Matter of: ARAMARK Services, Inc. File: B-282232.2 Date: June 18, 1999

Matter of: ARAMARK Services, Inc. File: B-282232.2 Date: June 18, 1999

DIGEST

Attorneys

DECISION

ARAMARK Services, Inc. protests that the specifications in request for proposals (RFP) No. FTC 99-3, issued by the Department of the Treasury, Federal Law Enforcement Training Center (FLETC) for the provision of food services at its training center in Glynco, Georgia, are defective because they are unduly restrictive, fail to provide sufficient information to allow vendors to compete intelligently and on a relatively equal basis, and impose undue risk on the contractor.

We deny the protest.

The RFP, issued on January 27, 1999 and amended three times, calls for offers to provide three meals per day in support of center training programs in a cafeteria-style dining facility on the center grounds. The center offers a variety of training programs that range from a single week to 18 weeks in duration, for a student population that fluctuates in size but typically numbers approximately 2,000. The students generally receive all their meals in the dining facility, and many staff members--of whom there may be as many as 2,000--eat some of their meals in the facility.

ARAMARK is the incumbent contractor, having been awarded the current food services contract for the center in February 1998. In January 1999, FLETC decided not to exercise the option under ARAMARK's contract, and that a new solicitation should be issued for this requirement because of substantial performance changes that the agency determined needed to be implemented. These changes include: lengthening the specified meal periods; changing meal portions and sizes; changing from a payment system based upon the average resident student population for the month to a system based on an actual count of meals served; and putting into service a scanning system capable of reading bar codes on student identification badges, in order to make the count.

The RFP contemplated the award of a contract for a 6-month base period, with four 1-year options. The RFP included, in section B, Contract Line Item Numbers (CLINs) that were to be priced in a variety of ways. In relevant part, CLINs 0005, 0006, and 0007, for food services for the three daily meals for the base period, /1/ called for offerors to provide prices on a per-person basis on a sliding scale, based on the average resident student population (ARSP) for any given week. /2/ Offerors were to fill in blanks indicating their proposed weekly rate per person, on a scale ranging (in 50-person increments) from an ARSP of "below 250" to "3751- 4000+." These CLINs stated that, while there would be no guaranteed student population, "the estimated average number of resident students is 2200." RFP Sec. B, at B-2. For evaluation and source selection purposes, the total amount for these CLINs was to be estimated as 2,200 students times the offeror's weekly rate times 26.1 weeks.

The pricing methodology changes under the option years. For each of the option years, the corresponding CLINs (1005, 1006, 1007; 2005, 2006, etc.) for meal services were to be based on a sliding scale, fixed-price rate for an actual count of meals per day. Offerors were to fill in blanks reflecting a sliding scale (ranging in 50-person increments from "below 750" to 11251 - 12000+"), with a daily rate per meal. These CLINs project an estimated count of 6600 meals per day, and set forth the basis for price evaluation and source selection as 6600 meals per day times the offeror's daily rate times 365 days. RFP Sec. B, at B-8, B-9, B-10.

The actual count system is described in the RFP in section C.34, which provides that all students will receive a FLETC-issued identification badge with a 10-digit bar code that will be presented at each meal for scanning. The RFP specifies what type of information each digit should provide, stating, for example, that the 9th digit will indicate the category of food services to which the badgeholder is entitled. The RFP requires the contractor to furnish equipment that will be able to:

1. Provide an adequate number of scanners (minimum of 4) which shall ensure that students are processed at a rate of at least 30 per minute per scanner.

2. Scan the bar code on the badge, interpret the 9th character in the bar code [indicating the category of entitled services] and give an audible signal if the student is not authorized to receive a meal during that meal period. . . .

3. Create a data file recording the entire ten (10) digit code, the date, the 24-hour (military) time of day and the station at which the bar code was scanned.

4. Provide a report of the actual count of students per meal period [in hard copy and on disk].

. . . . .

