1

Colorado Department of Transportation REQUEST FOR PROPOSAL #HAA RFP 14-048 BH

SUBMISSION DEADLINE: September 24, 2013 2:00 PM Proposals submitted to: Colorado Department of Transportation’s (CDOT) Purchasing; 4201 E. Arkansas, Room 200; Denver, CO 80222

Development of a Statewide Travel Model for Colorado

As corridor and other similar studies frequently have been conducted in Colorado, and such studies seem likely to be undertaken on a regular basis in the foreseeable future, the Colorado Department of Transportation (CDOT) has concluded that the quality and efficiency of future study conduct will be enhanced by the development of a statewide travel model. This Request for Proposal (RFP) includes description of the general approach that CDOT feels is appropriate to take in model development, discusses issues and questions associated with the approach to which proposers are encouraged to respond, and describes expected deliverables for each task in the expected work program. The combined effect of CDOT’s goals for this effort has led the project team to the conclusion that this project should begin by acquiring the “Focus” travel model recently built by the Denver Regional Council of Governments (DRCOG), and adapting that model for statewide use. Broadly described, the purpose of this project is to develop a statewide model that is useful to the Department “out of the box” for some applications (but not necessarily for all foreseeable applications), is upgradable in future years, and is feasible to operate by CDOT staff, and consultants and planning partners who may also use it on future projects

Proposer must be a subscriber to the State of Colorado’s BIDS system at the time of the RFP/bid opening for their bid to be considered. Please read this Request for Proposal (RFP) thoroughly before responding. Telegraphic or electronic bids (Fax, Western Union, Telexes, etc.) cannot be accepted directly in CDOT’s Purchasing Office as a sealed bid. Illegible responses may be rejected as non-responsive. The Colorado Department of Transportation (CDOT) reserves the right to reject any and all proposals or parts thereof, and to waive informalities or irregularities. By submission of a proposal, proposer agrees to the State of Colorado terms and conditions. By submission of a proposal, bid and/or quote, proposer agrees as follows: Except as replaced, modified, or supplemented by CDOT for this solicitation, all items in the State of Colorado Solicitation Instructions/Terms and

Conditions are considered part of, and are incorporated by reference into this document. Proposer testifies that bid prices were arrived at independently and there was no collusion involved. The Bidder/Proposer/Vendor guarantees to the State that they understand and agree to the terms and conditions of this RFP and that they will not

default from performance by virtue of a mistake or misunderstanding. Proposers shall seek clarification from CDOT of any specifications, terms and/or conditions that they determine to be unclear. The failure of a proposer to seek clarification may be deemed a waiver of any such clarification.

If applicable, low tie bids/proposals shall be decided in accordance with the provision of C.R.S. Section 24-103-202.5, as it currently exists or is hereafter amended, which gives a preference to resident bidders. Any bidder who wishes to be considered a “resident bidder” for purposes of the tie bid procedure provided in C.R.S. Section 24-103-202.5 shall include with their bid, proof that they meet the definition of resident bidder as set forth in either C.R.S. Section 24-103-101(6)(a) or C.R.S. Section 24-103-101(6)(b).

Pursuant to CRS 24-30-202.4 (as amended), the state controller may withhold debts owed to state agencies under the vendor offset intercept system for: (a) unpaid child support debt or child support arrearages; (b) unpaid balance of tax, accrued interest, or other charges specified in Article 22, Title 39, CRS; (c) unpaid loans due to the student loan division of the Colorado Division of Higher Education; (d) owed amounts required to be paid to the unemployment compensation fund; and (e) other unpaid debts owing to the state or any agency thereof, the amount of which is found to be owing as a result of final agency determination or reduced to judgment as certified by the controller.

This award shall be available for use by CDOT, other State Agencies and Institutions (with approval from State Purchasing), Local Governments and Political sub-divisions in the state of Colorado.

All proposals must be submitted on this form and signed in ink by an authorized officer or agent of the firm.

Proposer Name _________________________________ Signature ____________________________________

Proposer Address_______________________________ Name (Print)___________________________________

City, State, Zip ________________________________ Title Date___________

Proposer Phone Fax _______________ F.E.I.N./SSN ___________________________________ NOTE: Results will be posted on the BIDS System and/or sent via postal system but will not be discussed by phone except as noted in the RFP document.

NOTE: Proposers delivering their proposal in person must check into the CDOT’s Headquarters Building before being allowed to proceed to the Purchasing Office to submit proposal. Proposers should allow approximately 10 minutes in advance of proposaldeadline for the check in procedure. CDOT reserves the right to reject any and all proposals or parts thereof, and to waiveinformalities or irregularities. By submission of a proposal, proposer agrees to the State of Colorado terms and conditions.

2

REQUEST FOR PROPOSAL

THE COLORADO HIGH PERFORMANCE TRANSPORTATION ENTERPRISE

SECTION 1 ADMINISTRATIVE INFORMATION

1.1 ISSUING OFFICE:

This request for proposal is issued for the State of Colorado and the HPTE, by the Colorado Department of Transportation, Center for Procurement and Contract Services. All contacts regarding this RFP is to be directed to: Mr. Brian Hancock, Purchasing Agent Colorado Department of Transportation Center for Procurement and Contract Services 4201 East Arkansas Avenue, Room 200 Denver, CO 80222 Brian.Hancock@state.co.us. (303) 757-9131 FAX (303) 757-9669

1.2 PURPOSE: The purpose of this Request for Proposal (RFP) is to obtain competitive bid proposals from qualified firms interested in providing the Development of a Statewide Travel Model for Colorado to the State of Colorado. This RFP provides prospective proposers with sufficient information to enable them to prepare and submit proposals for consideration by CDOT to satisfy the needs as outlined in this RFP’s Statement of Work.

1.3 SCHEDULE OF ACTIVITIES: DATE: TIME (MST) 1. RFP published on BIDS August 22, 2013 N/A 2. Prospective proposer’s inquiry deadline September 3, 2013 4:00 P.M.

(No questions accepted after this date) 3. Response to proposer questions September 10, 2013 4:00 P.M. 4. Proposal submission deadline September 24, 2013 2:00 P.M. 5. Top consultants selected and notified of October 14, 2013 TBD interview (estimate), if appropriate

6. Oral interviews with a short list of consultants (estimate), if required – week of October 28, 2013 TBD

7. Firms selected (estimate) November 4, 2013 N/A 8. Desired date of executed contract December 4, 2013 N/A 1.4 PROPOSAL SUBMISSION:

All proposals must be received by the CDOT’s Center for Procurement Services, 4201 East Arkansas Avenue, Room 200 Denver, CO 80222, no later than the date and time shown in the Schedule of Activities, Deadline for receipt of proposals. Each proposal shall consist of one original (identified as such) and 5 copies of the proposer’s complete proposal. It is the responsibility of the proposer to ensure that their proposal is received in the Purchasing Office prior to the deadline. Proposers mailing their

3

documents should allow ample mail delivery time to ensure timely receipt of their proposals. PROPOSALS RECEIVED AFTER THE ABOVE DATE AND TIME WILL NOT BE CONSIDERED. Proposals must be clearly identified as a proposal for the RFP #HAA RFP 14-048 BH, and shall show such information on the outside of the proposal packet. Proposals will not be accepted by facsimile transmittal.

Proposers are advised that CDOT desires that proposals prepared in response to this RFP be submitted on recycled paper. While the appearance of proposals is important, and professionalism in proposal presentation should not be neglected, the use of non-recyclable or non-recycled glossy materials is discouraged. In addition, it is requested that proposals be in flat bound form to facilitate filing. Please do not submit proposals in loose-leaf binders.

1.5 INQUIRIES:

Prospective proposers may make written inquiries concerning this RFP to obtain clarification of requirements. No inquiries will be accepted after the date and time specified in the Schedule of Activities, Prospective proposer’s inquiry deadline. Questions must be submitted in writing via e-mail to: Mr. Brian Hancock, Purchasing Agent Colorado Department of Transportation Center for Procurement and Contract Services 4201 East Arkansas Avenue, Room 200 Denver, CO 80222 Brian.Hancock@state.co.us. (303) 757-9131 FAX (303) 757-9669

Subject line of the e-mail shall clearly state “Questions for HAA RFP 14-048 BH ” to facilitate handling

and distribution. Inquiries sent by fax will be accepted (fax number (303) 757-9669). An addendum responding to questions submitted regarding the RFP will be published on BIDS.

