STATE OF FLORIDA

DIVISION OF ADMINISTRATIVE HEARINGS

CALIPER SYSTEMS, INC., d/b/a CALIPER CORPORATION,

Petitioner,

vs.

DEPARTMENT OF TRANSPORTATION,

Case No. 18-0384BID

Respondent,

and

PTV AMERICA, INC.,

Intervenor.

/

RECOMMENDED ORDER

On February 26 and 27, 2018, Robert E. Meale, Administrative Law Judge of the Division of Administrative Hearings (DOAH), conducted the final hearing in Tallahassee, Florida.

APPEARANCES

For Petitioner: Frederick John Springer, Esquire

Elizabeth W. Neiberger, Esquire Bryant Miller Olive P.A.

101 North Monroe Street, Suite 900 Tallahassee, Florida 32301

For Respondent: Douglas Dell Dolan, Esquire

Department of Transportation

605 Suwannee Street, Mail Station 58

Tallahassee, Florida 32399-0450

For Intervenor: Bryan Duke, Esquire

Messer Caparello, P.A. 2618 Centennial Place

Tallahassee, Florida 32308 STATEMENT OF THE ISSUE

The issue is whether, due to the nonresponsiveness or misscoring of Intervenor's proposal, Respondent's intent to award a contract to Intervenor based on its proposal submitted in response to a request for proposals known as Florida Travel Demand Modeling Software and License (RFP) is contrary to

the governing statutes, rules or policies, or the RFP specifications, as provided by section 120.57(3)(f), Florida

Statutes.

PRELIMINARY STATEMENT

In November 2017, Respondent published the RFP. Respondent received three proposals by the stated deadline of December 7, 2017. On December 20, 2017, Respondent posted the proposed tabulation selecting Intervenor's proposal. On December 26, 2017, Petitioner filed a notice of intent to protest, and, on January 5, 2018, Petitioner filed a Formal Written Protest and required security.

The Formal Written Protest alleges generally that Intervenor's proposal was not responsive because it fails to comply with requirements of the RFP and that Respondent assigned an excessively high technical score to Intervenor's proposal.

For relief, the Formal Written Protest seeks a final order awarding the contract to Petitioner or rejecting all proposals and readvertising the RFP.

Respondent transmitted the file to DOAH on January 22, 2018. On January 25, 2018, the Administrative Law Judge issued a Notice of Hearing for February 12 through 14, 2018. On February 1, 2018, Petitioner filed an Unopposed Motion for Continuance of the Final Hearing, which sought the rescheduling of the final hearing for any two days during the week of February 26, 2018. By Order entered on the same date, the Administrative Law Judge granted the motion and reset the hearing for February 26 and 27, 2018.

At the final hearing, Petitioner called one witness and offered into evidence eight exhibits: Petitioner Exhibits 1, 2, and 6 through 11. Respondent called four witnesses and offered into evidence no exhibits. Intervenor called no witnesses and offered into evidence no exhibits. The parties jointly offered Joint Exhibits 1 through 13. All exhibits were admitted for all purposes, except Petitioner Exhibit 6, which was proffered, and Petitioner Exhibits 1 and 2, which were admitted, but not for the truth.

The court reporter filed the transcript on March 15, 2018. The parties filed their proposed recommended orders by March 23, 2018. On March 26, 2018, Petitioner filed an Unopposed Motion

to Reopen the Record to Supplement Joint Exhibit 1, which is

granted.

FINDINGS OF FACT

1. RFP and Proposals

   
   

   1. In November 2017, Respondent published the RFP. The RFP is divided into parts, including Special Conditions, Scope of Services, Price Proposal Form, and Introduction, which, according to Special Condition 36, are to be interpreted in this order in the event of conflicting provisions. The purpose of the RFP is to procure travel demand modeling software, which projects future service demands on a transportation system, so that transportation planners, engineers, and policymakers can design, schedule, prioritize, and budget transportation projects and expenditures.

   2. The Price Proposal Form is the first page of the RFP. It contains four columns to be completed by the proposer with dollar figures for year 1, year 2, year 3, and 3-year total. The Price Proposal Form contains five rows for the following prices: "Model Conversions," "Training," "Annual License Renewal," "Base Software Cost," and "OVERALL PRICE."

   3. The next part of the RFP is the Introduction.

      
      

      Introduction 1 invites interested persons to submit proposals "to provide travel demand modeling software and licensing in Florida for [Respondent], MPOs [Metropolitan Planning

      
      

      Organizations], local agencies and universities (teaching only)." The boldface language alerts prospective proposers that, although Respondent is conducting the procurement, the MPOs, local agencies, and universities in their academic capacity will be co-licensees with Respondent.

   4. Introduction 1 states that Respondent "intends to award this contract to the responsive and responsible Proposer whose proposal is determined to be most advantageous" to Respondent. Introduction 1 states that the estimated term of the contract is three years.

   5. Special Condition 1 warns that a proposer will be considered nonresponsive unless it is registered with the myfloridamarketplace system by the scheduled date for the opening of technical proposals. Special Condition 6 incorporates the Scope of Services. Special Condition 7 states that Respondent intends to award the contract to the "responsive and responsible vendor with the highest cumulative total points for the evaluation criteria." Special Condition 20 warns that a proposer may not apply "conditions . . . to any aspect of the RFP," and the placement of such conditions "may result in the proposal being rejected as a conditional proposal (see "RESPONSIVENESS OF PROPOSALS")."

