Road condition

What PAS 2161 means for road condition data

What PAS 2161:2024 requires, when it applies, which technologies are approved, and what non-accredited road condition data can and cannot be used for.

Disclosure, so you can read the rest of this page knowing where it comes from: Make Sense is not PAS 2161 approved. Our full position is at the end.

What PAS 2161 is

PAS 2161:2024, Road condition monitoring (RCM) — Data specification, was published by BSI on 30 September 2024, sponsored by the Department for Transport and technically authored by TRL (BSI, TRL). A PAS is a Publicly Available Specification — a fast-track document, not an Act of Parliament. The obligation does not come from the PAS itself. It comes from DfT attaching the PAS to how local highway authorities in England report road condition.

The substantive change is that PAS 2161 is technology-agnostic. It specifies the data, not the machine. Where SCANNER effectively mandated one survey type, PAS 2161 sets out condition categories, network coverage, processing and validation, reporting format and frequency, and a route by which any technology can demonstrate it produces comparable results (DfT guidance).

What changes in the numbers

Condition moves from red/amber/green to five categories, based on the treatment an engineer would judge necessary to return the carriageway to good condition:

No deterioration — no maintenance considered
Minor or aesthetic deterioration — light maintenance, minor patching
Moderate deterioration — localised or preventative treatment, surface dressing, crack sealing
Moderate to severe — rehabilitative treatment, resurfacing or multiple patching
Severe — structural maintenance, full resurfacing or reconstruction

DfT is explicit that the new data will not be directly comparable with SCANNER output, and that the categories do not map onto red/amber/green percentages. Expect a break in your time series, and plan how you will explain it to members.

Coverage requirements, for data collected within the previous two financial years: at least 90% by length of motorways and A roads in both directions, 85% of B roads in both directions, and 80% of C roads in one direction.

Who approves what

DfT owns the approvals process. In May 2026, TRL was appointed by DfT as PAS 2161 approvals auditor, running independent assessment through to 31 December 2030, on a two-year re-approval cycle (TRL). Approval is granted by survey or technology, not by supplier — a company being on the list does not mean everything that company sells is approved.

The first approved list was published in September 2025 and is maintained by UKRLG: nine technologies from seven suppliers, valid to 31 March 2027 — AISIN (Michilog), Gaist (Carriageway Condition), Metricell (two: an AI agent and a human agent variant), Route Reports (RouteReports RCM), Vaisala (RoadAI), WDM (PASWFE) and XAIS-PTS (MFVAI and AEI) (UKRLG). Check that page rather than this one before you procure — it is the authoritative list, and further demonstration trials are running in 2026 on networks in Cheshire West and Chester, Lincolnshire and Nottinghamshire.

The timeline

April 2025 – March 2026: collect under SCANNER, report spring 2026.
April 2026 – March 2027: collect in the 1–5 format using an approved technology. Report spring 2027; first PAS 2161-based official statistics expected autumn 2027.
April 2027 – March 2028: transition complete.

One carve-out matters. If you hold a multi-year contract that began before the approved list was announced, DfT allows you to see it out. Any new procurement must use an accredited survey.

What this means for data from a system that is not approved

Plainly, and without hedging.

What it cannot do. Your national condition return to DfT, and any condition figure you publish as your official statistic, must come from an approved survey. Data from an unapproved system cannot be reported as PAS 2161 condition categories, and no amount of internal confidence in it changes that. If it would appear in a return, in a transparency report, or in a business case as your certified condition position, it has to come from the approved list.

What it can legitimately do. PAS 2161 governs national reporting. It does not govern everything you know about your network. DfT states that the primary purpose of RCM data remains informing your own maintenance strategy, and that it sets no constraints on additional data collected alongside the minimum — content and format are for you and your supplier to agree. Nor does the PAS replace your safety inspection regime, which is set by your own policy. So non-accredited data can reasonably be used for internal screening and triage, for prioritising where officers look, for targeting where to send an accredited survey, for corroborating reports from the public, and on parts of the network — unclassified roads in particular — where no national reporting requirement bites.

The distinction is simple: unapproved data can help you decide where to measure. It cannot be the measurement.

Why an authority might want both. Accredited surveys are periodic by design; the coverage rules accept data up to two financial years old. Networks change faster than that. Cheap, frequent, unaccredited observation between accredited passes tells you where deterioration is running ahead of your programme — and then the accredited survey produces the number of record.

Where Make Sense sits

Make Sense is not PAS 2161 approved. We are not asking any authority to substitute data from us for data from an approved survey, and this page is not an argument that you should. It is an explanation of where the line sits, written by people who are on the unapproved side of it.

We are building a multi-sensor pod that mounts to vehicles already making their rounds, intended to record air quality, road surface condition and building heat loss, with provenance and confidence attached to every reading. The first system is in build. Nothing is deployed, and we have no accreditation, no PAS 2161 approval, and no published performance figures, because nothing has been validated yet.

What we are designing towards is the screening and targeting role described above — complementary to accredited survey cycles, not a replacement for them. We are not claiming it does that today, and we would rather you knew where we stand than find out later.

If the distinction between reportable and operational data is one you are working through, we are happy to talk about it without a pitch attached: get in touch.

Sources

Last reviewed 12 August 2026