OPINION OF THE COURT
ARNOLD H. POLLOCK, Hearing Officer.*226 RECOMMENDED ORDER
A hearing was held in this case in Tallahassee, Florida on January 13, 1988, before Arnold H. Pollock, Hearing Officer. The issue for consideration was which bidder, if any, should be awarded the contract to provide Health Maintenance Organization services to state employees in the Orlando service area.
BACKGROUND INFORMATION
On July 31, 1987, the Respondent, Department of Administration, (DOA), issued Request for Proposal No. 88-05, for the providing of health maintenance organization services in the Orlando service area. Proposals were submitted by, among others, Pru-Care Health Plan, (Pru-Care), Health Options, Inc., (Health Options), Health Alliance Plan, Inc., (Health Alliance), Cigna Health Plan of Florida, Inc., (Cigna), and Florida Health Care Plan. Respondent, DOA, announced its proposed award to Pru-Care and Health Options. Petitioner, Cigna, filed a formal protest challenging the proposed award. Health Options and Pru-Care intervened, as did Health Alliance and Florida Health Care Plan.
At the hearing, Petitioner presented the testimony of John F. Black', III, an administrator with the Department of Insurance; and Bruce Breckerink, Director of Planning for Cigna; and introduced Petitioner’s Exhibits 1 through 6. Respondent presented the testimony of Dennis E. Nye, Director of State Employees’ Insurance with DOA, and Intervenor, Pru-Care presented the testimony of John A. Joiner, Group Manager for Pru-Care in Orlando. Pru-Care also introduced its Exhibit 1.
Subsequent to the hearing, Petitioner, Respondent and Intervenors, Pru-Care and Health Options submitted proposed findings of fact which have been ruled upon in the Appendix to this Recommended Order.
FINDINGS OF FACT
1. On July 31, 1987, DOA mailed a Request for Proposal, (RFP), to various Health Maintenance Organizations, (HMOs), soliciting proposals for the providing of HMO services in the Orlando service area. Petitioner, Cigna, and the various Intervenors herein, submitted proposals which were opened by DOA on August 28, 1987, with a contemplated date of award of September 14, 1987 and an effective date of contract on January 1, 1988.
2. Section 2 of the RFP defined the general purposes of the *227procurement as being to meet benefit objectives of DOA and to provide high quality benefits and services to state employees. Specifically, the objectives of the RFP were:
A proactive approach to cost containment including an emphasis on aggressive claims management, utilization review, and superior statistical reporting.
Quality medical care which encourages health promotion, disease prevention, early diagnosis and treatment. Stability in the financial structure of offered health plans.
Professional, high quality service in all administrative areas including claims processing, enrollment, membership services, grievances, and communications.
Competitive premium rates which take into account the demographics and, if appropriate, the claims experience of state employees.
Other stated objectives included:
Have each county or contiguous group of counties be considered one service area.
Award no more than two contracts per service area; however, the awards will be based on the HMO’s ability to respond to the needs of employees and on accessibility by employees.
Have reciprocal agreements between locations, if an HMO has multiple service areas.
Enter into a two year, non-experience rates contract. A provision will be included tying renewal action at each of the two renewals to the consumer price index, (CPI), for medical care services.
In order to be considered as a “qualified” proposer, an organization had to be licensed by the Department of Insurance pursuant to Part II, Chapter 641, Florida Statutes.
3. Section IX of the RFP listed five major criteria for evaluation of the proposals. They were:
1. Premium cost
2. Extensiveness of service area — by county and/or contiguous counties.
3. Plan Benefits as follows:
a. Covered services
b. Limitations and exclusions
c. Co-payments, deductibles, and coinsurance features.
*228d. Range of providers including specialists and numbers of hospitals
e. Out of service area coverage
f. Grievance procedures.
4. Accessibility as follows:
a. Reciprocal agreements
b. Provider locations
c. Number of primary care physicians and specialists, in relation to
membership
5. Completeness of proposals
The first four of the above objectives were called for by the Legislative action providing for these procurements to be effective January 1, 1988. The fifth, completeness of proposals, was not identified by the Legislature but was added by DOA.