   6. Special Condition 21 is "Responsiveness of Proposals." Special Condition 21.1 states that a:

      
      

      responsive proposal is an offer to perform the scope of services called for in this [RFP] in accordance with all requirements of this [RFP] and receiving [70] points or more on the Technical Proposal. Proposals found to be non-responsive shall not be considered. Proposals may be rejected if found to be irregular or not in conformance with the requirements and instructions herein contained. A proposal may be found to be irregular or non-responsive by reasons that include . . . failure to utilize or complete prescribed forms, conditional proposals, incomplete proposals, indefinite or ambiguous proposals, and improper and/or undated signatures.

      
      

   7. Special Condition 22.1 calls for each proposer to submit, each in its own sealed package, a Technical Proposal and a Price Proposal. Special Condition 22.2 requires that the Technical Proposal be divided into six scored sections and

      30 unscored subsections; the six scored sections comprising five technical sections and one price section. The six scored sections are the six main sections of the Scope of Services, which is discussed below. Special Condition 22.4 states that "Technical Proposals should not exceed 30 pages in total."

   8. Special Condition 30 requires at least three evaluators with suitable experience and knowledge. Each evaluator will independently score each proposal, and the Procurement Officer will average the scores for each Proposer. During the evaluation process, the Procurement Officer is to examine the

      
      

      proposals for responsiveness and "automatically reject . . ." those that the officer finds are nonresponsive.

   9. Special Condition 30.2 explains that the technical evaluation "is the process of reviewing the Proposer's response to evaluate the experience, qualifications, and capabilities of the proposers to provide the desired services and assure a quality product." For the five technical sections making up the Technical Proposal, Special Condition 30.2.a assigns a maximum of 90 points, as follows:

      General Platform Capabilities	25	points
      Network	20	points
      Hardware Requirements and Options	10	points
      Development and Advanced Options	10	points
      Other Considerations	25	points

      
      

      These five sections are, respectively, Scope of Services 2, 3, 4, 5, and 7.

   10. Special Condition 30.2 states that, in evaluating the Technical Proposals, each evaluator is to use the following scale in assigning a single score for each section:

       Exceeds Reply fully meets all 4

       Expectations specifications and offers

       innovative solutions to meet specifications. Reply exceeds minimum specifica- tions and provisions in most aspects for the specific items.

       Meets Reply adequately meets the 3

       Expectations minimum described need, or

       provisions of the specific needs and is generally

       
       

       capable of meeting [Respondent's] needs for specific items.

       
       

       Partially Reply does not fully 2

       Meets address the need, one or Expectations more major considerations

       are not addressed, or is so limited that it results in a low degree of confidence in the [proposal]. Reply is lacking in some essential aspects for the specific items.

       
       

       Does Not Meet Reply fails to address the 1

       Expectations need, or it does not

       describe any experience related to the component. Reply is inadequate in most basic specifications or provisions for the specific items. Insufficient information provided to be evaluated.

       
       

   11. For the Price Structure, Special Condition 30.2.b states that the lowest Price Proposal earns 10 points and the other Price Proposals receive points based on a formula in which

       10 is multiplied by a fraction whose numerator is the lowest Price Proposal and whose denominator is the price of the subject Price Proposal. Thus, a Price Proposal with the lowest price

       $100,000 would earn 10 points, and a proposal with a price of

       
       

       $120,000 would earn 8.33 points ($100,000/$120,000 x 10).

       
       

   12. Scope of Services 1 notes that the Scope of Services is the product of input from the Florida Model Task Force (FMTF), which comprises members of the Florida modeling

       
       

       community. Scope of Services 1 describes the objective of the procurement:

       [Respondent] has for more than three decades promoted a unified statewide modeling approach for consistency to the application of engineering and planning travel demand modeling activities. As part of this effort [Respondent] makes available a common modeling software platform for use by all public agencies in Florida which includes [Respondent], . . . MPOs, County and City Governments and Regional Planning Councils. Additionally, Florida universities are provided a limited teaching license for teaching and research purposes.

       
       

       [Respondent] seeks to . . . select a travel demand software package and license for the purpose of meeting the stated objective of providing a common modeling platform. This platform is intended to support modeling activities in the state and represent the Florida-specific standardized modeling procedures outlined in the Florida Standard Urban Transportation Model Structure (FSUTMS).

       
       

       * * *

       
       

       This scope of services represents input from the Florida Model Task Force (MTF)[,] . . . whose mission is to advance model development and applications to serve the transportation planning needs of [Respondent], MPOs and local governments.

       The input from the Florida MTF serves as a guide for developing the model platform scope.

       
       

   13. No one challenged the specifications of the RFP. Proposals were submitted timely by Intervenor, Petitioner, and Citilabs, Inc., which is the present vendor of Respondent's

       
       

       travel demand modeling software. The Procurement Officer examined each proposal to ensure that it contained a Technical Proposal and a Price Proposal and determined that each Proposer was properly registered to do business in Florida. Without undertaking further analysis of responsiveness, the Procurement Officer distributed the proposals to the evaluators for scoring, assuming that any failure to meet RFP mandatories would result in a lower score.

   14. For the Price Proposals, Citilabs submitted the lowest price, which was $96,000, so it received 10 points. Petitioner submitted a price of $180,000, so it received 5.33 points. Intervenor submitted a price of $260,000, so it received 3.69 points. These scores are not at issue.

   15. For the Technical Proposals, Intervenor received 83.33 points, Petitioner received 78.75 points, and Citilabs received

       73.33 points. Thus, Intervenor received 87.03 points, Petitioner received 84.08 points, and Citilabs received 83.33 points. On December 20, 2017, Respondent published a notice of intent to award the contract to Intervenor. The intended award was protested by Petitioner, but not Citilabs.