1.6 AMENDMENTS TO RFP: In the event it should be necessary to revise any portion of this RFP, addenda will be published on the

State BIDS system. It is the proposer’s responsibility to monitor the BIDS System at the Internet site www.bidscolorado.com, and to acknowledge and/or comply with all addenda to this RFP.

1.7 RESPONSE MATERIAL OWNERSHIP: All material submitted regarding this RFP becomes the property of the State of Colorado. Proposals may

be reviewed by any person after the “Notice of Intent to Make an Award” letter has been issued, subject to the terms of Section 24-72-201 et. seq., C.R.S., as amended, Public (open) Records.

1.8 PROPRIETARY INFORMATION: All material submitted in response to this RFP will become public record and open to inspection after

Intent to Award notice is issued. Any material requested to be treated as proprietary or confidential must be clearly identified and easily separable from the rest of the proposal. Such request must include justification for the request. The request will be reviewed and either approved or denied by the CDOT Purchasing Director. If denied, the proposer will have the opportunity to withdraw its entire proposal, or to remove the proprietary restrictions. NEITHER COST NOR PRICING INFORMATION NOR A TOTAL PROPOSAL WILL BE CONSIDERED PROPRIETARY. Refer to Section 1.31 of this RFP for submission of Confidential/Proprietary information.

1.9 REJECTION OF PROPOSALS: CDOT reserves the right to reject any or all proposals received in response to this RFP, or to cancel this

RFP if it is in the best interest of the State to do so. Failure to furnish all information or to follow the proposal format requested in this RFP may disqualify the proposal. Any exceptions to the Statement of Work must be clearly identified in the proposal. Inclusion of exceptions does not guarantee acceptance by the State of such variation, and may instead lead to rejection of the proposal as non-responsive. (See further Section 1.12 of this RFP.)

1.10 INCURRING COSTS: CDOT is not liable for any costs incurred by proposers prior to issuance of a legally executed contract or

procurement document. All costs to prepare and submit a response to this solicitation shall be borne solely by the proposer.

4

1.11 EVALUATION CRITERIA: An evaluation will be made by a committee selected to evaluate the merits of all proposals received

according to the evaluation criteria defined herein (Section 3). The recommendations of this group will be forwarded to the Purchasing Director for approval.

1.11.1 Failure of the proposer to provide in his/her proposal any information requested in this RFP may

result in disqualification of the proposal. It is the sole responsibility of the proposing individual or firm to ensure all information requested in the RFP is included.

1.11.2 During the evaluation process, discussions/interviews may be scheduled with proposers who

submit proposals determined to be reasonably competitive for selection for award. It will be upon the recommendation of the evaluation committee if discussions/interviews for clarification are needed.

1.11.3 The sole objective of the evaluation committee will be to recommend the proposer(s) whose

proposal(s) is/are most responsive to CDOT’s needs within the available resources. The specifications within this RFP represent the minimum performance necessary for response.

1.11.4 Specific evaluation criteria are outlined in Section 3 of this RFP, entitled Evaluation Criteria.

1.12 ACCEPTANCE OF RFP TERMS: A proposal submitted in response to the RFP shall constitute a binding offer. Acknowledgment of this condition shall be indicated by the autographic signature of the proposer, or an officer of the proposer, legally authorized to execute contractual obligations. A submission in response to the RFP acknowledges acceptance by the proposer of all terms and conditions including compensation, as set forth herein. Any exceptions and/or variations to the terms and conditions presented in the RFP may be submitted as part of the proposal, with each such exception and/or variation identified clearly and thoroughly. Failure to identify any exceptions and/or variations in the submitted proposal shall be deemed a waiver of any rights to subsequently modify the terms of performance, except as outlined or specified in the RFP, and may result in cancellation of the award and such vendor may be removed from future solicitations. Submission of a proposal containing exceptions and/or variations does not guarantee of acceptance of such variations by CDOT, and may instead lead to the rejection of the proposal as non-responsive if the requested variations are determined to be extensive or unreasonable, by the evaluation committee assigned to this RFP solicitation.

1.13 PROVISION FOR REQUIRED INSURANCE: Award of a contract will be contingent upon the successful proposer submitting certificates of insurance in accordance with the provisions of the sample contract, Attachment B.

1.14 CONSULTANT CERTIFICATION: Proposers must submit a signed Consultant Certification Form, CDOT Form #637, with their proposal, Attachment A to this RFP.

1.15 CONFLICT OF INTEREST:

By submission of a proposal, proposer agrees that, at the time of contracting, the proposer has no interest, direct or indirect, that would conflict in any manner or degree with the performance of the required services. The proposer shall further covenant that, in the performance of the contract, it shall not employ any person having any such known interest. Any firm affiliated or related to an employee of CDOT shall be ineligible to submit a proposal for the required services.

1.16 REQUEST FOR PROPOSAL: The Request for Proposal Form - the cover page for this RFP - must be signed, in ink, by a person authorized to bind the proposer, and returned with the proposal.

1.17 AUDIT OF THE SELECTED PROPOSER:

Prior to final contract award, an audit may be conducted by the CDOT’s External Audit Branch of the selected proposer. This audit will be for the purpose of ensuring that the selected firm is financially capable of performing the contract, that the cost information and prices quoted are reasonable, and that the selected proposer has adequate accounting practices to assure accurate tracking of contract costs.

5

Prior to final acceptance of the contract work, a closing audit of the proposer may be performed by the CDOT External Audit Branch. This final closeout audit will be performed upon completion of the contract to verify the accuracy of the billings and compliance with the contract provisions.

1.18 BUDGETED FUNDS: The available funds for this solicitation is set at $1,200,000.00. 1.19 INTENT TO AWARD:

After a proposer is selected, an “Intent to Award” letter will be mailed to all firms who submitted a proposal. After intent to award has been issued, interested parties may review any/all the proposals by making an appointment with: Mr. Brian Hancock, Purchasing Agent Colorado Department of Transportation Center for Procurement and Contract Services 4201 East Arkansas Avenue, Room 200 Denver, CO 80222 Brian.Hancock@state.co.us. (303) 757-9131 FAX (303) 757-9669

1.20 PROTESTED SOLICITATIONS AND AWARDS: Any actual or prospective proposer or contractor who is aggrieved in connection with the solicitation or award of a contract may protest to Bob Corman, CDOT’s Purchasing Director. The protest shall be submitted in writing within seven (7) working days after such aggrieved person knows, or should have known, of the facts giving rise thereto. Protests received after the seven-working-day period shall not be considered. The written protest shall include, as a minimum, the following: A. The name and address of the protestor; B. Appropriate identification of the procurement by bid, RFP, or award number; C. A statement of the reasons for the protest; and D. Any available exhibits, evidence or documents substantiating the protest.

1.21 STANDARD CONTRACT: CDOT has provided as Attachment B a Sample Contract that it expects to enter into with the successful

proposer. 1.22 SELECTION OF PROPOSAL:

All proposers will be notified in writing regarding the results of the RFP evaluation. Upon review and approval of the evaluation committee’s recommendation for award(s), the CDOT Procurement Office will issue a “Notice of Intent to Make an Award” letter to the apparent successful proposer(s). Provided, however, that all proposers understand that such letter, by itself, does not grant any property interest or right of any nature in the RFP work/services or to a contract for the performance of such work/services. A contract must then be completed and signed by all parties and the State Controller, before any such right exists. Therefore, the apparent successful proposer(s) that receive a “Notice of Intent to Make an Award” letter shall not rely on that letter to make commitments to third parties, and the apparent successful proposer(s) shall not take any actions(s) to prepare for or start the performance of the RFP work/services until a contract is so negotiated and executed. In addition, a contract must be completed and signed by all parties concerned on or before the date indicated in the Schedule of Activities.

1.23 AWARD OF CONTRACT: The award will be made to that proposer(s) whose proposal conforms to the RFP, and is/are judged to be

the most advantageous to the State of Colorado and CDOT, price and other factors considered, subject to negotiation and execution of an acceptable contract as described above.

1.24 It is the intent of CDOT to select a vendor within 30 days of the deadline for receipt of proposals.

However, bid proposals must be firm and valid for award for at least 120 days after the deadline for receipt of proposals.

1.25 NEWS RELEASES:

6

News releases pertaining to this RFP shall NOT be made prior to execution of a contract, and then will be made only with the approval of CDOT.

