J.1_SOO_Parametric_Modeling_final_v3.pdf

U.S. Department of Transportation Office of Railroad Development Office (RRD) Railroad Parametric Capacity Model (RPCM)

Statement of Objectives (SOO)

1. Background

Since the 1975 publication of Federal Rail Administration’s (FRA) Parametric Analysis of Railway Line Capacity Final Report, parametric railroad capacity models have seen sporadic use throughout the North American railroad industry as a tool to support both short-term and long-term network, traffic, and capital planning. While some individual railroads have developed and maintained propriety parametric models for their own internal use, the publicly available 1975 FRA model, which provided the initial conceptual framework for this approach to capacity analysis, has become largely obsolete due to major changes that have occurred over the past 50 years including railroad operating practices, traffic types, physical plant characteristics, and capacity utilization.

During that same period, detailed railroad operations simulation emerged as the dominant tool for railroad capacity analysis. FRA has been a major proponent for the use of operations simulation an essential means of assessing the efficacy of proposed changes to railroad operations and infrastructure. Since most railway improvements require careful consideration of railway line capacity and operating performance, the need for an expanded set of tools for railroad capacity analysis has grown more important.

The U.S. Department of Transportation’s Federal Railroad Administration (FRA) RRD-20 seeks to develop a new Parametric Railway Line Capacity Model (PRLCM) which, like the 1975 model, would be made publicly available for use by both the railroad industry and by others with an interest in railroad capacity analysis (such as sponsors of potential railroad improvement projects).

The first step towards this objective is to develop the parametric railway line capacity model and test it prior to undertaking the project to publicly launch the tool.

Like the 1975 model, development of a new parametric capacity model would focus on an analytical examination of key factors affecting railway line capacity to:

• Determine the relative magnitude of the impact of varying key parameters such as track and signal system configuration and train operating policy in terms of their effect on capacity; and

• Develop a parametric model for general application to railroad capacity analysis.

As was the case in the development of the 1975 model, a key element of a new parametric model would be the use of a railroad operations simulation model to examine different combinations of track, signal, and train configurations and operating practices, with the resulting estimates of operating performance being used as the basis for a regression-based approach to establish the mathematically continuous relationships between various parameters and line capacity. Some of these relationships had only been previously addressed as discrete variables.

In addition to reflecting the contemporary characteristics of North American railroad operating practices, infrastructure, and traffic, the development of a new parametric capacity model would present the opportunity to incorporate advances in modeling simulation methods and address topics that were unable to be addressed as part of the 1975 model due to the limitations of the analytical tools available at the time.

FRA anticipates that a new parametric capacity model would have the potential to serve as an important adjunct to existing railroad operations simulation tools, to support high-level operations analysis for corridor- level investment studies. The evaluation of service alternatives necessitates the consideration of performance, financial, and environmental trade-offs associated with varying service levels. Equipping project sponsors and host railroads with analytical tools that can provide preliminary estimates of available capacity will help facilitate the advancement of appropriately scaled service options for funding consideration.

2. Purpose

The Contractor shall develop the Parametric Railway Line Capacity model (PRLCM), including a documented algorithmic specification, reproducible model source code, and supporting documentation sufficient for Government-led application development. The Contractor’s responsibility is limited to model design, estimation, testing, validation, and delivery of code and documentation; it does not include development, hosting, security accreditation, or operation of production desktop or web-based applications. Following delivery and conditional acceptance of the model and all associated artifacts, the Government will be responsible for any production application development, deployment, hosting, and public release derived from the contractor-delivered artifacts, unless the Government explicitly requests and separately authorizes additional contractor work via a contract modification or separate CLIN(s).

Offerors shall propose a Performance Work Statement (PWS) that aligns with the objectives and outcomes stated in this SOO. The Government seeks innovative, tailored approaches to achieving these outcomes and strongly discourages simple regurgitation of the SOO language (e.g., Scope of Services, Potential Deliverables, Potential Performance Metrics) or any previously released draft requirements documents (e.g., Draft SOW). The proposed PWS will be incorporated into the resultant order (if accepted by the Government) and should reflect the offeror’s recommended solution, methods, and performance approach, and include:

• Proposed tasks and activities designed to achieve the outcomes;

• Proposed deliverables and performance objectives/metrics aligned to their approach;

• A clear explanation of how the proposed solution supports DOT’s mission objectives for

Parametric Model.