2. Responsiveness

   
   

   1. Introduction

      
      

      1. The Procurement Officer's responsiveness review never went beyond a determination that each proposer was registered to

         
         

         do business in Florida and each proposal contained a Technical Proposal and a Price Proposal. None of the evaluators conducted any examination of the proposals for responsiveness or reduced any score of Intervenor for the two instances of nonresponsiveness discussed in this section of the recommended order. In order to apply the deferential standards discussed in the Conclusions of Law, it is necessary to deem that Respondent determined that Intervenor's proposal is responsive on the two issues discussed immediately below.

      2. Although the RFP could have more clearly presented its mandatories by setting them out separately, its failure to do so is irrelevant. Dispersed through the RFP are numerous requirements imposed upon a proposal that, if ignored or violated, would render the proposal nonresponsive. The items discussed in this section of the recommended order are mandatories in the RFP.

      3. In its proposed recommended order, Petitioner claims that Intervenor's proposal is nonresponsive in model conversions and special access to the software.

   2. Conversions of ABMs and Timeframes for Conversions of All 13 Models

      
      

      1. Except for three provisions, the RFP could easily be misconstrued to call for the submittal of travel demand modeling software on a platform that might or might not accommodate the

         
         

         platforms, and thus the travel demand modeling software, presently used by Respondent, the MPOs, and local agencies. The first of these exceptions is in the Price Proposal Form. The first of only four price categories in the Price Proposal Form is "Model Conversions," a prominent two-word reference that stands without explanation or context, although the plural form alerts the proposer to the need to price more than one conversion.

      2. Nearly as laconic, Scope of Services 7.3.2 requires each proposer to "outline a plan for implementation of the software and/or software updates." An understandably puzzled proposer asked, "Is this about conversion plan for [Respondent] or general software update plan as a whole?" Failing to seize upon the opportunity to elaborate on conversion requirements, in Addendum No. 1, Respondent replied only, "The intent was to form a conversion plan."

      3. In Scope of Services 6, Respondent abandons its reticence and describes the conversion responsibilities in reasonable detail. As noted above, Scope of Services 6 is Price Structure, which describes each of the four price components included in the Price Proposal Form or Price Proposal.

      4. In its proposed recommended order, Respondent argued that responsiveness requirements for the Technical Proposal may not be culled from the portion of the RFP detailing the Price

         
         

         Proposal. Given the failure of the remainder of the RFP to detail conversion requirements, Respondent's argument is burdened by the fact that, if the argument were to prevail, Respondent would be deprived of the only provisions anywhere in the RFP to enforce important conversion responsibilities undertaken by the ultimate vendor. But Respondent's argument finds no support in the RFP itself.

      5. Scope of Services 6.1 addresses model conversions as

         
         

         follows:

         
         

         It is the mission of the [FMTF] that every travel forecasting model in Florida operates from the same software platform. These models are validated to standards established by the [FMTF]. The Vendor is expected to convert these models to the selected platform such that the converted models are provided as validated models. A timeframe and conversion methodology is required. While conversions are not expected to precisely meet the outputs of the original model, they are required to meet validation standards consistent with guidelines established through National Cooperative Highway Research Program (NCHRP) Report 716 and other resources identified on the FSUTMSOOnline.net modeling website.

         Specific requirements will also include recoding ancillary modeling scripts into the selected platform or to a more common, standardized programming language such as Python.

         
         

         Updates to socioeconomic data inputs, local travel demand variables and network coding are not required through this RFP.

         
         

         The vendor must provide a cost estimate for the conversion of seven (7) 4-step models

         
         

         (Florida Statewide Model, Florida Turnpike Model, Northwest, Capital Region, Gainesville, DS, and D1); four (4) ABM [activity-based models) models (Southeast, Tampa Bay, Northeast and Treasure Coast); and two (2) training models.

         
         

      6. Scope of Services 6.1 not only informs proposers what they need to include in their cost projections for Model Conversions, but, in so doing, also informs them of their obligation to convert Respondent's Citilabs model, ten local models, and two training models. Except for Scope of Services 6.1, the requirements of the RFP, as distinct from the mission statements contained in Scope of Services 1, might be misinterpreted as specifications for the procurement for Respondent of a travel demand modeling software on a platform whose compatibility with the platform presently used by Respondent and platforms presently used by the MPOs and local agencies is irrelevant.

      7. Most importantly, Respondent's argument ignores Special Condition 21.1, which identifies the entire RFP as a source of mandatories. Without regard to Special Condition 21.1, Special Condition 22.2 lists Scope of Services 6 within the Technical Proposal, which, Respondent would concede, is an obvious source of mandatories. Scope of Services 6 is merely the fifth of six sections to be scored by the evaluator.

         
         

         Respondent's argument to disregard Scope of Services 6 as a source of mandatories is a misreading of the RFP.

      8. Intervenor's proposal, which refers to its traffic demand modeling software as "Visum," responds to Scope of Services 7.3.2 by proposing to convert Respondent's present Citilabs model, but not all of the models currently used by the MPOs and local agencies:

         We understand that successful model conversion only can be achieved through a collaborative relationship in between [sic] [Respondent] (and affiliated agencies), local consultants, and the software provider. Therefore, we propose a process that all three parties can contribute to this process and ensure all local modeling and software expertise can be fully utilized for this process. The overall conversion process is divided into four tasks below:

         
         

         1. Kick-off meeting with [Respondent's] Central office: First, we will work with [Respondent's] Central office to come up with a set of basic templates which will be applicable to four-step models as well as ABM models. In this way, we can come up with set standard that can be applied to all models that need to be converted and/or new models that need to be developed in the future. Details on model conversion schedule and prioritization of each model will be discussed and decided based on required model update (for LRTP) schedule and similarities of models.