1.26 CERTIFICATION OF INDEPENDENT PRICE DETERMINATION:

1.26.1. By submission of this proposal each proposer thereto certifies as to its own organization, that in connection with this procurement:

(a) The prices in this proposal have been arrived at independently, without consultation, communication, or agreement, for the purpose of restricting competition, as to any matter relating to such prices with any other proposer or with any competitor;

(b) Unless otherwise required by law, the prices which have been quoted in this proposal have not been knowingly disclosed by the proposer and will not knowingly be disclosed by the proposer prior to opening, directly or indirectly to any other proposer or to any competitor; and

(c) No attempt has been made by the proposer to induce any other person or firm to submit or not to submit a proposal for the purpose of restricting competition.

1.26.2 Each person signing the Invitation for Bid form of this RFP certifies that: He/she is the person in the proposer’s organization responsible within that organization for the

decision as to the prices being offered herein and that he/she has not participated, and will not participate, in any action contrary to 1.26.1 (a) through (c) above.

or He/she is not the person in the proposer’s organization responsible within that organization for the decision as to the prices being offered herein but that he/she has been authorized in writing to act as agent for the persons responsible for such decision in certifying that such persons have not participated, and will not participate, in any action contrary to 1.26.1 (a) through (c) above, and as their agent does hereby so certify; and he/she has not participated, and will not participate, in any action contrary to 1.26.1 (a) through (c) above.

1.26.3 A proposal will not be considered for award where 1.26.1 (a) and (c), and 1.26.2 above, have

been deleted or modified. Where 1.26.1 (b) above has been deleted or modified, the proposal will not be considered for award unless the proposer furnishes with the proposal a signed statement which sets forth in detail the circumstances of the disclosure and the head of the CDOT’s Purchasing Office, or designee, determines that such disclosure was not made for the purpose of restricting competition.

1.27 TAXES The State of Colorado, as purchaser, is exempt from all Federal taxes under Chapter 32 of the Internal Revenue Code (Registration No. 84-730123K), and from all State and Local Government Use Taxes (Ref. Colorado Revised Statutes Chapter 39-26.114[a]). Proposer is hereby notified that when materials are purchased in certain political subdivisions the seller may be required to pay sales tax even though the ultimate product or service is provided to the State of Colorado. This sales tax will not be reimbursed by the State.

1.28 FUNDS AND COMPENSATION:

The funds payable after the current fiscal year are contingent upon funds for that purpose being appropriated, budgeted, and otherwise made available. In the event funds are not appropriated, any resulting contract will become null and void, without penalty to the State of Colorado or CDOT.

1.29 BACKGROUND, OVERVIEW, GOALS:

As corridor and other similar studies frequently have been conducted in Colorado, and such studies seem likely to be undertaken on a regular basis in the foreseeable future, the Colorado Department of Transportation (CDOT) has concluded that the quality and efficiency of future study conduct will be enhanced by the development of a statewide travel model. This Scope of Services (the Scope) describes the general approach the CDOT feels is appropriate to take in model development, discusses issues and questions associated with the approach to which proposers are encouraged to respond, and describes expected deliverables for each task in the expected work program.

7

In the process of developing a statewide travel model, the Department seeks to forward several goals: Reduce the cost and shorten the schedule of studies requiring models by making available a model

for consultants to use in support of study analysis. At present, it is typical for the Department to require consultants to begin studies by building a new modeling tool for use in each study. In most cases, this increases project budget and lengthens schedule.

Enhance consistency in modeling approach between one study and another. When new models are developed for each study, it is almost inevitable that different approaches will be taken in each study, making forecasting inconsistency between studies also almost inevitable.

Make maximum practical use of past modeling efforts, and modeling efforts undertaken by partner agencies. At present, CDOT has very little ability to internalize and maintain past modeling work, or efforts by other agencies, so that work can be deployed on future projects. A CDOT statewide model can serve as a platform for retaining such improvements, and for leveraging the work of partner agencies.

Build a first version of the model that is usable on at least one class of projects. As a rule, even mature models can’t serve all conceivable planning or engineering needs, and attempting to build a brand new model that can serve a wide spectrum of needs is a risky strategy. However, building a “version 1.0” model that can’t serve any practical needs also is a risky strategy, as such a course is likely to damage agency support for the effort.

Emphasize ease and speed of model use, while building a model that takes advantage of recent leading-edge work in Colorado. The easier a model is to use and the faster it runs, the more it will be used. Ease of use will be particularly important in the early stages of a CDOT modeling program, when staff are likely to be inexperienced at operating and maintaining such models. However, the Department should not pursue these goals to such an extent as to the harm the usefulness of the model it builds for anticipated projects, or to fail to take advantage of the many advances taking place in the modeling field today.

Foster cooperation and collaboration between CDOT, MPOs, member governments, etc. Development of a statewide model that makes use of model advances created by other planning partners in the state, and is of value to those partners, is key to fostering such collaboration.

Enhance CDOT’s ability to operate and maintain the model when it is delivered. CDOT has found that the best way to forward this goal is to structure the project so that CDOT staff do an appropriate amount and type of development work as the project proceeds. In this way, CDOT staff will better understand the model when it is completed.

The combined effect of these goals has led the project team to the conclusion that this project should begin by acquiring the “Focus” travel model recently built by the Denver Regional Council of Governments (DRCOG), and adapt that model for statewide use. While older, aggregate trip-based models are simpler in their structure, and take less time to run, they are less able to support planning efforts in a long list of ways. In addition, many metropolitan planning organizations (MPOs) and several state DOTs are shifting from their older trip-based models to the newer activity-based model (ABM) technology; if CDOT adopted the older trip-based design as its platform, there is an excellent chance that the agency would find itself well behind its peers very shortly after having completed this project. Adopting an older platform also will hinder the goal of cooperation and collaboration with partner agencies in the state.

A number of characteristics of the Focus model also are susceptible to enhancement and simplification, adding further support to its selection as the foundation of a CDOT statewide model. For example, while Focus has been proved effective as a planning support tool through two rounds of DRCOG regional transportation plan (RTP) development (in 2010 and 2012), it is still a relatively new technology, and many options are available to improve its speed of operation. In addition, Focus includes a number of capabilities that CDOT is not likely to need in the short-term; Focus can be adapted to simplify these capabilities, enhancing ease of operations and reducing run time speed. For example, in typical Focus operation at DRCOG, great care is taken to depict the locations of households and jobs to a high degree of detail and accuracy. In the short run, CDOT is likely to find that a more approximate approach to this type of model input will be sufficient, saving considerable time and effort in operating the model. Numerous other model simplifications and enhancements are possible, that could greatly enhance that model’s usefulness at the statewide level.

8

While DRCOG’s Focus model has many desirable features, any regional-level model will require some modifications simply to support basic state-scale functionality. Such modifications will be a key element of this project. For example, modifications will have to be made to better support rural and inter-city/town travel, as regional-scale models rarely have to address this issue. Other analogous adaptations are sure to be required.

The purpose of this project is to develop a statewide model that is useful to the Department “out of the box”, is upgradable in future years, and is feasible to operate by CDOT staff, and consultants and planning partners who may also use it on future projects. Instructions to Proposers Respondents should note that CDOT will cooperate closely with the state’s Office of Information Technology (OIT) on some aspects of this project, and that OIT has on-call IT consultants who may be called in to assist on aspects of this project (particularly those related to IT-specialty capabilities, such as relational database optimization and administration, code performance optimization, etc.)

CDOT expects a variety of skill sets to be necessary to conduct the tasks in the scope below, and will select a single firm or team of firms under this solicitation, to conduct these tasks as CDOT assigns. The project will be conducted in the following tasks:

Task 1 – Review CDOT requirements, and group by priority and feasibility Task 2 – Preparation of input/estimation/calibration/validation data Task 3 – Adaptations of regional model for state-level application Task 4 – Enhancing Focus functionality for state-level modeling Task 5 – Enhancing automation of model input development Task 6 – Software upgrades for faster, easier model operations Task 7 – Model installation and implementation support

Skill areas that CDOT seeks in responding consultants include:

For scope tasks 1 through 5, expertise in all aspects of travel demand forecasting, such as:

o Design and estimation of activity-based travel models o Development of detailed input data for activity-based travel models o Calibration and validation of activity-based travel models o Development of travel models at the statewide level o Development of software to implement activity-based travel models, including model

execution, model input development, model output summary, and model deployment o Use of and development in model software environments such as TransCAD, Cube, etc. o Integration of activity-based travel models with traffic simulation tools, dynamic traffic

assignment tools, and advanced land use models. For tasks 6 and 7, and for advising on parts of task 1, expertise in software development, including:

o Microsoft .NET Framework , o Application architecture, o Successfully designing and deploying large scale Microsoft SQL Server environments, o Application integration, o Managing a complete software development lifecyle, o Familiarity with developing Cloud available applications, o SQL Programing (including T-SQL), o User interface design and development, o Setting up, managing, and using a source code management solution (code check in, check

out, review, etc…), o Creating test protocols, and environments (to test software functionality).