The Government encourages offerors to fully exercise their flexibility to propose a PWS that demonstrates creativity, efficiency, and effectiveness.

Mission-Aligned Objective This effort will result in a railroad parametric capacity model that will be validated based on “real world” operations data.

3. Strategic Outcomes

The Contractor shall support DOT and RRD in achieving the following outcomes:

• Deliver a validated, reproducible Railroad Parametric Capacity Model (RPCM), associated documentation and code such that the Government can independently develop any production desktop or web-based applications.

• Develop an RPCM model that provides functionality needed to support operational analysis at early stages of corridor planning.

4. Objectives

Objective Area 1: Parametric Methodological Scoping and Development

Objective 1.1: Establish Methodological Guidelines

Desired Outcome: To provide transportation practitioners and researchers with a consistent methodology for the evaluation of the quality of service/capacity of the network on the rail network. The Contractor shall adopt an iterative development approach that supports regular Government reviews and feedback.

Specific expectations include:

• Monthly progress reports and demos of incremental model functionality.

• Delivery of interim drafts for major deliverables per schedule in the SOO/PWS, with specified

Government review windows.

• Maintain an action item tracker and documented responses to Government comments.

• Participation in scheduled working sessions with FRA subject matter experts for model scoping, validation test case selection, and final acceptance reviews.

Scope:

DOT seeks support to determine analytical needs and functionality for an updated Parametric Model

Potential activities to achieve the desired outcome may include, but are not limited to, the following:

• Identify critical external stakeholders (Class I railroads, DOTs, etc)

• Review and assess key elements and prior experience and industry practice

• Identify objectives and desired tradeoff logic related to contextual factors such as freight demand and fluidity, maintenance access, prioritization, and variable speeds

• Develop scenario configurations and use cases to be incorporate at the design level

• Assess and select railroad simulation software to generate inputs for estimating the Parametric model.

Potential Deliverables:

• Technical report Assessing prior efforts and the State of the Practice for Parametric Models

• Technical report on the Assessment and selection of simulation modeling software to support

Parametric model development

• Technical report on desired functionality and performance measures based on stakeholder input

• Report summarizing strategies for engaging stakeholders (e.g. Class I railroads, state DOTs)

Potential Performance Metrics

• 95% of planning deliverables accepted after no more than two (2) submissions for review to FRA

• Milestone alignment confirmed by Contractor’s Program Manager within 10 business days of schedule changes

Intended Roles:

• Contractor leads development of all materials in collaboration with Government

• Contractor manages data in space that meets FRA IT Standards

• Government COR serves as technical lead, reviews and approves/rejects planning deliverables

Objective 1.2: Develop Model Specifications, Define Inputs and Outputs Measures

Desired Outcome: Scoping of methodological approaches, develop model structure and assess data sources and variables to be use in model development.

Scope of Services:

Identify Parametric model structure, parameters and inventory of model variables and approach for conducting simulations across a range of train operating (baseline) conditions

Potential activities to achieve the desired outcome may include, but are not limited to, the following:

• Determine parametric modeling methodology

• Develop roster of input variables and approach for conducting simulations o FRA considers the following variables as candidates for inclusion within the Parametric model framework:

Train heterogeneity: power-to-weight ratio, priority, and level of service.

Effects of higher-speed passenger trains (79–110 mph) and interaction across speed ranges and track configuration.

Long trains that exceed the length of most passing sidings on a route.

Number of long sidings that can facilitate meets between over-length trains, and sidings equipped with mid-siding crossovers to facilitate meets between shorter trains.

Schedule flexibility to account for variation in train departure time.

Operating strategies such as train fleeting that can increase capacity of bottlenecks.