4. The Department reviewed and evaluated all the proposals submitted by Petitioner and the various Intervenors. Each proposer was evaluated by three individual evaluators. Two separate sets of evaluations were performed; the second coming upon the direction of the Secretary who, after the first evaluation and recommendation of award, concluded the standards for evaluation had been too subjective and directed a second evaluation utilizing more objective standards. During this second evaluation process, after the actual evaluation had been done but before the recommendation was forwarded to the Secretary, several computer treatments of the raw scores were accomplished by Mr. Nye because of additional unidentified factors brought to his attention. The final computer run identified that Central Florida Physicians, not a part to this action, received the highest point total followed by Health Options, Pru-Care, and Petitioner, Cigna. Mr. Nye, who had designed and supervised the evaluation process, recommended to the Secretary that Central Florida Physicians, Health Options, and Pru-Care receive the award even though the guidelines called for only two recommendees. Central Florida Physicians was recognized to be in financial difficulties though it received the highest rating, and in order to provide two viable candidates in the event that provider should be disqualified, Health Options and Pru-Care were added. Central Florida Physicians was, in fact, subsequently disqualified due to financial insolvency. This left Health Options and Pru-Care as the two providers with the highest evaluations and the Secretary made the award to them. At the final count, Health Options received a point total of 64.635; Pru-Care, 57.415; and Cigna, 56.83, or a difference of .585 between Pru-Care and Cigna.
*2295. According to Mr. Black, an administrator with the Department of Insurance and responsible for the licensing of HMOs and other health care facilities, as of January 12, 1988, Pru-Care was not licensed in Volusia or Lake Counties and department records show that Pru-Care has never been or requested to be licensed in those counties.
6. Mr. Beckerink, the Director of Planning for Cigna of Florida, who oversaw Cigna’s proposal for the Orlando area and who reviewed DOA’s evaluation of the various proposals submitted, carefully examined the evaluation forms for both Cigna and Pru-Care and concentrated on scores relating to costs, benefits, accessibility, service area, and completeness. He noted that Pru-Care received 10 points for proposing service in Orlando, Seminole, Osceola, Lake, and Volusia Counties though it is not licensed in the latter two, whereas Cigna received only 4 points for Orange and Seminole Counties. Cigna is licensed in all five counties and has hospitals and physicians in Seminole, Osceola, and Orange Counties. He contends Pru-Care received credit by the evaluators for five counties when it is licensed only in three, an unearned award of 4 points, and Cigna was awarded credit for only two counties when it is licensed in five, an improper denial of 6 points.
7. According to Mr. Nye, the award to Pru-Care was based on its representation it would provide service in five counties. The Department of Insurance did not tell him, at the time, in which counties PruCare was licensed. As a result, he took the proposal, which indicated the five counties, at face value. Credit was given only for full counties to be served and Cigna’s proposal indicated it would deliver service to two full counties and to only portions of three counties.
8. The evidence indicates that Pru-Care’s facilities are primarily in Orange and Seminole Counties with some service offered in the extreme northern portion of Osceola County, too far away for those individuals living in the southern portion of that county reasonably to take advantage of it. Mr. Nye indicates that driving time, which would be the problem here, is not a consideration in assessing accessibility, but merely a factor in quality of service. The department is not concerned with whether it is convenient for the employee to get to the service but merely whether the service will be offered to anyone residing in the county. For this reason, Pru-Care was awarded credit for Osceola county since it proposed to enroll any eligible employee living in the county whether service was convenient to that part or not, whereas Cigna, which limited its enrollment in certain counties to those personnel living in only a part of the county, was not given any *230credit for those partially served counties. Mr. Nye admits that had he known Pru-Care was not fully licensed, he would have deferred to legal counsel, but would most like not award points if a provider is not licensed in a county for which it proposes service.
9. Mr. Breckerink identified additional area in the evaluation wherein he believes errors were made, the correction of which would result in an adjustment of the award of points. For example, in evaluating plan benefits, the evaluator gave Pru-Care 20 points when only 10 points are available for award without a demonstration of additional services. For emergency room availability, Cigna was awarded 5 points when it should have received 10. In the area of co-payments, Cigna was awarded 13 points and should have received 23. Concerning range of providers, Cigna’s proposal lists seven hospitals yet the evaluation form only reflects six, resulting in a shortage of 10 points. As to turn around time, Cigna indicated it would accomplish payment in 60 days whereas Pru-Care indicated it would in “an average” of two weeks. As a result, Mr. Breckerink, who points out Cigna’s actual time is 30 days and it therefore should have been given 30 points, contends there is no opportunity for a valid comparison here since Pru-Care’s answer is not responsive to the RFP’s call for an “expected” time. His point is well taken.