            
            

         2. Kick-off meeting with [Respondent's] District office(s): Based on priority list provided from previous step, we will set up individual kick-off meetings with each district. We expect to meet with local model coordinators as well as local

            
            

            consultants with local modeling knowledge (up to two consultants selected by [Respondent]) to learn about the model that needs to be converted. This will give us a background on special features of the existing models, expected run-time, memory requirements and current shortcomings. All data and documentation necessary for model conversion should be provided at the meeting so that it can be reviewed by conversion team. At the end of the meeting, conversion team will come up with initial model conversion plan and shared [sic] with model coordinator and invited consultants.

            
            

         3. Basic Model Conversion: Basic components in the existing model will be converted to Visum by [Intervenor] at no additional cost. This conversion includes network (traffic and transit) conversion for the base year model, 4-step procedures, trip tables, and any special scripts used in the current model (to model trip adjustments, special assignments, skim averaging, etc.). In case of models integrated with third- party ABM, we will provide network (traffic and transit) conversion for the base year model, assignment and skimming procedures, and scripts necessary for the ABM interface on the Visum side (any modifications required for the ABM side, i.e., code within the ABM is beyond the scope of the basic conversion process).

            
            

            Once the basic model conversion is completed, we will host a hand-over meeting to the model coordinator and selected local consultant (e.g. on-call consultant). At the meeting, we will present the process that was undertaken and detailed information on new attributes, calculations and overall model operation. We will also provide model conversion report so that [Respondent] and consultants can use it to understand converted model.

            
            

         4. Model fine-tuning and final delivery: [Intervenor] will take the lead along with [Respondent] model coordinator (or selected consultant with local knowledge) on this final model fine-tuning process that includes calibration and validation of the 4-step models along with [Intervenor]. The calibration and validation will be conducted based on guidelines/standards provided on NCHRP Report 716. For the ABM interface, the local consultant is expected to re-

            write/modify the code with the ABM system in order to successfully interface it with Visum ([Intervenor] will provide full support on the Visum side required in this process.) As a software expert, [Intervenor] will support [Respondent] model coordinator (or selected consultant), local model expert, to complete fine-tuning and localization process and attend meetings (as necessary) to provide continuous feedback.

            
            

      9. By contrast, Petitioner's proposal responds to Scope of Services 7.3.2 with an unconditional undertaking to convert, not just Respondent's Citilabs model and local nonABMs, but also local ABMs:

         In this section, we present our approach to and time frame for the model conversions.

         Quite obviously model conversions are the principal obstacle to a successful transition to new travel demand modeling software. We will not be taking on this task from scratch, as we have already converted a number of current Florida models and, upon selection, would aggressively ramp up the model conversion efforts.

         
         

         No one has more experience in converting models from Citilabs software to another platform than we do, as we have been doing it for more than two decades. Recently we converted the NFTPO [North Florida Transportation Planning Organization]

         
         

         activity-based model to run on TransCAD. In the process, we improved the models in several respects. First, we replaced the stick road network with an accurate HERE network that was already licensed. We then recreated the transit network so that the buses run on the correct streets in the road network. In doing so, we also fixed errors in both networks. We also identified and fixed a variety of errors in the model scripts and significantly reduced the run times for both models. We also converted the statewide model and the Olympus training model as part of the aborted ITN process. [The "aborted ITN process" refers to an earlier, unsuccessful effort by Respondent to procure the subject software by an invitation to negotiate.]

         
         

         At the outset of the conversion process, we will meet with the stakeholders for each model to be converted to understand their priorities and preferences and to develop a mutually acceptable approach to the model conversion. We will welcome the participation of involved consultants as well as agency managers in these discussions.

         
         

         We will use templates for FSUTMS in TransCAD to facilitate the conversion process. These will consist of a standard flowchart interface and the identification of the specific macro functions to be used for trip generation, trip distribution, model choice, and assignment. Highly experienced staff will then perform the conversions and test the results to ensure a successful outcome. Significant discrepancies will be investigated and resolved in a technically proficient manner, consulting with agency representatives if errors are found that need to be corrected. Each and every conversion will ensure that similar results are obtained, may at the option of each model stakeholder have obvious scripting errors corrected, and will improve upon

         
         

         validation measures and run much faster than the current Cube version model.

         
         

         Each conversion will be accompanied by a technical memorandum detailed the conversion effort, changes made, and validation achieved. The conversion effort will be further strengthened and memorialized in the creation of standard scripts for FSUTMS in TransCAD, which will be published and shared with users statewide.

         
         

         We estimate that we will be able to complete all the conversions in a 6- to 12-month time frame. Based on our prior experience, we know that different agencies will have different timetables for this work, and we intend to work with [Respondent] and other model stakeholders to schedule the work effort to reflect these schedules and [Respondent] priorities.

         
         

         We will be mindful of the improvement and standardization opportunities afforded by the conversion effort and will work close with [Respondent] and MPO staff to incorporate some upgrades to the models as part of the process.

         
         

      10. Upon close analysis, the promise of kick-offs featured in Intervenor's proposal fade to a more prosaic element of the kicking game, as Intervenor fails to convert and punts its responsibilities to Respondent, local agencies, and even unspecified private consultants. In three ways, Intervenor's proposal comes up short as to conversion, so as to deprive Respondent of much of the benefit of the bargain that is the purpose of the procurement.