New control systems using moving blocks and different densities of virtual blocks.

Incremental capacity of additional double and triple track.

Models or technology that optimize dispatching to minimize delays Increased time requirements for on-track maintenance.

• Review and define potential quality of service and capacity measures for intercity passenger rail operations

• Develop methodology for conducting simulation runs that replicate variability across a broad range of infrastructure and operating conditions

• Identify data sources needed for model validation.

• Report assessing potential measures of railroad capacity to be estimated by Parametric model (in consultation with FRA)

• Report identifying key variables to be tested in model development (for example: operational, technological, train types)

• Data management plan that identifies resources to support the development and validation of the Parametric model o Note: The plan should address management and documentation of all input data, output data, and code including data and code maintenance, storage, security, access, sharing, automation, commenting, reproducibility, quality control, and supporting information and metadata collected and generated in completing the tasks of this contract. FRA will assist the contractor on activities related to the collection and management of data resources needed develop and validated the Parametric model.

• Report describing a technical approach develop ‘baseline’ scenario(s) and microsimulation approach to capture variability and interactions between input variables (identified in objective 1.1).

• Report summarizing the approach for assembling and translating simulation outputs into a model estimation data set.

Ongoing Adaptive Support:

• Coordination with Class I railroads, state DOTs

Potential Performance Metrics:

• Monthly dashboards delivered on time with accurate data

Intended Roles:

• Contractor tracks, maintains, and reports on risk and performance.

• COR and Government Program Manager review, validate, and escalate as needed.

Objective 1.3: Develop Parametric Model Prototype

Desired Outcome: Deliver a foundational Parametric capacity methodology that includes a core algorithm, and non-production demonstration user interface.

Scope of Services:

Potential activities to achieve the desired outcome may include, but are not limited to, development of a Parametric model prototype equipped with the following features:

• Model should be able to rapidly assess corridor capacity and throughput for a baseline case (observed conditions) based on inputs defined in step 1.2.

• Model should be able to estimate impacts based on user defined inputs, reflective of relevant corridor characteristics, and enable comparisons between a ‘baseline’ and ‘build’ alternative. For example, testing of alternative strategies (lengthening sidings, improving train control, modifying freight traffic characteristics) to examine impacts on available capacity.

• Non‑Production Demonstration Interface: Contractor shall provide a limited, non‑production demonstration interface solely for internal Government validation of the Railroad Parametric Capacity Model. The demonstration interface will allow users to input sample parameters and view or export model outputs in a basic, non‑production format. All such demonstration artifacts must be documented as non‑production, will not be deployed for public use, and will be delivered together with code and documentation to RAD‑20 for further development, integration, and production deployment under separate Government purview.

• Model code base developed on open-source scripting language (with sufficient labeling to enable further updates or de-bugging)

• Documentation of model methodology, including model inputs, outputs and key parameters (for FRA internal use)

• Identify requirements for model development environment that complies with FRA IT security and confidentiality requirements.

o All work must be conducted within approved environments. Furthermore, the contractor's access to data is strictly limited to the duration of the Non-Disclosure Agreement (NDA) which is in effect for the current contractual year only. Compliance with these conditions is mandatory for the protection of sensitive information and adherence to agency protocols.

Ongoing Adaptive Support:

• Coordination with Class I’s, project sponsors, state DOTs

Potential Performance Metrics:

• Monthly dashboards delivered on time with accurate data

Intended Roles:

• Contractor tracks, maintains, and reports on risk and performance.

• COR and Government Program Manager review, validate, and escalate as needed.

Objective Area 2: Model Validation

Objective 2.1: Parametric Model Validation

Desired Outcome: Ensure that the Parametric reliable replicated observed, or ‘real world’ conditions and demonstrates a plausible response and sensitivity to changes in input variables. The Contractor shall perform model verification, validation, and acceptance testing using clearly documented development, testing, and validation data partitions. The Contractor shall deliver a Model Validation and Test Report that documents the datasets used, the rationale for the partitioning, the validation scenarios executed, and the quantitative results.