To simplify CDOT’s review of contractor proposals, proposers should separate their response to each task below as follows: a description and accompanying budget for their approach to development of a “basic package” model (as discussed at various points in the scope), and their approach and accompanying budget to developing whatever capabilities beyond that package that they may wish to

9

propose (if any). To help clarify these instructions for bidders, bidders should note that the basic package model will be, simply described, DRCOG’s Focus activity-based travel model, adapted to serve at the full statewide level the same capabilities that Focus serves in the DRCOG region. While this list is not intended to be a complete description of Focus, the basic capabilities include: Modeling of typical weekdays in the spring and fall.

Modeling trips in tours, rather than in separate, disconnected origin-destination trips.

Modeling trips in a detailed set of real trip/tour purposes (e.g., work, school, shopping, etc.) rather than in the abstracted “purposes” used in older trip-based models (e.g., “home-based work”, “home-based non-work”, “non-home-based”, etc.)

Trip and tour time-of-day choice capability.

Modeling of trips/tours in an overall day pattern.

1.30 STATEMENT OF WORK: Task 1 – Review CDOT requirements, and group by priority and feasibility The purpose of this task is to review CDOT’s requirements for a statewide model, and provide recommendations that will guide the model development and implementation tasks of this project. Recommendations will be developed in four general categories:

Detailed recommendations of how to implement the “basic package” model (adapting DRCOG’s Focus model for statewide use), which will form the core of this project, including budget (to be carried out in Task 3); and

Detailed recommendations of extensions to the basic package model, including separate budget to implement each recommendation. Upon reviewing these recommendations, CDOT will choose from among the consultants proposed extensions those (if any) that CDOT wishes to include in this project (and be implemented under Task 4).

Detailed recommendations for automation of model inputs, including separate budget to implement each recommendation. Upon reviewing these recommendations, CDOT will choose from among the consultants proposed input automation approaches those that CDOT wishes to include in this project (and be implemented under Task 5).

Detailed recommendations for software upgrades to permit faster, easier model operations, including an overall budget for such activities (that is, CDOT is not asking a list of separate optimization activities with separate budget for each, but a list of such activities with an overall budget that covers the entire recommended package of software optimization.) These optimization activities will be implemented under Task 6.

Task 1.1 Review of CDOT Analytical Requirements and Potential Capabilities of a CDOT Statewide Model The purpose of this sub-task is to review past (and likely future) CDOT analytical needs that can or should be supported by a statewide model, and develop plans for modeling functionality to be implemented in tasks 3 and 4, including:

The “basic package”: adaptations of the DRCOG Focus model needed to make it work at a statewide scale – implemented in Task 3.

Model extensions: enhancements that are not currently supported by the DRCOG Focus model, but that are desirable features of a statewide model, and that CDOT may decide to include in this project – implemented in Task 4.

CDOT has in the past conducted numerous studies and other projects for which travel demand forecasting models were required. Examples include the North I-25 EIS, the mountain I-70 Programmmatic EIS and its

10

precursor studies, the currently on-going Colorado Interregional Connectivity Study (ICS) and the Advanced Guide-way System (AGS) studies, the Ports to Plains Study, and many others. These studies required analysis from travel forecasting models of varying degrees of detail and rigor, but frequently required highly detailed multi-modal analysis. Few if any statewide or regional models in the world today are able to respond to the full range of requirements that these CDOT studies have involved, and indeed almost all past CDOT studies involved the development of modeling tools specialized to the needs of the particular project. At this moment, it is unknown whether or not a single modeling system can satisfy the requirements of all likely future projects, but past experience suggests this is unlikely to be the case, and that such an omni-competent model also is not necessary. The goal of this project is to develop a modeling tool that when its first version is complete, will serve SOME of those likely requirements (especially those that are likely to confront CDOT in the nearer term), and serve as a foundation on which other capabilities can be built over time, as, if and when necessary. Various regulatory requirements also must be considered in developing an initial model (MAP21, etc.) To serve that purpose, this task will review and assemble a thorough list of CDOT’s past and present travel forecasting model needs, and evaluate as far as possible CDOT’s likely future needs. This analysis will be used to develop a list of modeling needs to which this project may respond, and to categorize those needs by likely time-frame (short, medium, and long-term), model type (regional demand model, dynamic traffic assignment model, micro-simulation model, etc.) and other dimensions that the project team may consider necessary. At minimum, this task will involve:

Review of reports from past projects that included a forecasting component;

Interview of selected individuals, including interview of past and present staff from CDOT, RTD, MPOs in Colorado, etc.;

A discussion process with project staff and others to develop the modeling requirement dimensions and categorize requirements by those dimensions;

A similar discussion process to select modeling capabilities to be addressed in this project, and to distinguish them from capabilities that will have to be deferred to other projects, or possibly later phases of this project; and

A discussion and conclusion concerning whether or not any model estimation should be performed on this project (and if so, for which components), as opposed to simply re-calibration/validation, together with more ad-hoc model adjustments that fall short of re-estimating any model components.

A variety of core challenges have been encountered in past CDOT modeling projects, which will inform the process of identifying likely future modeling needs. These challenges have included:

Complex recreational trip/tour purposes

Travel demand on weekends, and demand varying by season

Demand suppression/inducement resulting from existing conditions and base and build-scenarios

Detailed effects of lane and other pricing measures, and of other corridor and facility management techniques.

The effects of and demand for transportation facilities unlike any that exist in the region (and in some cases the world) at present.

The “Basic Package” As the central approach of this project is to acquire the DRCOG Focus model and adapt it for statewide use, the project team will begin this task with an analysis of the capabilities of that model, and their ability to address key CDOT needs. A number of basic capabilities are almost certain to be present in any such adaptation of the Focus model. Examples of such capabilities include:

Modeling of typical weekdays in the spring and fall.

Modeling trips in tours, rather than in separate, disconnected origin-destination trips.

11

Modeling trips in a detailed set of real trip/tour purposes (e.g., work, school, shopping, etc.) rather than in the abstracted “purposes” used in older trip-based models (e.g., “home-based work”, “home-based non-work”, “non-home-based”, etc.)

Trip and tour time-of-day choice capability.

Modeling of trips/tours in an overall day pattern.

Adaptations of these capabilities necessary to support needed statewide function are described below in Task 3 of this scope of services (though that list is not assumed to be exhaustive.) Model Extensions In addition to these basic package capabilities, the consultant shall describe a recommended approach, data needs and budget for the following extensions to the basic package:

Weekend travel. While the Focus model includes an explicit social/recreational trip purpose for weekday travel, many recreational trips take place on weekends. The contractor shall describe approaches to improving the model’s ability to address weekend travel, separately describing approaches that would require additional data, and approaches that would not.

Summer and winter travel. The heaviest recreational trip volumes in key corridors (most prominently the I-70 mountain corridor) do not occur during spring and fall (the seasons that typical regional models address) but in both summer and winter. The contractor shall describe approaches to improving the model’s ability to address summer/winter travel, separately describing approaches that would require additional data, and approaches that would not.

In addition to these capabilities, the contractor may provide approach and budget for other extension capabilities that it feels will be helpful in meeting CDOT’s goals for a statewide model.

Depending on available funding, budgets for the proposed model extensions, and evaluation of the likely effectiveness of any proposed extension, CDOT will choose to move forward with none, some, or all proposed extensions. Task 1.2 Enhancement of Model Operations The purpose of this subtask is to develop detailed plans for:

automating the development of key inputs needed by the model; and

shortening the run time of the model.

Disaggregate models such as DRCOG’s Focus model require much more detailed input data files than older aggregate models do. An on-going task with these disaggregate models is to develop tools that make them easier and faster to run. Such improvement is a key area of emphasis for this project, as ease and speed of model operations is a high priority for CDOT. Such enhancements will take two forms: software tools to speed and simplify the work of developing input data (to be implemented in Task 5) such as the synthetic population, the point location dataset for each job and household, etc., and software improvements to make the model run faster (to be implemented in Task 6.) Lists of example elements of these enhancements, not intended to be exhaustive, are provided below under tasks 5 and 6. The contractor shall prepare detailed plans and budgets for each of these two aspects of model enhancement. Task 1 Deliverables:

A detailed description of the contractor’s plan (approach, data needs and budget) for developing basic package of statewide model capabilities,

Separate detailed descriptions of the contractor’s plan (approach, data needs and budget) for developing the model extensions listed above; and

Separate detailed descriptions of the contractor’s plan (approach, data needs and budget) for developing any other model extensions the contractor feels will help address CDOT’s goals in developing a statewide model.