          
          

      11. First, Intervenor's proposal does not undertake the conversion of the four travel demand ABM models, which include the heavily populated areas of southeast Florida and Tampa Bay. Instead, Intervenor shifts the responsibility for converting the ABMs, so as to enable them to interface with Visum, to local consultants who are, in the RFP, third-party beneficiaries of the procurement, not the vendor or its subcontractors. Intervenor's unwillingness to convert the ABMs evidences the difficulty of converting this type of model, as borne out by Petitioner's proposal. Petitioner has considerable experience converting Citilabs' travel demand modeling software, so Petitioner's conversion of Respondent's Citilabs model, which Intervenor also has agreed to do, should not be difficult; the open-ended timeframe to which Petitioner committed for converting all of the models--6 to 12 months--likely reflects the difficulty of converting the ABMs, which Intervenor has expressly declined to do.

      12. Second, Intervenor fails adequately to describe exactly what it will undertake as to the conversion of ABMs. For these four models, including two with very large service bases, the last sentence of the above-quoted excerpt from Intervenor's proposal offers only Intervenor's support of the "localization" efforts of other parties. Failing to define "localization," Intervenor nonetheless has made it clear that it

          
          

          does not accept the RFP requirement that it convert the four ABMs. To this requirement, Intervenor has attached a condition that relieves Intervenor of the responsibility for the final step or steps necessary for local agencies' travel demand models, which will share the new platform of Respondent's software, actually to work. By so doing, Intervenor has declined unconditionally to assume the daunting tasks of calibration, in which each model is adjusted to force results that match real-world conditions, and validation, in which the model is tested by performing a model run for an historic period, for which the actual data is known, to confirm that the model's output compares favorably to actual results--although, as described in Scope of Services 6.1, quoted above, validation in this RFP also may mean the ability of the model to reproduce the outputs of the model that it is replacing.

      13. Third, Intervenor's proposal does not contain the required timeline for the conversion work that Intervenor has undertaken to perform. Intervenor has not imposed upon itself the required timeline for any of the 13 models required to be converted. The materiality of this omission is underscored by Petitioner's warning, "Quite obviously model conversions are the principal obstacle to a successful transition to new travel demand modeling software."

          
          

      14. Intervenor's nonresponsiveness to the conversion requirements in Scope of Services 6.1 and 7.3.2 confers upon Intervenor a competitive advantage. Conversion, calibration, and validation of the 13 travel demand models are time- consuming, expensive processes, which are at the core of the services for which Respondent is paying in this procurement, so that a proposal that incompletely undertakes these responsibilities confers upon the proposer a significant competitive advantage. Intervenor has also undermined Respondent's ability to enforce the contract in case of incomplete work by shifting to Respondent and private consultants the final stages of the conversion of the ABMs and omitting a timeframe within which to complete any of the 13 conversions.

   3. Access as a Co-Licensee for Universities in their Teaching Capacity and Affordable Access for Universities as Consultants and Private Consultants

   
   

   1. Petitioner argued in its proposed recommended order that Intervenor's proposal is nonresponsive due to inadequacies in its undertaking to provide access to the travel demand modeling software for universities and certain private modeling consultants. As the heading indicates, there are two distinct aspects to this challenge.

      
      

   2. Scope of Services 7.4 provides:

      
      

      While [Respondent] makes the modeling software available to other public agencies (and Universities acquire no-cost teaching licenses), selection of the software will consider the costs to private industry working in Florida. Private industries and Universities work in collaboration with [Respondent] and Florida's public agencies. It is important to ensure that these industries, particularly smaller firms, have affordable access to the selected software.

      
      

   3. In Addendum No. 1, Respondent responded to a vendor's question of how and where to present pricing information pertaining to the specifications contained in Scope of Services 7.4. Respondent replied: "Please present a price, a discount, or your approach as to how these entities will have affordable access to the selected software in section 7.4."

   4. Intervenor's proposal responds to Scope of Services 7.4 as follows:

      [Intervenor] has been providing a separate pricing structure for academic users.

      First, all academic users in Florida will get access to not only Visum licenses as a part of this contract but also, for each semester, they will be eligible for additional classroom licenses for up to 60 students per request. If they would like to acquire separate licenses, they will be eligible for academic pricing where we provide all four off-line software that [Intervenor] provides.

      
      

      For smaller firms in Florida, we will apply maximum multiple license discount (50%) from first license; however, we will require them to submit Florida DBE [Disadvantaged

      
      

      Business Enterprise] certification to ensure their eligibility. In addition, we will offer a lease-to-own option as well as making Visum license to be even more affordable to them. Leased licenses will be fully functional with an expiration date.

      Upon expiration, user will be able to choose whether they would like to purchase a license and the full amount that they have paid until then (within 1-year) will be applied as a credit toward their purchase.

      In this way, we can provide affordable access to users with smaller companies.

      
      

   5. Petitioner's proposal, which refers to its travel demand modeling software as "TransCAD" and its traffic simulator software as TransDNA and TransModeler, responds as follows:

      Our offer will actually lower the cost to Florida consultants and university researchers. Many, of course, already have our software and will not need to acquire additional licenses. For those that will need licenses, we will provide TransCAD free of charge, but expect that the normal annual support fee of $1,200 be paid up front to receive the software. We will limit this offer to two copies per consulting firm for use in Florida and for work performed for Florida public agencies. Similarly, we will offer one optional TransModeler license to Florida consultants and university researchers for work performed in Florida for free but with the normal annual support fee of $1,500 per year to be paid in advance.