5. Provide all system operations maintenance, training and supplies.

RFP Secs. C.34(a), at C- 35, C-36. By amendment responding to prospective offerors' questions, the RFP identified the manufacturer of the cards that the contractor would be required to scan and stated that the government had no preference for any particular brand or type of scanning equipment.

ARAMARK protests that the "specifications for an Actual Count card scanning system improperly omits specific, required information denying prospective offerors the ability to compete intelligently and on a relatively equal basis." Protest at 4. ARAMARK explains that off-the-shelf card scanning systems operate by comparing information previously loaded into memory--which the protester calls "the required database for the system"--with new information scanned at the terminal, Protest at 5, and complains that "[t]here is no information in the solicitation suggesting that the Department of Treasury will, or is able, to provide the contractor with the database required by off-the-shelf systems." Id. The agency report treats this complaint as questioning "whether the specifications in the solicitation are sufficiently adequate to allow the contractors to provide a scanning system capable of reading the student identification badges and producing a daily report of meals served." Memorandum of Law at 3-4. The report states that the "specifications are clearly performance based, i.e., the Government provides the input (the bar codes on the Student ID), and the contractor is to provide the necessary scanners and equipment to provide the output," and notes that ARAMARK "has not attempted to identify any short coming in the specification or recommend any additional specifications." Memorandum of Law at 4-5. In its comments on the agency report, the protester complains that the agency has misconstrued its basis for protest. ARAMARK alleges that it has discussed FLETC's inability to provide accurate student count information with the contracting officer on several occasions during the term of the current contract, and that the contracting officer has cited this inability as a principal reason for requiring an electronic card-scanning system to capture actual counts of students participating in meal service at FLETC. The protester asserts that "[b]ased on the Contracting Officer's previous statements [regarding the difficulty in accurately estimating the number of students receiving meals at the center] ARAMARK believes that FLETC is incapable of providing the contractor with the required information from the students' identification cards for the system prescribed by the specifications." Protester's Comments at 5.

ARAMARK's specific basis for protest in this regard is difficult to discern from its submissions. ARAMARK appears to be speculating that the agency will not be able to fulfill its obligation to provide properly encoded student identification cards under the contract. /3/ There is nothing in the record which supports this prediction. ARAMARK appears to be construing the agency's inability to predict with certainty how many students will actually take their meals in the center as an inability to determine (at the time student identification cards are issued) whether a student is entitled to receive meals at the center. However, nothing in the record suggests that the agency cannot determine whether students enrolled in training are entitled to meals.

To the extent that ARAMARK is objecting that the agency cannot determine in advance with precision how many students will actually receive meals, this appears to be true but it does not provide a basis to sustain the protest. As the protester points out in its comments, students who are entitled to take their meals at the center do not always do so, for a variety of reasons, and it is virtually impossible to predict the exact number of students who will avail themselves of the food services at any given time. Protester's Comments at 2. It is undisputed that the agency's determination to change from the current payment system to a system based on an actual count of meals served has been generated, in part, by the agency's inability to accurately project how many students will actually make use of the provided food services at any given time. ARAMARK alleges that the disparity between the estimated number of students who will be served and the actual number who are served can vary, sometimes significantly, and that the disparity causes the contractor to incur excessive food costs, higher food waste, and higher labor costs which do not get reimbursed. Protest at 8. In terms of the protest, we fail to see how this adversely impacts ARAMARK, which, as the incumbent, is in the best position of any offeror to accurately estimate the extent of that disparity.

ARAMARK protests, however, that the specifications place the offeror "at undue financial risk." Protester's Comments at 7. ARAMARK also is concerned that, as the incumbent, only it is aware of the risks involved in pricing this work and that it will be underpriced by firms which do not price for the risks involved. In this regard, ARAMARK's protest includes objections to the base period pricing methodology as well as the actual-count pricing method to be used in the option years, but does not specify what methodology would be "proper." In essence, ARAMARK objects to the imposition of any risk on the contractor, and seeks to have the agency restructure the contract in a manner that would provide revenue guarantees or other safeguards for the contractor.