12

A detailed plan for development of software/tools to enhance automation of travel model inputs.

A detailed initial plan for development and implementation of software enhancements to reduce the model’s run time. Note that it is assumed that this plan likely will be modified during Task 6 below, when the plan is implemented, as the contractor learns through experience what improvements are more effective than others.

Task 2: Preparation of input/estimation/calibration/validation data. The purpose of this task is to develop data needed for input/estimation/calibration/validation of all model elements selected by CDOT in Task 1. The results of Task 1 will provide a list of data needed to support the project’s agreed-upon design elements. In this task, that data will be acquired and prepared for use in the project. Extensive data is available for this project, but there are data gaps, and it is possible that new data will be needed. However, given the project schedule, absent very strong arguments to the contrary, the presumption is that the model will be designed and implemented with data already available, either to CDOT or to its various partners, or available “off the shelf” from one or another source (e.g., Census data, private vendor data, etc.) Examples of data available for this project include (but are not limited to):

Traffic counts – CDOT and its partners gather and maintain an extensive set of traffic counts, including classification counts, throughout the state.

Transit ridership data – this is available from various transit providers, MPOs, etc.

Speed data – CDOT has roadway speed data in at least two forms: speed-delay “floating car” data gathered between 2007 and 2009, and NAVTEQ data.

Front Range Travel Counts (FRTC) survey data, including a household diary survey of weekday travel for 12,000 households across the Front Range, and cordon survey data covering the Pueblo, Colorado Springs, and Denver metropolitan area MPOs. Cordon survey data is also available through a separate survey for the North Front Range region. The household survey also includes a supplemental survey of long distance trips (defined as trips of over 50 miles.)

Highway and transit networks. CDOT currently is working on developing statewide highway and transit networks. These will be available in either complete or partial form when this project commences.

Base and future year employment/household/population estimates from the Colorado State Demographers’ Office (DOLA.)

US Census data of a variety of types.

The consultant may also suggest data from various private sources as it feels appropriate to do so (with the cost to be described in an overall project budget). Private sources are available for several categories of data, including speed, demographic, business, and other data.

This task will involve, in some cases as approved by CDOT, acquiring the data, and working with CDOT to format the data for use. At task commencement, the contractor and CDOT will discuss and come to agreement concerning division of labor between CDOT and consultant staff for this effort. However, any data prepared by CDOT will be provided to the contractor, who will combine that data with its own data, and deliver the final complete datasets. Examples of data formatting that may be necessary include (but are not limited to):

Defining geographic districts (aggregations of traffic zones) throughout the FRTC survey area, and summarizing trip/tour data into these districts for use in adjusting and calibrating trip location choice models.

Combining household long-distance trip survey data with the long-distance supplemental survey to produce a more reliable table of long-distance trips, also at the district level.

13

Working with CDOT staff to consolidate the highway network’s traffic zone system to produce a level of zonal detail sufficient for statewide model needs, balancing the need for analytical detail and the need for reasonable model run times.

Attaching traffic count and speed data to the model highway network.

Task 2 Deliverables: A plan acceptable to CDOT for division of labor between CDOT and the contractor for development of

input data needed to run the statewide model, including a base year for calibration purposes, and a forecast year for validation and reasonableness checking.

all data needed for agreed-upon model estimation, calibration and validation, both for all basic package model components, and for any model extensions agreed upon by CDOT, including all meta data (data documentation) needed to describe and work with the data, according to the plan described in the first Task 2 deliverable.

Task 3: adaptations of regional model for state-level application. The purpose of this task is to implement the basic package of adaptations of DRCOG’s Focus model, as described in the Task 1 deliverables, necessary to make Focus operable at the statewide level. In this task, the contractor shall implement the Focus adaptations agreed upon in Task 1. While CDOT and the contractor will develop a detailed approach to such adaptations in Task 1, the following discussion (not intended to be exhaustive) is provided to illustrate CDOT’s understanding of these issues at the time of project initiation, as a guide to the contractor’s work:

Networks and zones. CDOT is working on combining the MPO and other networks into a statewide network, starting by retaining the zone system of those networks. This is may result in a network with more zones than are practical from several perspectives: model time to create skims, size of skim files, etc. Some degree of zonal consolidation therefore may be necessary. Depending on the progress CDOT staff have made by project notice to proceed, some of this work may be conducted by the consultant team. Considerable network checking also is likely to be necessary, to find and correct problems. Tests may include developing a draft trip table, assigning it to the network, and examining the results for anomalies. Skims may also be generated and examined for anomalies.

Modeling longer trips. A statewide model will have to model inter-city and long rural trips, something that metro-scale models are unlikely to do without adjustment. In some models, separate trip “purposes” are established for long distance trips, making a sharp distinction between them and intra-regional trips. However, it seems likely that such an approach will require quite significant modification to the Focus model structure, particularly in the day pattern model, which selects sets of stops/tours from a set of seven purposes (work, school, escort, personal business, social/recreation, meal, and shop.) As this set of purposes specifically addresses key purposes for which people make long-distance trips (particularly work, personal business, and social-recreational), perhaps an effective approach that would be simpler to implement in the short run would be to retain this same set of purposes, and adjust location choice models to handle longer trips. Existing Focus location choice models include a rich set of independent variables that may be adapted for this purpose. A number of other approaches may also be used, including adding variables to the utility expressions, including district-to-district constant terms, etc. Adjustments to the intermediate stop and location choice models may also be necessary in this approach, if the data show that longer-distance trips exhibit different patterns in these aspects of tours.

Non-motorized trips. While the Focus model does a good job of depicting bicycle and pedestrian trips in the DRCOG area, how it will perform in other areas is more uncertain, particularly in rural areas and smaller towns and cities away from the Front Range. Some approach to addressing this issues seems likely to be necessary, from broadly calibrating bike/ped mode share in parts of the state and admitting that the results are not very accurate, to eliminating bike/ped from the mode choice set in some areas.

Land use – related variables. The Focus model was built to be much more sensitive to detailed land use patterns than were older trip-based models, and to produce this sensitivity includes numerous land-use-related variables in numerous utility functions. These variables allow Focus to show land use effects on transportation choices at very small geographic scales. It is CDOT’s assumption that a “version 1.0” Focus model for statewide application may not need to be capable of such fine-grained analysis, partly

14

because such geographic detail may not be necessary in a state-scale model, and also because producing such detailed input variables at a state scale is likely to be very challenging. Various approaches to managing such variables are possible, including modifying them or eliminating them from utility expressions, or simplifying the approach to developing them (noting that they must be developed both for base and forecast years). Examples of such variables include:

o Network-derived variables – primarily intersection density variables used as surrogate for pedestrian friendliness, and are derived from a GIS-based street map. A simple spreadsheet model is used to estimate density in forecast years for TAZs where there is little or no network today.

o Variables derived from point location of households and jobs. Various formulations based on the point locations are derived for utility expressions. Examples include retail density within ½ mile of the geographic center of employment in each TAZ, and a measure of mixed use around the same center (with mixed use calculated as a formula based on households and jobs in the area.)

Zone-level land use estimation and forecasting. The basic land use data file input to DRCOG’s Focus model is produced at the TAZ level (follow-on procedures are then executed to place each job and household at points in the zone.) An approach to developing a TAZ-level land use data file also will have to be developed for the state-scale model. Elements of an approach to the problem may include, for example, taking the MPO forecasts in the areas covered by those models, and using one of a number of possible datasets to develop land use for the rest of the model. Sources include the US Census, the Colorado State Demographer’s office, the Quarterly Census of Employment and Wages (QCEW, generated by the Colorado Department of Labor and Employment), and a number of private data vendors.