      
      

   6. Intervenor's proposal is nonresponsive in two respects. First, Scope of Services 7.4 clearly identifies as co-licensees local public agencies and universities in their teaching capacity. This is consistent with Introduction 1,

      
      

      which, as noted above, alerts in boldface that Respondent is acquiring the software and license for itself, the MPOs, local agencies, and universities in their teaching capacity. The university's teaching of traffic demand modeling is not feasible if only the professor were to be entitled to a free copy of the software, which students would be required to purchase at a cost of tens of thousands of dollars per copy. Attaching an impermissible condition to the requirement to treat the university in its teaching capacity as a co-licensee, Intervenor's proposal limits the free student copies to 60 per semester and offers additional student copies at an unspecified academic discount. Thus, Intervenor's proposal is nonresponsive to Scope of Services 7.4 and the Introduction in its treatment of universities in their teaching capacity as a co-licensee.

   7. As to Scope of Services 7.4, Petitioner's proposal is also nonresponsive because it imposes substantial "annual support fees" on all "free" university licenses--even though the above-quoted Price Proposal Form clearly includes the price of the "Annual License Renewal" for three years. Additionally, Petitioner's proposal fails to provide any free copies of the software for students.

   8. Second, regardless of whether they are private entities or universities, consultants, who are not co-licensees, are assured by Scope of Services 7.4 affordable access to the

      
      

      software. This assurance does not impose much of a burden upon a proposer. As amplified by Respondent's response to the second question in Addendum No. 1, each proposal was required to "present a price, a discount, or your approach as to how these entities will have affordable access to the selected software in section 7.4." Contrary to Petitioner's contention, a discount without a price against which to apply the discount is facially sufficient, so Intervenor's proposal is responsive to this requirement.

   9. However, Intervenor's proposal is nonresponsive because Intervenor inexplicably failed to offer its vague promise of preferential pricing to the class of users to whom Scope of Services 7.4 assures affordable access. Rather than extend its discount to all private and university consultants, Intervenor's proposal limits its discount to private consultants that are certified as DBEs, which is likely a small fraction of private consultants and, of course, improperly ignores all universities in their capacity as consultants.

   10. Intervenor's nonresponsiveness to these requirements confers upon Intervenor a competitive advantage. The advantage from failing to treat the universities in their teaching capacity as co-licensees means that every dollar exacted from students or universities in their teaching capacity for the term of the RFP is unearned because Respondent has already paid for

       
       

       these licensing rights in this procurement. The advantage from extending a discount to a small fraction of the class of persons entitled to the discount means that Intervenor will improperly realize thousands of dollars on the sale of undiscounted software to consultants that are not DBEs.

3. Scoring

A. Introduction

1. The evaluators were T. Hill, T. Corkery, and

   
   

   1. Tabatabee (respectively, Evaluator 1, Evaluator 2, and Evaluator 3). The evaluators were not trained in the RFP, and they did not communicate with each other while scoring the three proposals. The evaluators worked briskly, completing their evaluations within two weeks.

2. Evaluator 1 has been Respondent's state modeling manager for the past five years and has prior experience with Respondent in transportation modeling in a district office. He has a total of 18 years' experience in transportation modeling. Evaluator 2 has been employed by Respondent for 25 years. He is presently a senior travel demand modeler, in which capacity he has served for ten years. Evaluator 2 previously served as a transportation modeler for Respondent. Prior to his employment with Respondent, Evaluator 2 worked as a travel demand modeling consultant for seven years. Evaluator 3 lacks experience in modeling, but instead is experienced in statistics and the

   
   

   development of Respondent's traffic data system, which supplies the data used for traffic modeling.

3. As noted above, none of the evaluators lowered a score of Intervenor's proposal due to its nonresponsiveness, but neither did they lower a score of Petitioner's proposal due to its nonresponsiveness. In any event, these omissions have not rendered the scoring clearly erroneous.

4. Oddly, Evaluator 3 may have lowered a score of Petitioner for complying with an RFP provision. Evaluator 3 testified that Petitioner improperly included a price within its Technical Proposal, even though, as noted above, Respondent instructed the proposers to do so in Addendum No. 1. However, this act has not rendered Evaluator 3's scoring clearly erroneous.

5. In contrast to the clear, confident testimony of Evaluators 1 and 2, who demonstrated fluency with the RFP and reasonable familiarity with the proposals, the testimony of Evaluator 3 was often vague, sometimes confusing, and, at least once, as noted in the preceding paragraph, confused. Perhaps due to his unique expertise, Evaluator 3 was not as conversant as the other evaluators with the RFP or the proposals. But Evaluator 3's shortcomings do not render his scoring clearly erroneous, although it inspires less confidence than the scoring of Evaluators 1 and 2. In any event, Petitioner would have lost

   
   

   to Intervenor even if Evaluator 3's scores had been discarded. Averaging the scores of Evaluators 1 and 2, Intervenor outscored Petitioner on the Technical Proposal 86.875 to 83.125, so the addition of Intervenor's Price Proposal score of 3.69 and Petitioner's Price Proposal score of 5.33 would have yielded a final score of 90.565 for Intervenor and 88.455 for Petitioner.

6. Moreover, the scoring of the two sections at issue-- Scope of Services 3 and 7--did not reveal that Evaluator 3 was much of an outlier. For Scope of Services 3, Evaluators 1 and 2 assigned a 4 to both proposals, and Evaluator 3 assigned a 3 to both proposals. For Scope of Services 7, Evaluator 3 assigned to each proposal the same score as one of the two other evaluators: for Intervenor's proposal, Evaluators 1 and 3 assigned a 4, and Evaluator 2 assigned a 3, and, for Petitioner's proposal, Evaluator 1 assigned a 4, and

   Evaluators 2 and 3 assigned a 3.