The Contractor shall address all FRA comments following the agreed upon PWS..

Scope:

DOT seeks support associated with the validation and reasonableness checking of Parametric model outputs.

Potential activities to achieve the desired outcome may include, but are not limited to, the following:

• Model validation plan than includes an assessment of statistical methods to verify model accuracy.

• Application of the model for current freight and passenger operations. The number of validation tests should be sufficient to verify model readiness for use across a range of freight and passenger operating conditions.

• Validation Criteria

• Model validation plan

• Model validation results

• 95% of planning deliverables accepted after no more than two (2) submissions

• Milestone alignment confirmed by Contractor’s Program Manager within 10 business days of schedule changes

Intended Roles:

• Contractor leads development of all materials in collaboration with Government

• Contractor manages data in space that meets Federal/DOT/FRA IT and Cyber Security

Standards

• Government COR serves as technical lead, reviews and approves/rejects planning deliverables

Objective 2.2: Parametric Model Documentation and Resources

Desired Outcome: Prepare model for use by FRA. Identify all data, including computer software, computer database, technical data, computer software documentation, code, documentation, datasets, etc.

necessary for the performance of the contract. The Contractor shall produce and deliver the following primary artifacts in electronic form to the Government-designated repository (FRA GitHub or other Government-designated repository) in the folder structure and formats agreed in advance with FRA.

• All data rights (i.e. ownership, licenses, etc.) necessary for the Government to use the parametric model and for the Government to move the parametric model to the production phase. The Government anticipates that Objectives 1.2 and 1.3 (i.e., the parametric model) will be subject to FAR 52.227-17 Rights in Data-Special Works, and all other data / Objectives shall be subject to FAR 52.227-14 Rights in Data-General.

o Document any data and software limitations/restrictions and provide a plan for the Government’s use post-Contract.

• Model documentation report that includes a description of model input variables, and output measures.

• Algorithm Description Document: Full technical specification of the RPCM model including purpose, conceptual model, mathematical/statistical formulations, inputs/outputs, assumptions, parameter definitions and units, expected ranges, and a flow diagram for model execution.

• Parametric Model Source Code: Reproducible, modular, tested, and fully commented source code written in R or Python (contractor to specify language choice during proposal); includes unit tests or scripts demonstrating model execution and sample runs.

o Visual Basic solutions are not acceptable.

• User’s Guide and README: Step-by-step instructions for installing dependencies, setting up the execution environment, running the model on sample data, interpreting outputs, and troubleshooting common errors.

• ETL scripts and sample data: Reproducible, well‑commented ETL scripts (R or Python) that transform supplied raw/example inputs into the model’s required input format, plus a small non‑sensitive sample dataset sufficient to run the model examples; include a README with run commands and dependency notes

• Non-production user interface (described under objective 1.3): All such demonstration artifacts must be documented as non-production, will not be deployed for public use, and will be delivered together with code and documentation to RAD-20 for further development, integration, and production deployment under separate Government purview."

o Examples of acceptable optional non-production demonstration interfaces include:

A local spreadsheet workbook (Excel, Google Sheets) with input fields, macros/scripts, and basic output charts A local R Shiny or Python Streamlit app executed in the contractor’s or

Government’s approved environment A configuration-file driven script that generates a static results table or chart

• Technical Reports and Final Publication Materials, including interim reports and final report(s) documenting methodology, results, stakeholder engagements, and recommended use-cases.

• Code base in open – source programming language made accessible on FRA’s GitHub page

• Conduct peer review and assessment to verify sufficiency of the model use by the broader industry. The peer review should include a range of external subject matter experts and industry partners identified in the stakeholder engagement plan. Comments and recommendations will be summarized in a report and shared with FRA to address any updated or revisions.

