reduction in overhead associated with this change.
Regarding NUSC's adjustments to the subcontractor's labor escalation rate, the record shows that during discussions NUSC questioned the subcontractor's offer of a substantially lower escalation rate than that established by the Office of the Assistant Secretary of the Navy (Shipbuilding and Logistics) (ASN(S&L)). /13/ While Purvis's subcontractor responded that its proposed labor escalation rate reflected its actual escalation of salaries for 13 months, it provided historical data for just three labor categories that the subcontractor intended to use in performing the subcontract. The agency found that the information provided was too limited in time and scope to demonstrate the reasonableness of its lower proposed escalation. /14/ Given that the offerors and other subcontractors-- all who must perform the required services in the New London, Connecticut, labor market area - utilized the same rate as the ASN(S&L) rate and in the absence of more adequately documented support for the subcontractor's proposed labor escalation rate, we think the agency reasonably adjusted Purvis's costs to reflect the ASN(S&L) rate. See Raytheon Servs. Co.] Informatics Info. Sys. Co., Inc., 59 Comp.Gen. 316 (1980), 80-1 CPD Para. 214.
Purvis also protests NUSC's source selection determination as unreasonable. Specifically, the protester complains that the NUSC in its best value determination did not accord sufficient weight to cost and that the agency, in balancing technical considerations with cost, employed an unstated evaluation factor-- efficiency.
Agency officials have broad discretion in determining the manner and extent to which they will make use of technical and cost evaluation results. Cost/technical tradeoffs may be made, and the extent to which one may be sacrificed for the other is governed only by the test of rationality and consistency with the established evaluation factors. Grey Advertising, Inc., supra. Award may be made to the higher rated, higher priced offeror where the decision is consistent with the evaluation factors and the agency reasonably determines that the technical superiority of the higher-priced offer outweighs the cost difference. See Oklahoma Aerotronics, Inc.-- Recon., B-237705.2, Mar. 28, 1990, 90-1 CPD Para. 337.
Here, the record shows that the agency properly weighed cost and technical considerations in performing its best value determination. Technical considerations were stated to be more important than cost, although the importance of cost could increase relative to the perceived equality of technical offers. Here, ATI's proposal was evaluated to be technically superior to the Purvis proposal, /14/ and NUSC concluded that this evaluated superiority would result in greater efficiency during contract performance. NUSC balanced ATI's evaluated superiority against Purvis's 2 percent realistic cost advantage, in part, by numerically comparing the additional labor Purvis's cost advantage could purchase against ATI's superior proposal. The agency determined that the "technical gain offered by ATI's proposal was worth the anticipated premium in costs," even considering the possible additional labor that Purvis's cost advantage could purchase. We find the agency's determination that ATI's technical superiority was worth the associated cost premium was reasonable and supported by the record.
Purvis argues that NUSC employed an unstated evaluation factor in considering the performance efficiency offered by ATI's superior proposal. We disagree. Rather than being employed as an evaluation factor, efficiency in performance was simply and properly used as a tool in assessing whether Purvis's cost advantage outweighed ATI's technical superiority. /15/ See generally Dalfi, Inc., B-224248, Jan. 7, 1987, 87-1 CPD Para. 24, aff'd, B-224248.2, Feb. 19, 1987, 87-1 CPD Para. 186.
Purvis finally protests NUSC's approval of ATI's small and small disadvantaged business (SDB) subcontracting plan goals because ATI's plan provided goals of less than 5 percent. Purvis argues that since the RFP provided for the payment of a bonus for exceeding subcontracting plan goals, ATI received an unfair price advantage when NUSC approved ATI's plan with goals below the 5 percent level in Purvis's plan.
The RFP incorporated by reference Department of Defense FAR Supplement (DFARS) Sec. 252.219-7009 (June 1988), which provides for incentive awards to contractors that exceed subcontracting plan goals by a stated percentage; the size of the incentive awards is dependent upon whether the stated subcontract goal was greater or lesser than 5 percent. Since Purvis offered 5 percent goals, Purvis, if it were to exceed its goals, would have been entitled to receive a larger incentive award than ATI will under its stated plan goals. Thus, we do not see how approval of ATI's plan gave it an unfair price advantage. To the extent Purvis believes the agency should not otherwise have approved a subcontract plan with goals of less than 5 percent, we point out that DFARS Sec. 252.219-7009, as incorporated in the RFP, specifically recognized that subcontracting plan goals could be less than 5 percent. See also FAR Sec. 19.705-4(c).
The protests are denied.
/1/ Portions of the protest record are subject to a General Accounting Office protective order to which counsel for Purvis and ATI have been admitted. Our decision, which is based upon protected, confidential information, is necessarily general.
/2/ The other offer was rejected as technically unacceptable. /3/ NUSC compared Purvis's proposed rates with its actual rates through April 1990, which had been approved for use in the firm's 1991 fiscal year.
/4/ A "forward pricing rate agreement" is a written agreement negotiated between the government and a contractor concerning the billing rates to be utilized by a contractor for a specified period of time. Federal Acquisition Regulation (FAR) Sec. 15.801.
/5/ CAS-covered contractors are required to file a disclosure statement, describing the firms' cost accounting practices and procedures, with their cognizant ACOs and contract auditors. See FAR Sec. 30-202.
/6/ Purvis's indirect cost rates in its BAFO were only slightly lower than that offered with its April 3 revised proposal but were significantly lower than those offered in the initial proposal.
/7/ NUSC also adjusted Purvis's proposed costs for material handling and facilities capital cost of money. The dollar amount of these adjustments is insignificant relative to Purvis's total proposal cost. NUSC also made relatively minor adjustments to ATI's proposed costs.
/8/ The facilities technical evaluation factor was the only one under which Purvis's proposal was rated technically equivalent to ATI's.
/9/ There is no support in the record for Purvis's allegation that NUSC required the reassignment of these personnel.
/10/ In this regard, Purvis alleges that its higher indirect cost rates for the prior 2 years were the result of "one-time events which are not likely to recur," but has provided no evidence or explanation for what these events were or why they will not recur.
/11/ This might reflect Purvis's declining business.
/12/ DCAA does not have authority to enter into forward pricing rate agreements with contractors; this authority is accorded to the contracting officer. See FAR Sec. 15.809 (FAC 90-3).
/13/ ASN(S&L)'s labor escalation rates are intended to provide guidelines to contracting officers in negotiating contract prices. /14/ Purvis and its subcontractor-- a long time government contractor-- could have provided other evidence to support its labor escalation rate history, such as the labor escalation history for the company as a whole or for other similar cost centers.
/15/ Purvis complains that the agency erred in computing Purvis's overall numerical technical score and that it was entitled to a 1 point higher technical score. NUSC admits that it incorrectly computed Purvis's overall numerical technical score and that the protester was entitled to an 86 instead of an 85 point score. The record shows, however, that Purvis was not prejudiced by this scoring error. Numerical scores were only used as evaluation guides and the record shows that the source selection was based on a determination that ATI was technically superior to Purvis in all but one of the evaluation areas. That is, the source selection was based on ATI's evaluated technical superiority rather than the point scores. The record also does not support Purvis's allegation that efficiency had been evaluated as a part of the agency's evaluation of technical proposals.