Other land use details. Because a standard input to disaggregate models like Focus is a synthetic population, one or another population synthesizer must be set up to run at the state-scale level. At present, Focus uses a population synthesizer called PopSyn, which performs reasonably well, but is now somewhat antiquated, both in function and software. Different population synthesizers are available and may be desirable for use in CDOT’s state-scale model. Similarly, each job and household must be given a point location. The approach taken with Focus by DRCOG is highly detailed and time-consuming, based on parcel-level zoning provided by all counties within the DRCOG region, an approach that is unlikely to be practical at state scale. It is of course the case that even old, aggregate, trip-based models give a precise point location to each job and household in each zone (the zone centroid), so the goal here is to do as much better a job than that as is practical (in cost, operational effort, etc.), providing as much model accuracy and sensitivity as possible for as little cost and operational effort as possible, recognizing that further enhancements may be implemented at later dates.

Commercial vehicle – The DRCOG Focus model includes a commercial vehicle component, and such capability must also be included in the state-scale model. The Focus commercial vehicle component still uses the old aggregate trip-based approach, and CDOT assumes that this approach will also be used at the state scale for the first version of the state model. CDOT is a funding participant in the Front Range Travel Counts project, and the next survey to be implemented in that project will be a survey of commercial vehicles, so any enhancement to the model’s commercial vehicle component should be deferred until that survey is completed. Various methods of calibrating such a model at the state scale are available. For example, CDOT purchases TRANSEARCH commodity flow data each year. Such data will be a key input to the calibration process.

IE/EE trips. The FRTC project is gathering IE/EE data for the Front Range area, but this data will be of limited use in developing an IE/EE set of trip tables at the state scale. Data to calibrate an IE/EE gravity model will likely be sparse, so a key issue for this model element will be identifying practical data sources and methods of employing that data. CDOT also expects that, at the state scale, IE/EE trips will constitute a very small fraction of overall trips, so overall errors due to errors in this category will be small. Potential exceptions to this claim will involve cities and towns close to the Colorado boarder, particularly those with sizable cities and towns near them in adjacent states. Examples may include Durango, CO and Farmington, NM, Fort Collins, CO and Cheyenne, WY, Trinidad, CO and Las Vegas, NM, and Burlington, CO and Goodland, KS.

15

Theoretical structure adaptations. It may be that some aspects of DRCOG’s Focus model should use a different theoretical structure in a statewide adaptation, to support faster run times or other performance goals. For example, while it may be that a location choice model that uses choice set sampling will be faster than the existing Focus approach of including all zones in the choice set, it may also be the case that a gravity model approach will be faster than any location choice model. And while location choice models are widely considered to be theoretically superior to gravity models, there still is some debate on this topic in the modeling community.

Task 3 Deliverables: all necessary model development, estimation, calibration and validation necessary to produce a functional version of the DRCOG Focus model at the statewide level, as described in the Task 1 deliverables, together with all computer code and data necessary to implement the model, and detailed documentation of all steps in their development, testing, and operations. Task 4: enhancing Focus functionality for state-level modeling. The purpose of this task is to implement any model extensions selected by CDOT as identified in the deliverables for Task 1, beyond those implemented in Task 3. MPO models such as Focus are primarily useful for typical fall/spring weekday travel. While such a model is useful for a wide variety of state-scale purposes, some key corridors in the state require analysis of other types of trips. Examples include:

Weekend travel. While the Focus model includes an explicit social/recreational trip purpose for weekday travel, many recreational trips take place on weekends. The proposer should describe approaches to improving the model’s ability to address weekend travel, describing approaches that would require additional data, and approaches that would not.

Summer and winter travel. The heaviest recreational trip volumes in key corridors do not occur during spring and fall (the seasons that typical regional models address) but in both summer and winter. The proposer should describe approaches to improving the model’s ability to address summer/winter travel, describing approaches that would require additional data, and approaches that would not.

In this task, the contractor will implement all model extensions (beyond the basic package implemented in Task 3), as selected by CDOT and identified in the deliverables of Task 1. Task 4 Deliverables: all statistical models, software necessary to implement them, input data necessary to operate them, estimation/calibration/validation results, and detailed documentation of all steps in their development, testing, and operations. Task 5: enhanced automation of model input development. The purpose of this task is to develop software/tools to automate as much as possible the development of often complex input data needed to run a scenario with the Focus model. Disaggregate models such as DRCOG’s Focus model require much more detailed input data files than older aggregate models do. An on-going task with these disaggregate models is to develop tools that make them easier and faster to run, and this is likely to be a key area of emphasis for this project, as ease and speed of model operations will be a high priority for CDOT. In addition to the work of Task 6, which will develop enhancements to the Focus software to permit faster operation of the model itself, attention also must be paid to the time and effort required to develop model inputs, to speed and automate that process where possible. Possible approaches (provided here for purposes of discussion and clarification, to indicate the core of CDOT’s interest in this task, and not intended to be exhaustive) could include:

The point locations of each household or job. As discussed above, DRCOG’s approach to developing these points is highly detailed, and makes use of parcel-level zoning provided by county assessor’s offices. This project should develop methods of automating the development of such points so that their development is fast and easy, and requires less complex input data. For example, perhaps it would make sense to use the DRCOG points in that model area, and use some other approach outside the DRCOG region. Other approaches might include simply randomly distributing the points in each zone outside the DRCOG model area. Or perhaps there is some other kind of intelligence that can be introduced with relative simplicity. Perhaps if zones are more numerous and smaller in towns, points can be randomly distributed there, and points in larger zones adjacent to towns can be distributed in weighted random fashion, with more points located nearer the towns. Many approaches are possible.

Development density variables. As discussed briefly above, Focus includes ten to twenty variables that depict the density of development at trip origins and destinations. Typically, these are characterized as

16

the density of job or household points (or both) within ½ mile of the geographic center of households (or employment) in the zone. Preparing such variables involves several calculations, usually simple, and this project should include methods of automating their creation.

Intersection density variables. These are used as surrogates for pedestrian friendliness, and derived based on a count of intersections by type (four-way, three-way, and cul-de-sac) within ½ mile of the household and employment centers of mass in each zone.

Task 5 Deliverables: software and other tools necessary to implement the automation of all input data included in the input automation plan described in the Task 1 deliverables, and detailed documentation of all those tools. Task 6: software upgrades for faster, easier model operations. The purpose of this task is to identify and implement upgrades to the Focus model software, for use at a statewide level, that will result in shorter run times of the model. When software was being developed for DRCOG’s Focus model, considerable attention was paid to model run times. At project inception (2005), the initial target was a model that would run once through in 24 hours, and (with the aid of a powerful server) this target was achieved. However, several factors suggest that there still is much practical room for improvement in Focus run times: the development team was not expert in optimization of software/database interaction, a central aspect of Focus operations; software development for multi-processor computers was quite new at the time, and much progress has been made in that area since then; and various algorithmic and theoretical enhancements have been made since that time to activity-based travel models that promise real run time improvements. Examples of possible performance upgrades include:

The Focus model location choice models include all zones in the choice set. There is some indication, based on the experience of other ABMs, that location choice models that use choice set selection run faster than those using all zones.

Discussions with others in the travel modeling field suggest that there may be faster algorithms to implement Monte-Carlo choice selection (an operation that runs billions of times in models like Focus, and therefore is a possible source of run time enhancement.)

Support for “threading” (i.e., distributing calculations to multiple processors) in programming languages (in the case of the Focus model, the C# language) was in its infancy at the time that Focus was developed. Programming languages have made significant upgrades to their support for threading since that time, which may translate into faster performance.

Many applications and data warehouses are being moved to the “cloud”. There is some possibility that such an operational environment, allowing the use of potentially very large cloud hardware arrays, can also permit significant improvement in model run times.

Focus is written at present to make frequent database reads and writes with small amounts of data. Speed may be enhanced by storing more model inputs and outputs in memory and making fewer database calls.

The contractor shall work with CDOT staff to develop a list of software and hardware enhancements to be investigated, which shall at minimum include those listed above. The contractor shall then investigate and test software enhancements for all elements included in this list. The performance target that will guide this work is that the statewide model should run once through (i.e., one run with no speed balancing iteration) in 24 hours or less. It should be noted that CDOT anticipates wide use of the statewide model by partner agencies, consultants, etc., so cost of any suggested hardware upgrades should be investigated carefully, as in some scenarios for deployment of the model to users, the overall cost of the operational environment may be multiplied by the number of users in the community. As part of this task, the successful proposer will obtain the Focus model from DRCOG and install the model on a development environment selected by CDOT, to permit operational tests of software enhancements they suggest during the project. Task 6 Deliverables:

A list of software enhancements to be investigated for their performance improvement effects, to be approved by CDOT.

For all enhancements from the above list that are selected for inclusion in the final model, all software necessary to implement the enhancements, fully incorporated into the overall model software system, and detailed documentation of the enhancements and their effects.