   
   

7. Petitioner's evidence of clearly erroneous scoring takes two forms. First, Petitioner relies mostly on the testimony of its principal, who is extremely knowledgeable about travel demand modeling, but equally interested in the outcome of the case. Second, Petitioner relies on a few internal inconsistencies in scoring that are not so grave as to render the scoring clearly erroneous. Petitioner's task of proving clearly erroneous scoring was undermined by the strong testimony

   
   

   of Evaluators 1 and 2, the open-ended nature of the scoring criteria driving a single score for each section, and, for Scope of Services 7, the large number of unweighted subsections. It is a daunting task for a party challenging a proposed award in a highly technical procurement to set aside scoring as clearly erroneous without the testimony of at least one independent expert witness, who is well informed of the facts of the case.

   1. Scoring of Scope of Services 3: Network

      
      

8. Scope of Services 3 comprises two subsections:

   
   

   1. True Shape Network--At a minimum, the vendor's software must efficiently accommodate true shape networks.

      
      

   2. Integrated Advanced Network Capabilities--Inefficiencies of contemporary modeling networks have made it challenging to share data among models and have led to duplication in data collection. This results in less than optimal model execution times and consequently reduced capacity to develop multiple scenarios efficiently. The vendor's software shall include access to integrated advanced networks and capabilities that promote a unified network platform for all travel demand models in the state and promote more efficient and flexible networks.

   
   

9. These subsections generally ask each evaluator to assess how efficiently the proposed software accommodates true shape networks, which capture the actual geometry of roads rather than invariably representing them linearly as sticks, and the accessibility of the proposed software to integrated

   
   

   advanced networks and capability that promote a unified network platform for all travel demand models. The phrasing of these criteria introduces an element of flexibility in the scoring of the proposals under Scope of Services 3, although this section is much less open-ended than Scope of Services 7 and its myriad criteria.

10. Evaluator 1 testified to no significant differences between the proposals of Intervenor and Petitioner in handling true shape networks and integrating advanced networks. Evaluator 2 testified that the proposals of Intervenor and Petitioner offered true shape networks and also did well in importing other map-based information on top of the road

    information, which evidences the integration of advanced network capabilities. This testimony is credited, and Petitioner has failed to prove that the scoring of Scope of Services 3 was clearly erroneous in favor of Intervenor's proposal.

    1. Scoring of Scope of Services 7: Other Considerations

    
    

11. Scope of Services 7 comprises nine subsections:

    
    

    1. Support Needs and Integration with Other Florida Models

    2. Model Flexibility

    3. Implementation and Collaboration

    4. Private Industry and University Consideration

    5. Comprehensive Documentation

    6. Training Plan

    7. Consultant Support

       
       

    8. Consultant Work Experience

    9. Addressing Florida's Future Modeling Needs

    
    

    Three of these nine subsections have a total of seven subsubsections, so a total of 16 separate scoring criteria are found in Scope of Services 7, which, like other scoring sections, is ultimately assigned a single score of 1 through 4.

12. For Scope of Services 7, Intervenor's proposal received an average of 22.92 points, and Petitioner's proposal received an average of 20.83 points. As noted above, Intervenor's proposal is nonresponsive to Scope of Services 7.3 and 7.4, although Petitioner's proposal is nonresponsive to Scope of Services 7.4. Intervenor's proposal also offers one year, not three years, of training, so as to earn a relatively low score on Scope of Services 7.6 and describes less work experience than that described in Petitioner's proposal. However, the open-endedness of Scope of Services 7 requires deference even to Evaluator 3's enthusiastic endorsement of Intervenor's proposal's response to Scope of Services 7.6 for its division of the state, for personnel training, by latitude, not longitude, exactly as Evaluator 3 does.

13. Nothing in the RFP compels a specific weighting of the

    
    

    16 scoring criteria in Scope of Services 7. Addressing this point in its proposed recommended order, Petitioner argued that for a score "to be true of the overall whole [section,] it must

    
    

    also be true of a fair number of its parts." The deferential standards discussed in the Conclusions of Law undermine this assertion by reducing a "fair number" to a very low number.

    Although Evaluators 1, 2, and 3 struggled to justify their scores for Intervenor's proposal as to Scope of Services 7, as compared to the explanations offered by Evaluators 1 and 2 as to Scope of Services 3, Petitioner failed to prove that their scores were clearly erroneous in favor of Intervenor's proposal.

    CONCLUSIONS OF LAW

    
    

14. DOAH has jurisdiction of the subject matter.

    
    

    §§ 120.569 and 120.57(1) and (3), Fla. Stat. (2017). Any person "adversely affected" by proposed agency action to award a contract in a competitive procurement is entitled to an administrative hearing. § 120.57(3)(b). A person is adversely affected if the person has submitted a proposal. Advocacy Ctr. for Pers. with Disab. v. Dep't of Child. & Fam. Servs., 721 So.

    2d 753 (Fla. 1st DCA 1998).

    
    

15. In a competitive procurement case that does not involve the rejection of all bids, the Administrative Law Judge conducts a "de novo hearing" to determine whether an agency's proposed action is "contrary to the agency’s governing statutes, the agency’s rules or policies, or the solicitation specifications." § 120.57(3)(f). The standard of proof is whether the person challenging the intended award has proved

    
    

    that the agency's proposed action is "clearly erroneous, contrary to competition, arbitrary, or capricious" (Clearly Erroneous Standard). Id. In general, though, administrative

    proceedings are governed by the preponderance standard of proof.

