GovKM
Information Continuity

FHWA Experiment Records: Make the Supporting Evidence Reachable

GAO found that practitioners lacked clear instructions for obtaining FHWA experiment details. GovKM proposes a maintained route from database entries to supporting records, accessible responses, and accountable ownership.
Conceptual illustration of a transportation experiment database entry connected to research plans and supporting records, with a pedestrian crossing in the background; not actual FHWA systems or records.
Expand image

A transportation engineer finds an experiment in the Federal Highway Administration's Official Rulings database. The entry concerns a traffic-control device the engineer wants to understand. Before preparing a new request, the engineer needs the research plan, the conditions under which the device was tested, and the reported results.

A searchable entry helps identify the experiment. The information service must also explain how to obtain the material behind it.

The access problem GAO identified

In its November 16, 2026 report, GAO found that FHWA's database offered limited experiment detail. FHWA told GAO it supplied additional documents upon request, but the website did not clearly explain that option and many interviewed practitioners were unaware of it. Some case-file links had previously been removed; officials cited server capacity and the resources needed to make externally supplied documents accessible. [1]

GAO recommended prominently displaying contact information and clear instructions for obtaining more detailed experiment information. DOT agreed. The recommendation remained open when checked November 17, 2026. The interviews were a non-generalizable sample, and the report does not establish that the underlying records were lost.

The continuity challenge: preserve the route from discovery to evidence

GovKM interprets this as a specific information-continuity problem. A public index and the records it describes can remain available in separate places while the route between them becomes difficult to use. A change in hosting, document format, or staff responsibility can interrupt that route.

The likely desired outcome is straightforward: a practitioner can identify an experiment, understand what supporting information may be available, and reach an accountable office that can provide an authorized response. The next employee handling that service should inherit the pending requests and the means to locate the records.

This work should begin with GAO's requested communication improvement. The agency can assess additional record links and workflow controls against actual demand, accessibility obligations, and available resources.

GovKM's recommended solution within a governed federal environment

GovKM would propose a bounded pilot covering a small set of experiment entries and the process used to respond to information requests. Program personnel would work with the agency's web, accessibility, records, security, and disclosure specialists to establish the applicable requirements and approve the service design.

Explain the request at the point of discovery

Each selected database entry would retain its experiment identifier and provide a prominent route to request instructions. Those instructions would describe the kinds of supporting material that may be available, the identifier a requester should supply, the responsible office, and how to seek assistance with an accessible format.

The instructions would distinguish requests for existing experiment information from applications to conduct a new experiment. FHWA's experimentation page describes the latter process and states that agency approval is required before an experiment begins. Obtaining another jurisdiction's research material does not grant that approval. [2]

The service owner would review the instructions when contact arrangements or repository locations change. A shared, agency-managed channel could support continuity through staff turnover where approved. A named backup role would prevent an employee's departure from leaving requests without an owner.

Connect the index to controlled supporting records

For each pilot entry, an authorized employee would establish which request, approval, research, progress, or final-result records are actually held. The inventory would identify a controlled location, document date and version, responsible custodian, and any release or accessibility review still needed. A document that cannot be located would be recorded as unresolved.

Public metadata and internal working records could remain separate. The public entry would disclose only approved information; authorized staff would retain the relationships needed to retrieve the underlying material. Direct publication would follow accessibility and disclosure review. Where publication is not appropriate or feasible, a maintained request route would remain visible.

Consider a fictional city requesting a final report for a past pedestrian-signal experiment. Staff locate an interim progress report but cannot initially establish whether a final report was received. The response should accurately identify the available document and its status. The unresolved search would retain an owner and supporting history. Labeling the interim report as final to complete the request would undermine the service.

This example illustrates GovKM's proposed controls. It is not an additional finding about FHWA.

Maintain the service through change

An approved response would record which documents were supplied and which questions remained unresolved, with access to requester information restricted as appropriate. A later employee should be able to continue the request without relying on the former employee's personal files or recollection.

Repository migration would include checks that experiment identifiers still resolve to the correct records and that public instructions remain usable. Document replacement would preserve the earlier record and its status under applicable retention requirements. Accessibility remediation would retain the relationship to the source document and verification of the resulting content.

Where implemented and validated, OrgAI could help maintain these relationships and flag affected entries when records, contact responsibilities, or document status change. Authorized officials would continue to make release, engineering, and compliance decisions.

Make compliance an evidenced operating condition

The federal records basis includes 44 U.S.C. § 3101, which requires adequate documentation of agency work, and 36 CFR § 1236.10, which addresses controls for electronic records. The pilot would apply the agency's approved schedules, access rules, and preservation requirements to the records within its scope.

Accessibility must be included in the workflow. The Access Board's Revised Section 508 Standards establish applicable federal ICT accessibility requirements. Specialists would determine the requirements for the particular website, documents, and delivery method and preserve the results of appropriate verification. A request channel should not be treated as a blanket exemption from those requirements.

GAO's 2025 Green Book, effective beginning in fiscal year 2026, provides the federal internal-control framework. The pilot would connect each applicable requirement to the responsible control, test evidence, and determination. GovKM's proposed configuration would require that assessment before any scoped compliance claim.

Demonstrate that a new user and a successor can complete the task

Acceptance would use a small set of approved or synthetic cases. A practitioner unfamiliar with the website would try to find request instructions from a selected experiment entry. A responding employee would locate the correct document and distinguish its status. An accessibility specialist would verify the applicable delivery requirements.

A handoff case would transfer an unfinished request to a successor. A migration case would test whether the same experiment identifier still leads staff to the correct supporting records. The evidence would capture failures and limitations as well as successful results.

GovKM could support this limited information-service assessment, record mapping, workflow configuration, and validation. The deliverable would show whether practitioners can reach the evidence and whether the agency can maintain that service as people and systems change. Traffic engineering judgment and permission to deploy devices would remain with the responsible authorities.

Sources and scope

[1] GAO, Traffic Control Devices Manual: Selected Practitioners Reported Never or Rarely Deviating from DOT Standards, GAO-27-108431, November 16, 2026. The full report discusses the database, removed case-file links, request pathway, recommendation, and agency response. Recommendation status checked November 17, 2026. This article characterizes the reported findings; it does not claim a complete audit of FHWA's current website.

[2] FHWA, Experimentations. The procedure for seeking approval to conduct an experiment is distinct from requesting information about an existing one.

The workflow and examples are GovKM proposals, not an implemented FHWA solution, agency endorsement, or confirmed procurement opportunity.

‍

RELATED KNOWLEDGE

Continue Exploring

Explore related research, framework domains, and continuity concepts.
CONTINUE WITH THE FRAMEWORK

Explore the continuity relationships that support trustworthy organizational intelligence.

Continue through the GovKM Framework to examine the doctrine, knowledge, and implementation guidance behind Organizational Continuity.