17

Task 7: model installation and implementation support. The purpose of this task is to evaluate deployment options for the statewide version of Focus, to select one of those options, and to deploy the model according to the selected approach. The final step of the project will be to implement the model on the selected platform. Several implementation solutions are possible (such as cloud-based operations.) The contractor will work with CDOT and State of Colorado IT personnel to install the model and test it to ensure that it is operating correctly. The contractor also will provide initial training in model operation to CDOT staff identified by the project team. This task will involve several test model runs, with analysis of the results provided to demonstrate its proper function. The goal of this task is to implement a deployment solution that will permit practical use of the model by CDOT staff, consultants that CDOT hires for corridor and other studies, partners agencies, etc. Note that, at CDOT’s discretion, on-call consultants available to OIT may assist in the analysis, selection and implementation of a deployment approach. Activities in this task shall include:

Close examination of model characteristics relevant to the choice of deployment strategy. Characteristics and/or issues likely to be relevant include (but are not limited to):

o Licensing requirements for the several commercial off-the-shelf (COTS) software elements used in the model (particularly TransCAD and Microsoft SQL-Server.)

o Installation challenges associated with the custom C# code, and with the COTS softwares.

o Model run time requirements. CDOT has established a minimal target of 24 hours to execute one complete iteration of the model (not including any speed balancing iterations that may be necessary), with faster performance than this being highly desirable. Note that the Focus model currently runs in 24 hours for the DRCOG region, using a 32-core server with 64 GB RAM. Faster performance will be required to achieve 24-hour performance at the larger state scale.

o Model disk storage requirements. At full scale (including all input and output files with no zipping or other compression approaches) one model run requires approximately 70 GB of hard drive space, and planning and engineering projects on which the model may be employed sometimes execute over 100 model runs.

o Threaded/distributed processing. The Focus model currently is written to operate simultaneously on multiple processors, and to detect and use however many processors are available. These processors need not be on the same “box.” However, it is likely that the Focus code can be re-factored to perform this function faster and more efficiently.

o Interaction between the algorithm code (primarily in C#) and the data repositories in TransCAD proprietary matrix formats and in Microsoft SQL-Server. As currently written, the Focus model performs many thousands of data repository read/write operations. It also is likely that this aspect can be re-factored for enhanced efficiency.

o Need for and approaches to deploying updated versions of the model as they become available.

Review of deployment options in light of the above model characteristics. Options to be reviewed will include (but may not be limited to):

o Install the model on the hardware of users (typically consultants hired by CDOT or partner agencies.) This was the approach taken by DRCOG with its older generation of models (prior to 2010), and DRCOG currently is testing this approach on a project using the Focus model. Among other issues, automating model installation as much as possible will be key to this option’s feasibility.

o Implement the model as a hosted solution on the hardware of the Colorado Office of Information Technology (OIT). A hosted solution will involve development of a number of capabilities, including account/login and data management for users, provision of data upload/download capabilities, COTS licensing associated with multiple simultaneous users, etc. Use of OIT

18

hardware will involve conformity with OIT security and other requirements. A system to charge users also will be considered and evaluated for feasibility.

o Implement the model as a hosted solution on the hardware of one or another cloud vendor. All the “hosted solution” issues discussed above are still relevant for this option, as are costs associated with the cloud vendor (provision of requested hardware/performance, data upload/download costs, etc.)

Selection of one of these options (or other that the proposer may suggest), and implementation of the selected option.

o CDOT will lead the selection of one of these options, with technical support from the contractor.

o Contractor will lead the implementation of the statewide model on whatever deployment platform CDOT selects.

Task 7 Deliverables: A report documenting the process and findings of evaluating the deployment options, with “pro and con”

analysis, and the contractor’s recommendations for a deployment approach.

All software developed to support and facilitate installation of the model on the selected deployment environment, together with detailed documentation describing the steps in deployment, supporting software, etc.

A fully-installed model on the selected environment.

1.31 CDOTs Desired schedule for project.

SCHEDULE

1st qrtr

2nd qrtr

3rd qrtr

4th qrtr

5th qrtr

6th qrtr

7th qrtr

8th qrtr

Task 1 – Review CDOT requirements, and group by priority and feasibility

Task 2: Preparation of input/estimation/calibration/validation data

Task 3: Adaptations of regional model for state-level application

Task 4: Enhancing Focus functionality for state-level modeling

Task 5: Enhanced automation of model input development

Task 6: Software upgrades for faster, easier model operations

Task 7: Model installation and implementation support

19

1.32 Submission of Confidential/Proprietary Information

The State neither requests nor encourages the submission of confidential/proprietary information in response to this RFP. Information submitted will be open for public inspection. However, written requests for confidentiality can be submitted to the CDOT Purchasing Director, provided that the submission must be in STRICT accordance with the following procedures. The submission of information in strict accordance with such procedures shall be the SOLE RESPONSIBILITY of the proposer.

PROCEDURE FOR SUBMISSION:

A. A written request for confidentiality shall be submitted, by the proposer with the proposal documents. B. The written request will be enclosed in an envelope marked “REQUEST FOR CONFIDENTIALITY”,

and attached to the cover of the ORIGINAL copy of the proposer’s proposal that contains the invitation for proposal page with the proposer’s ORIGINAL autographic signature.

C. The written request must state SPECIFICALLY, AND IDENTIFY BY PAGE NUMBER, what elements of the proposal are to remain confidential. The request must also IDENTIFY THE BASIS for the claim of confidentiality, OTHER than a recitation of a SPECIFIC State or Federal statute.

D. Confidential/propriety information MUST be readily IDENTIFIED, MARKED and SEPARATED/PACKAGED from the rest of the proposal. Co-mingling of confidential/propriety information and other information is NOT acceptable.

E. The CDOT Purchasing Director will make a written determination as to the apparent validity of any request for confidentiality. The written determination of the Purchasing Director will be sent to the proposer.

E. Proposals that are determined to be at variance with this procedure may be declared non-responsive by the Purchasing Director, and not given further consideration.

1.33 ORAL PRESENTATION/SITE VISITS:

Proposers may be asked to make oral presentations to the evaluation committee. Such presentations will be at the proposer’s expense.

1.34 PROPOSAL PRICES: Estimated proposal prices/amounts are not acceptable. Best and final offers may be considered in determining the apparent successful proposer, if requested, by the evaluation committee after oral presentations.

1.35 RFP CANCELLATION: The State reserves the right to cancel this Request for Proposal at any time, without penalty. 1.36 PARENT COMPANY:

If a proposer is owned or controlled by a parent company, the name, main office address and parent company’s tax identification number shall be provided in the proposal.

1.37 ASSIGNMENT AND DELEGATION:

Except for assignment of antitrust claims, neither party to any resulting contract may assign or delegate any portion of the agreement without the prior written consent of the other party.

1.38 VENUE:

The laws of the State of Colorado shall govern in connection with the formation, performance and the legal enforcement of any resulting contract. Further, Title 24, C.R.S. as amended, Article 101 through 112 and Rules adopted to implement the statutes govern this procurement.

20

SECTION 2.0 INFORMATION REQUIRED FROM PROPOSERS

General Proposer Response Format

2.1 BACKGROUND:

This section of the proposal, should demonstrate the proposer’s understanding of, and approach to, the described services, specifically addressing how each element of the Statement of Work will be accomplished. The proposal also should include details of the following:

2.2 CONFLICT IDENTIFICATION:

Identify all current and former contract activity with any existing State agency or transportation authority related to this Project. Indicate when involvement occurred and length of activity, type of activity with such transportation authority, and indicate extent of involvement with such entities.

2.3 WORK EXPERIENCE:

The proposer’s work experience described in the proposal should include: Designing and implementing activity-based models Designing and implementing statewide models Software development experience, including object-oriented code and relational database design

and optimization Designing and implementing software deployment systems, including multi-user hosted systems Use of data sources that support state and larger-scale modeling, particularly covering

geographies for which standard household diary survey data is not available.