The contracting agency, not the protester, is responsible for determining its needs and the best means of meeting those needs since the agency is most familiar with the conditions under which supplies or services are to be used, and our Office will not question this assessment unless it is shown to be unreasonable. Kastle Sys., Inc., B-231990, Oct. 31, 1988, 88-2 CPD Para. 415 at 5. The fact that ARAMARK would prefer a more favorable payment structure does not call into question the reasonableness of the agency's determination that its needs are best met by the assessment processes set forth in the RFP.

Solicitations must contain sufficient information to enable bidders to compete intelligently and on a relatively equal basis. RMS Indus., B-248678, Aug. 14, 1992, 92-2 CPD Para. 109 at 2. There is no requirement, however, that a solicitation be drafted so as to eliminate all possible risk to the contractor. Id. An agency may properly impose substantial risk upon the contractor and minimal risk upon itself, and offerors reasonably are expected to use their professional expertise and business judgment in anticipating risks and preparing their offers. AT&T Corp., B-270841 et al., May 1, 1996, 96-1 CPD Para. 237 at 8. Here, the risk imposed on offerors appears to affect all offerors equally and all offerors are capable of taking those risks into account in preparing their proposals; ARAMARK, as the incumbent, is most familiar with the recent historical pattern and in this respect should be particularly able to calculate the risk factor in preparing its proposal. It is proper for agencies to impose reasonable risks on contractors in order to limit the burdens on the government. International Creative and Training, Ltd., B-245379, Jan. 6, 1992, 92-1 CPD Para. 26 at 5. Under the actual-count system, offerors bear a risk that fewer students than anticipated may avail themselves of the food and that the contractor will be able to bill the government for only that smaller number of meals. If that risk did not fall on the contractor, the government would bear it: that is, either the contractor risks incurring costs for meals that are not served, or the government risks paying for meals in excess of the actual number served. It is within FLETC's discretion, in the exercise of its business judgment, to impose those risks, which ARAMARK has not shown to be unreasonable.

ARAMARK also objects to the requirement at C.34(a), which states as follows:

The Contractor shall provide equipment which can . . .[p]rovide an adequate number of scanners (minimum of 4) which shall ensure that students are processed at a rate of at least 30 per minute per scanner.

RFP Sec. C.34(a), at C-35.

ARAMARK protests that this specification is unduly restrictive of competition because it requires the capability to process students at a rate of at least 30 per minute per scanner. The protester calculates that this requires the contractor to scan 1 student per scanner every 2 seconds, and that the minimum requirement of four scanners would result in an ability to allow 14,400 students to pass through the system during a 2-hour meal period. Protest at 5. ARAMARK concludes that this requirement overstates the minimum needs of the agency and is logistically impractical because of factors beyond the control of the offeror, such as the speed with which students actually present their cards and pass through the scanner.

This allegation is without merit. Regarding the argument that the requirement unduly restricts competition, ARAMARK concedes that commercial off-the-shelf systems have the capability of processing scanned data in microseconds, and never alleges that ARAMARK or any other offeror is unable to provide equipment capable of scanning at this speed. Protest at 6. ARAMARK alleges that "[t]he plain meaning of this requirement is that the contractor will have to provide enough scanners to ensure that students can be processed through the scanners at the rate of 30 students per minute/per scanner," and argues that "[I]industry representatives who have toured the site and observed the meal period indicate that a number of scanners far greater than the four estimated by FLETC would be required to meet this requirement." Protester's Comments at 7.