10. With regard to accessibility, Mr. Breckerink states that Cigna got only 20 points for its two allowed counties but should have received 30 points since it has hospitals in three counties in the service area. DOA’s rationale on this point is identical to that on the issue of full counties served. He also alleges that Cigna was shortchanged by at least 2 points on the number of counties in which specialty providers are represented and by at least 1 point on the number of providers.
11. Mr. Nye admits Pru-Care should have received 5 points instead of 10 for benefits. This would reduce its’ raw score in this area from 258 to 253 points. Nye contends, however, that the points awarded Pru-Care for its’ turn around time were correct. He does not consider the question to be a bad once since it was asked equally of all providers and each responded as it saw fit realizing that its response might become a part of a contractual obligation. This reasoning is specious at best and does not address the real question of the fairness and appropriateness of the question asked. Further, Mr. Nye also admitted that under certain circumstances, if Pru-Care were to lose credit for those two counties in which it was not shown to be licensed, the change could result in a difference sufficient to reverse the relative standings of Pru-Care and Cigna.
12. Mr. Breckerink alleges, and Mr. Nye admits that multiple *231computer runs were made utilizing the raw scores developed by the evaluators before the recommendation as to award was forwarded to the Secretary. On the first rim for the second evaluation, Cigna was in second place with a point total of 71.1 and Pru-Care was third with 65.86 points. On the second run, which Nye contends was done to make the computer run consistent with what had been said at the prebid conference and in the RFP, Cigna dropped from second place to third place with 58.2 points and Pru-Care went from third to fourth with 57.195 points. In the third run, which ultimately formed the basis for the award, the positions of Cigna and Pru-Care reversed with Cigna dropped to 56.83 points and Pru-Care rising to 57.425. Central Florida Physicians remained in first and Health Options in second. When Central Florida Physicians dropped out due to insolvency, Health Options became number one and the other two each went up one place in the standings without changing relative positions.
13. According to Mr. Breckerink when the mistakes were identified and changes made in the raw scores, Cigna got a total of 23 more points but Pru-Care still got 16 more points than it should have. He contends that if the mistakes were accurately corrected, if Cigna were to get all the points it should and Pru-Care lose all it should not legitimately have, Cigna would come out higher in the overall ranking than Pru-Care. However, he admits there are factors involved about which he does not know which may affect the standings. What is clear is that while Mr. Breckerink could not clearly follow the evaluation procedure, neither can others charged with evaluating it.
14. What is more, notwithstanding the direction given in the objectives of the procurement that only two providers be awarded contracts, the department continuously has been unable to abide by this guidelines. In its September 11, 1987 recommendation after the first evaluation sequence, Mr. Nye recommended, for the Orlando service area, awards to Central Physicians, Cigna, and Pru-Care for a part of the service area and an additional award to Health Options and Florida Health Care for other counties in the service area. When the Secretary directed the objective second evaluation, no change was made to the number of providers to be recommended (two), but again, on October 6, 1987, Mr. Nye recommended three providers, Central Florida Physicians, Health Options, and Cigna. No evidence was presented as to why this recommendation was not implemented, but it is seen that on October 26, 1987, Mr. Nye submitted his third set of recommendations to the Secretary, this time recommending only Central Florida Physicians, and Health Options. Being still unable to finalize the process, on October 30, 1987, Mr. Nye submitted his fourth set of *232recommendations to the Secretary recommending, for the most part, three providers, but specifically recommending Pru-Care for award in Lake and Volusia counties, where it was arguably not even licensed. No justification or explanation for this vacillation was forthcoming from the Department and the exercise appears to have been clearly capricious.
CONCLUSIONS OF LAW
The Division of Administrative Hearings has jurisdiction over the parties and subject matter in this case. Section 120.57(1), Florida Statutes.
The threshold question in this proceeding is the applicability of Chapter 287, Florida Statutes, to the procurement in question. Prior to the Legislature’s enactment of Chapter 87-156, Laws of Florida, Section 110.123(d), Florida Statutes, the pertinent statute relevant to state employee membership in HMOs, provided that any employee eligible to participate in the state group health insurance plan may, instead, elect to participate in a HMO. By enacting Chapter 87-156, the Legislature amended 110.123(d) to add that, effective January 1, 1988, the Department of Administration should contract with HMO’s to participate in the state plan through the competitive bid process based on cost, service area, plan benefits, and accessibility.