• 95% of planning deliverables accepted after no more than two (2) submissions

• Milestone alignment confirmed by Contractor’s Program Manager within 10 business days of schedule changes

Intended Roles:

• Contractor leads development of all materials in collaboration with Government

• Contractor managed model data resources for model development and validation, model source code and outputs from model estimation in a storage environment that complies with FRA IT and Security Standards and requirements

• Government COR serves as technical lead, reviews and approves/rejects planning deliverables

5. HANDOFF & TRANSITION

The Contractor shall deliver all source code, scripts, workflows, metadata, documentation, and final reports to the Government-designated repository in an agreed structure. Acceptance of deliverables by FRA constitutes transfer of the model and supporting materials to FRA for Government use and further development.

Transition Assistance (Post‑Delivery): Following conditional acceptance of the Railroad Parametric Capacity Model, the Contractor shall provide a time‑boxed Transition Assistance service to support Government engineers and the RAD‑20 application development team. The Transition Assistance shall be limited to 80 hours of technical support (remote) over a 60‑day period following acceptance unless otherwise agreed in writing. Activities may include clarifying code implementation, reproducing example runs, assisting with running ETL scripts and unit tests, and walkthroughs of the Algorithm Description Document and README materials. This Transition Assistance does not include production application development, hosting, or security accreditation work and is limited to knowledge transfer and troubleshooting of contractor‑delivered artifacts.

A. Transition Assistance (60 days)

Transition Assistance (Post-Delivery) — Following conditional acceptance of the Railroad Parametric Capacity Model, the Contractor shall provide a time-boxed Transition Assistance service to support Government engineers and the RAD-20 application development team. The Transition Assistance shall be limited to 80 hours of technical support (remote) over a 60-day period following acceptance unless otherwise agreed in writing. Activities may include clarifying code implementation, reproducing example runs, assisting with running ETL scripts and unit tests, and walkthroughs of the Algorithm Description Document and README materials. This Transition Assistance does not include production application development, hosting, or security accreditation work and is limited to knowledge transfer and troubleshooting of contractor-delivered artifacts..

6. CYBERSECURITY & DATA HANDLING

Cybersecurity, Data Handling, and Environment Requirements — The Contractor shall comply with FRA/DOT IT security standards and Government policies for the handling, storage, processing, and transmission of Government and third-party data. In particular:

- Work involving FRA or third-party sensitive or production datasets shall be performed only within Government-approved secure environments and under the applicable Non-Disclosure Agreement (NDA);

the Contractor shall not download, copy, or store such data on external systems, personal devices, or third-party cloud services unless explicitly authorized in writing by FRA IT.

- All data at rest and in transit shall be protected using approved encryption standards and access controls.

The Contractor shall implement role-based access and maintain logs sufficient for auditing access to sensitive data.

- Upon contract termination or expiration, and as required by the NDA, the Contractor shall permanently delete all copies of sensitive or production data from any temporary workspaces and certify in writing to FRA that deletion has been completed and no copies were retained.

- Geospatial data usage shall comply with the Geospatial Data Act (GDA): the Contractor must search and consider existing Federal, State, local, or private geospatial data sources before recommending or performing new geospatial data collection.

- Any software artifacts delivered for Government use must be accompanied by a security considerations section describing known vulnerabilities, dependencies, and recommended security hardening steps for production deployment.

7. SECTION 508 / ACCESSIBILITY

Section 508 Accessibility — Any contractor-delivered Information Communication Technology (ICT) artifacts, user-interface prototypes, or documentation intended for public consumption must comply with Section 508 Revised Standards (36 C.F.R. § 1194). If the Contractor produces a demonstration prototype (non-production), the Contractor must (a) document accessibility limitations and known non-compliances,

(b) provide remediation guidance and estimated effort required to reach full Section 508 conformance for production deployment, and (c) ensure that all publicly distributed documentation is provided in accessible formats.

8. PERSONNEL EXPECTATIONS

The Government places high importance on subject-matter readiness; therefore, any contractor staff without demonstrated experience in railroad operations, railroad simulation modeling and development of statistical and empirical methods for freight operational analysis, will not be deemed qualified to undertake this work.

Personnel: Team Lead It is expected that the contractor’s Technical Team Lead will demonstrate proven in-depth knowledge of core subject matter areas noted above.