While we think this specification is inartfully drafted, we do not agree with ARAMARK's interpretation, which expands the requirement beyond its intended meaning. The number of students that can be processed per scanner, per minute, remains the same, no matter how many scanners are provided: this requirement specifies only the rate at which each scanner processes cards. While the RFP requires that the contractor "[p]rovide an adequate number of scanners (minimum of 4)," it also seems to relate that requirement to a processing rate--i.e., per scanner--that is, however, wholly independent of the number of scanners. Nonetheless, this internal inconsistency does not provide any basis for ARAMARK's conclusion that the RFP requires offerors to provide "a number of scanners far greater than the four estimated by FLETC." Fairly construed, the solicitation reasonably requires that the contractor provide a minimum of four scanners, each capable of processing cards at a rate of 30 per minute. The requirement primarily pertains to the provision of adequate scanning equipment, not to the contractor's ability to move students during its performance of the contract. Contrary to ARAMARK's assertions, the specification calls for a scanner with the capacity to process cards at a certain rate in order to ensure that the scanner will not create a bottleneck situation when a number of students are lined up waiting to pass through the scanner. Memorandum of Law at 7. It does not require that the number of students passing through the scanners constantly be 30 per minute per scanner for the duration of the lunch service period, as ARAMARK suggests when it calculates a total of 14,400 students per meal period. Rather, the agency is reasonably requiring a system capable of keeping the line moving, without creating backups during periods of heavy usage. Since the protester concedes that commercially available scanners easily meet the specification, we see no merit to its objection.

Finally, ARAMARK alleges that the projected ARSP levels provided in the RFP are inherently deceptive as to the true risks and workloads to be associated with the proposed contract. /4/ Protester's Comments at 8. ARAMARK alleges that as the incumbent, its superior knowledge of the costs and risks involved effectively denies it a chance to compete on an equal basis. In essence, ARAMARK is arguing that because it is the only offeror with direct experience in providing these services, it will not cost its offer on inaccurate assumptions, as other potential offerors would, and it is thus likely to be underpriced by offerors who do not have the same knowledge of the risks of performance.

Where estimates are provided in a solicitation, there is no requirement that they be absolutely correct; rather, they must be based on the best information available and present a reasonably accurate representation of the agency's anticipated actual needs. DSP, Inc., B-220062, Jan. 15, 1986, 86-1 CPD Para. 43 at 3. From the record here, we cannot conclude that the data relied upon in the agency's requirement projections is incomplete or misleading, or that the requirements of the solicitation are not based upon the best available information. The RFP includes historical information concerning the average weekly payment for students in training from 1992 through the end of 1998, showing the estimated amounts based on ARSP projections at the time, and the actual amounts paid based on actual ARSP for each contract period, as well as ARSP figures by week and partial week for fiscal years 1993-1998. RFP Technical exh. 27, at 1-6. During the term of the contract, the agency provides a projected estimate of the number of students in training in advance, first providing estimates 6 weeks prior to the actual training week and revising the figures 3 weeks, 2 weeks, and 1 week before the training week. The agency notes in its report that these projections are not 100 percent accurate, and that no reasonable contractor would assume that the projections would be perfectly accurate when preparing its offer. While ARAMARK complains that under the current contract, the number of students actually in training was often lower than the projected ARSP figures, it has neither shown that the historical data in the RFP is inaccurate, nor has it shown that more accurate data is available. In short, this aspect of ARAMARK's protest is without merit. G & B Packing Co., Inc., B-204192, April 20, 1982, 82-1 CPD Para. 359.

The protest is denied.

Comptroller General of the United States

1. Each of these CLINs called for the offeror's weekly rate per person, for all three meals; the different CLINs correspond to different hours of service that might be required under the contract, thus allowing the vendor to adjust its price to a longer or shorter meal service time requirement.

2. This pricing methodology for the base period is the same as it is under ARAMARK's current contract.

3. ARAMARK also complains that the solicitation fails to provide necessary data to the offerors (sometimes referred to in the protester's pleadings as the "required database"). See, e.g., Protester's Comments at 3-4. The protester has never adequately explained what that missing data (or database) is and we therefore conclude that it has failed to state a valid basis of protest in this regard.

4. ARAMARK initially also alleged that the RFP omitted important information concerning wages paid to hourly employees, Protest at 8, but conceded this issue in its comments. Protester's Comments at 11.