2.4 FORMAT FOR PROPOSALS – Firms responding to this RFP should address the following items in no more than 30 pages (excluding appendices) per submission – for example 30 pages for cover letter, description of staffing and brief biographies of key staff members, description of firm experience (including short project summaries), description of project approach, description of proposed schedule, etc. While the 30-page limit is not a hard one, proposers are discouraged from exceeding it, from the use of small fonts and small page margins, etc. Appendices may include full resumes of proposed participants, descriptions of relevant projects from the firms’ experience, and brief technical memoranda from such projects that the proposers may wish to include. Proposals must be flat bound (i.e., no three ring binders or similar binding methods.) Either in appendices or in the main body of the proposal, responses should include:

Staffing/experience

Names and experience of project/task leaders. Names and experience of key “hands-on” staff who will be conducting the bulk of the technical

work of the project. Relevant experience of the broader firm(s), which may potentially be used to support successful

project completion. Project Approach

A detailed description of the proposer’s approach to all work program tasks listed in Section 1.3. Discussion of the proposer’s overall vision of an easier-to-use statewide activity-based model,

and of their approach to achieving that vision Discussion of the proposer’s estimate of fastest feasible run time (achievable during the next

three to five years), and of their approach to achieving that run time. Discussion of what applications the proposed model configuration will support, and what

applications it will not. Discussion of content of and approaches to implementing future enhancement, including

enhancement through future corridor (and similar) studies. References

Please list three relevant references, for which you have provided similar services during the last five years. Include the name of the organization, contact person, phone number, e-mail address, contract number and a brief description of the services provided.

21

2.5 FEES AND EXPENSES Proposers should describe in detail their costs to implement their proposed project approach, including:

Separate costs for each task listed in Section 1.3. Costs for travel and other major direct cost categories. Hourly rates for all project/task leaders, and any other staff with significant commitment to the project, and

either specific rates or averaged rates by employee category for all other staff who may participate. Approximate total committed hours for all staff with significant commitment to the project. Overhead rates and fee for team firms.

2.6 VETERANS PREFERENCE:

Pursuant to C.R.S. 24-50-511, the State shall give consideration to proposers utilizing a preference for hiring veterans of military service only in the following manner:

To break a tie between proposals following review, scoring and ranking by the evaluation committee. Such tie shall broken by awarding the resulting contract to the proposer utilizing the greatest quantitative (numerical) preference for veterans in the hiring of its employees.

Veterans’ preference will not be used as a scored criterion in the evaluation and ranking of proposals received in response to this RFP solicitation.

22

SECTION 3

EVALUATION CRITERIA

3.1 AWARD OF BID:

This section will outline the evaluation criteria to be used by the evaluation committee in the review, rating, and selection of submitted proposals. After evaluation of the written proposals, CDOT may request oral presentations from top-ranked proposers. The highest ranked proposals will be given fair and equal treatment during the second (optional) phase of the evaluation. Oral presentations will not be scored separately as part of this solicitation. Oral presentations will only be used to adjust the proposal scores accordingly (per the same evaluation criteria), based upon the information discussed at the presentation. The top-ranked proposal(s) (following conclusion of all phases of the evaluation) shall be recommended, by the evaluation committee, to enter into contract negotiations. The contract(s) will be awarded to those proposer(s) whose proposal (conforming to the RFP) is/are deemed to be the most advantageous to the State of Colorado, price and other factors considered.

3.2 EVALUATION PROCESS:

3.2.1 Each member of the evaluation committee will first independently evaluate the merit of proposals received in accordance with the evaluation factors defined in the RFP, followed by panel discussion and final scores ranking. The recommendations of this committee will then be forwarded by the purchasing agent, to the CDOT Procurement Director for review and final approval.

3.2.2 Failure of the proposer to provide any information requested in the RFP may result in

disqualification of the proposal as nonresponsive. It is the responsibility of the proposer to provide all information required by this RFP.

3.2.3 The sole objective of the evaluation committee will be to recommend the proposal most

responsive to the State of Colorado’s needs. The specifications detailed in this RFP represent the minimum performance necessary for such response.

3.2.4 The top ranked proposal(s) (highest score(s)), following independent review and panel

discussion, will be recommended either for award or, if the evaluators deem in appropriate, to make an oral presentation.

3.2.5 Proposal Scoring: The sole objective of the evaluation committee will be to score the responses

and recommend the proposer(s) whose proposal is/are most advantageous to the State of Colorado, taking into consideration all evaluation factors set forth herein. Following independent review and panel discussion, the successful proposer(s) will be the one(s) accumulating the highest number of points (of a maximum 100) at the conclusion of the final stage of the selection process and whose proposal(s) is/are deemed most advantageous to the State, and who successfully negotiates the ensuing contract.

3.3 EVALUATION CRITERIA:

The complete proposal package will include, but not be limited to, evaluation using the factors listed below. These factors are designed to incorporate specific evaluation of the items presented in Section 1.30 and Section 2 of this RFP. As stated in Section 2.1, proposals should not simply repeat what is written in Section 1.30 of this RFP – the Statement of Work, but rather evidence the proposer’s understanding of the State’s requirements and its ability to provide the services needed within a clearly defined and cost-effective budget. (Refer to Section 2 of this RFP).

23

1. Understanding and Approach to the Project: (30%) a. Understanding of principal modeling and planning issues across Colorado that the

statewide model must support, and approaches to model development that will support them in the most practical and effective manner possible.

b. Approach to data acquisition/development/use to support model development and calibration/validation outside the Front Range region.

c. Approach to adapting a region-level activity-based model for use at a statewide scale.

d. Understanding of and approach to automating model input data, and other steps to make the model as easy to operate as possible.

e. Understanding of and approach to optimizing activity-based model software to improve run times.

f. Understanding of approaches to enhancing the model in future years/projects.

2. Overall Experience & Capabilities: (30%) a. Experience with design and implementation of activity-based models. b. Experience with design and implementation of statewide models. c. Experience with software design/development and performance optimization. d. Experience with relational database design/development and performance

optimization. e. Experience with software deployment systems, including separate installations on

local systems, and hosted systems remotely accessed by multiple users.

3. Qualifications and Ability to Provide Services: (20%)

a. Qualification and commitment of the firm(s) b. Qualification and commitment of lead personnel. c. Qualification and commitment of hands-on technical and support personnel

4. Budget and Narrative: (20%) (can be more, but not less)

a. Information is provided as requested in Section 2.5 and no unacceptable

modifications to the terms and conditions outlined in this RFP are proposed.

5. Feasibility and Completeness: (5%)

a. The proposal is both adequate and complete, as defined through the RFP. b. The proposal inspires confidence in production of a quality-required product, solicited

under the RFP.

24

Attachment A

COLORADO DEPARTMENT OF TRANSPORTATION

CONSULTANT CERTIFICATION CDOT Form #637

Consultant firm name Consultant firm complete address Authorized representative name (print)

Title

I certify that neither I nor the above firm I represent has: - employed or retained for a commission, percentage, brokerage, gift, contingent fee or other consideration, any firm or person (other

than a bonafide employee working solely for me or the above consultant) contingent upon or resulting from the award or making of this contract, or to solicit or secure this contract:

- agreed as a known or implied condition for obtaining this contract, to employ or retain the services of any firm or person in connection with carrying out the contract; or

- paid, or agreed to pay, to any firm, organization or person (other than a bonafide employee working solely for me or the above consultant) any fee, contribution, donation, or consideration of any kind for, or in connection with, procuring or carrying out the contract;

I further certify that: - any exceptions to the conditions listed above are: - wage rates and other factual unit costs supporting the compensation to be paid under this contract are accurate, complete and

current. I acknowledge that this certificate is to be furnished to the Colorado Department of Trasportation and the Federal Highway Administration, U.S. Department of Transportation, in connection with this contract involving participation of Federal Aid highway funds, and is subject to applicable State and Federal laws, both criminal and civil. I DECLARE UNDER PENALTY OF PERJURY IN THE SECOND DEGREE, AND ANY OTHER APPLICABLE STATE OR FEDERAL LAWS THAT THE STATEMENTS MADE ON THIS DOCUMENT ARE TRUE AND COMPLETE TO THE BEST OF MY KNOWLEDGE. Firm representative signature Date

CERTIFICATION OF THE COLORADO DEPARTMENT OF TRANSPORTATION

I certify that I am the duly authorized representative of the Colorado Department of Transportation and, that the above Consulting firm or its representative has not been required, directly or indirectly as a known or implied condition in connection with obtaining or carrying out this contract to: - employ or retain, or agree to employ or retain, any firm or person; or - pay, or agree to pay, any firm, person, or organization, any fee, contribution, donation, or consideration of any kind; I further certify that: - any exception to the conditions listed above are: I acknowledge that this certificate is to be furnished the Federal Highway Administration, U.S. Department of Transportation, in connection with this contract involving participation of Federal Aid highway funds, and is subject to applicable State and Federal laws, both criminal and civil. CDOT representative signature and title

Date