    
    

    § 120.57(1)(j). The difference between these evidentiary standards is significant. A preponderance of the evidence is the greater weight of the evidence, see, e.g., Gross v. Lyons,

    763 So. 2d 276, 280 (Fla. 2000), but the Clearly Erroneous Standard requires proof that the agency's determination is not "within the range of possible and reasonable interpretations." See Cagle v. St. Johns Cnty. Sch. Dist., 939 So. 2d 1085, 1089 (Fla. 5th DCA 2006).

16. Evidentiary, basic, or direct facts, such as whether a bid contained an attachment when submitted, are governed by the preponderance standard. See, e.g., Asphalt Pavers, Inc. v. Dep't of Transp., 602 So. 2d 558, 561 (Fla. 1st DCA 1992).

    Determinations of ultimate facts, mixed questions of fact and law, and technical facts drawing on the expertise of the agency, which typically drive an agency's proposed action, are governed by the Clearly Erroneous Standard. Compare State Contr. & Eng'g Corp. v. Dep't of Transp., 709 So. 2d 607, 609 (Fla. 1st DCA

    1998). Determinations of whether a proposal deviates from a request for proposals and, if so, whether the deviation is material also fall under the Clearly Erroneous Standard. Id.

    
    

17. "Contrary to competition" probably derives from the longstanding requirement of Florida courts that the bidding process assures "fair competition" to all bidders. As stated in Wester v. Belote, 138 So. 721, 723-24 (Fla. 1931), the effect of this standard is:

    to protect the public against collusive contracts; to secure fair competition upon equal terms to all bidders; to remove not only collusion but temptation for collusion and opportunity for gain at public expense; to close all avenues to favoritism and fraud in its various forms; to secure the best values for the county at the lowest possible expense, and to afford an equal advantage to all desiring to do business with the county, by affording an opportunity for an exact comparison of bids.

    
    

18. "Arbitrary" requires that the proposed agency action is "supported by logic or the necessary facts," and capricious precludes proposed agency action that is taken "without thought or reason or is irrational." Hadi v. Liberty Behavioral Health

    Corp., 927 So. 2d 34, 38-39 (Fla. 1st DCA 2006); §§ 120.56(1)(a) and 120.52(8)(e).

19. Any deviation from a requirement in a procurement document may render the bid or proposal nonresponsive, even if the document fails to identify the requirement as a mandatory item on which responsiveness will be determined. See, e.g., State Contr., 709 So. 2d at 609. Deviations from mandatories

    are divided into material variances, which the agency may not

    
    

    waive, and minor irregularities, which the agency may waive. As the court explained in Tropabest Foods, Inc. v. Department of General Services, 493 So. 2d 50, 52 (Fla. 1st DCA 1986):

    although a bid containing a material variance is unacceptable, not every deviation from the invitation to bid is material. It is only material if it gives the bidder a substantial advantage over the other bidders and thereby restricts or stifles competition.

    
    

20. The Clearly Erroneous Standard necessarily applies to actions taken by Respondent in determining the responsiveness and scoring of Intervenor's proposal--here, a deemed determination of responsiveness. For the reasons stated in the Findings of Fact, Petitioner has proved that Respondent's determination of responsiveness is Clearly Erroneous as to the conversions of models, access as co-licensees for universities in their teaching capacity, and affordable access for universities as consultants and private consultants. Respondent's clearly erroneous determinations of responsiveness on these two points resulted in an intended award that is contrary to competition and contrary to the above-cited provisions of the RFP.

21. For the reasons stated in the Findings of Fact, Petitioner has failed to prove that Respondent's scoring of the proposals of Intervenor and Petitioner was Clearly Erroneous.

RECOMMENDATION

It is

RECOMMENDED that the Department of Transportation enter a final order rejecting Intervenor's proposal as nonresponsive.

DONE AND ENTERED this 20th day of April, 2018, in Tallahassee, Leon County, Florida.

S

ROBERT E. MEALE

Administrative Law Judge

Division of Administrative Hearings The DeSoto Building

1230 Apalachee Parkway

Tallahassee, Florida 32399-3060

(850) 488-9675

Fax Filing (850) 921-6847 www.doah.state.fl.us

Filed with the Clerk of the Division of Administrative Hearings this 20th day of April, 2018.

COPIES FURNISHED:

Douglas Dell Dolan, Esquire Department of Transportation

605 Suwannee Street, Mail Station 58

Tallahassee, Florida 32399-0450 (eServed)

Frederick John Springer, Esquire Elizabeth W. Neiberger, Esquire Bryant Miller Olive P.A.

101 North Monroe Street, Suite 900 Tallahassee, Florida 32301 (eServed)

Bryan Duke, Esquire Messer Caparello, P.A. 2618 Centennial Place

Tallahassee, Florida 32308 (eServed)

Andrea Shulthiess, Clerk of Agency Proceedings

Department of Transportation Haydon Burns Building

605 Suwannee Street, Mail Station 58

Tallahassee, Florida 32399-0450 (eServed)

Erik Fenniman, General Counsel Department of Transportation Haydon Burns Building

605 Suwannee Street, Mail Station 58

Tallahassee, Florida 32399-0450 (eServed)

Michael J. Dew, Secretary Department of Transportation Haydon Burns Building

605 Suwannee Street, Mail Station 57

Tallahassee, Florida 32399-0450 (eServed)

NOTICE OF RIGHT TO SUBMIT EXCEPTIONS

All parties have the right to submit written exceptions within

15 days from the date of this Recommended Order. Any exceptions to this Recommended Order should be filed with the agency that will issue the Final Order in this case.

Docket for Case No: 18-000384BID

Orders for Case No: 18-000384BID