DOA determined to solicit competitive proposals pursuant to Chapter 287, Florida Statutes, which, at subsection 287.057 authorizes the use of competitive sealed proposals if the agency determines that the use of sealed bids is not practicable. The term “request for proposals” is defined in Section 287.012(11), Florida Statutes. Section 287.057(16) provides for evaluation of proposals by:
A selection team of at least three employees who have experience and knowledge in the program areas and service requirements for which contractual services are sought shall be appointed by the agency head to aid in the selection of contractors.
DOA failed to make a written determination that the use of competitive sealed bidding was not practicable. In this case, however, that omission is not considered to be of sufficient magnitude or gravity to effect the validity of the process itself.
Cigna contends that DOA’s evaluation of the proposals submitted was arbitrary and capricious and bore no rational relationship to the statutory and public policy objectives of the procurement or to the objectives or criteria stated in the RFP. Several factors exist which tend to lend credence to Cigna’s contention.
*233It was established through the testimony of Mr. Nye that the evaluators, for the past part personnel technicians in his Division, were poorly trained and not familiar with the HMO concept, the terms of the RFP, or the evaluation form or process they were to use. In addition, examination of the evaluation forms indicates that the questions thereon were, in many instances, poorly framed and not clearly defined, and failed to give the evaluators a fair and clear indication of what they were to evaluate.
There was substantial testimony to indicate that as a result of the two above-cited factors, numerous arithmetic and interpretive errors were made by the evaluators in the course of their consideration of each of the proposals and while Mr. Nye may contend in good conscience that even if corrected, the net change would be insufficient to result in a relative change of position among the proposers, this is not sufficiently established by evidence of record. The fact that slightly over one half of one point differentiated the successful provider, PruCare from the unsuccessful Petitioner, Cigna, demands that clear and unequivocal evidence be available to do away with any reasonable suspicion of error. This is so, particularly in light of Mr. Nye’s admission that a removal of Lake and Volusia counties from the award to Pru-Care could well reverse the relative standings between Pru-Care and Cigna.
Finally, taken as a whole, the evaluation procedure appears tainted. In addition to the above-cited factors, it is clearly seen that the Secretary rejected all recommendations and evaluations after the first series of evaluations returning the package to Mr. Nye for accomplishment of a new evaluation based on different standards. According to Mr. Nye, even on the second evaluation, three separate computer runs had to be accomplished in order to come up with a reasonably accurate depiction of the evaluation process. Even that is in question. In addition, notwithstanding the provision in the policy guidelines for this procurement that only two contracts be awarded in each service area, the fact that four separate recommendations were made to the Secretary, only one of which contained no more than two recommended providers per service area, would tend to indicate that the Department ignored its own policies and guidelines throughout the evaluation and award process.
It is well settled that contract award decisions made by state agencies will ordinarily be upheld unless the decision is arbitrary, capricious, or beyond the scope of agency discretion. Systems Development Corp. v Department of Health and Rehabilitative Services, 423 *234So.2d 433 (Fla. 1DCA 1982); Capeletti Brothers v State Department of General Services, 432 So.2d 1359 (Fla. 1DCA 1983). In Systems Development, the court stated:
We are constrained to review the agency’s decision under these circumstances only so far as to determine whether the decision was arbitrary, capricious, or beyond the scope of its discretion, which discretion is very broad:
So long as the public agency acts in good faith, even though they may reach a conclusion on facts upon which reasonable men may differ, the courts will not generally interfere with their judgment even though the decision reached may appear to some persons to be erroneous. Volume Services v Canteen Corp., 369 So.2d 391, 395 (Fla. 2DCA 1979).
In this case, the question is not so much whether the agency’s decision appears to be erroneous but that the decision making process was exercised in an arbitrary and capricious fashion. For the reasons previously stated, this factor is considered to have rendered invalid the agency’s proposed award decision.
RECOMMENDATION
In view of the foregoing, it is, therefore:
RECOMMENDED that the Department of Administration issue a Final Order rejecting all proposals submitted for the Orlando service area and readvertise for new proposals if deemed appropriate.
RECOMMENDED this 19th day of April, 1988 at Tallahassee, Florida.