Personal injury collision analysis in highway evidence

Highway objections are rarely about capacity. They are about safety, and personal injury collision analysis planning evidence is what decides whether a highway officer can support an access or has to object to it. Capacity arguments turn on modelling that both sides can rerun. Safety arguments turn on five years of recorded collisions and a set of measured distances, and those are facts.

Which is good news for applicants, because facts can be gathered before the design is fixed.

Why the safety section carries the objection

National policy sets two separate tests. Development should be refused on highways grounds where there would be an unacceptable impact on highway safety, or where the residual cumulative impacts on the road network would be severe.

The safety test has no qualifier. There is no severity threshold to reach and no cumulative element to argue about. If the highway authority concludes that the access is unsafe, the objection is close to fatal, and inspectors give it weight accordingly. The capacity test is far more forgiving, which is why a well-advised applicant spends the transport budget on the access rather than on junction modelling.

So the safety chapter of a Transport Assessment is the section to write first, not last.

Getting five years of collision data

Ask the local highway authority's road safety team for personal injury collision records covering the study area for the most recent five complete calendar years. Some authorities charge, some publish, and the national road safety statistics released by the Department for Transport cover the same underlying records.

Specify what you want rather than accepting whatever arrives:

• The extent of the study area, defined by junction and link rather than by radius.
• Date, time and day of week for each record.
• Severity, split into fatal, serious and slight.
• Number and type of casualties, including pedestrians and cyclists.
• Vehicle manoeuvres and, where recorded, contributory factors.
• A plotted map, or the coordinates to plot them yourself.

Five years is the convention because shorter periods produce noise and longer ones capture a highway that has since been altered. If a junction was rebuilt three years ago, say so and split the data, because a cluster before the works tells the officer nothing about the road today.

Beware of the empty return. Zero recorded collisions is a finding, but it is also what you get when the study area was defined too tightly.

Reading clusters against the proposed access

A plot on its own is decoration. The analysis is the sentence that connects each record to the thing your development would change.

Work through the records one at a time and ask whether the manoeuvre involved is one the development affects. A collision between two vehicles travelling straight through a junction 200 m away, at night, in the wet, is not evidence about your right turn. A pattern of shunts on the approach to the bend your access sits on is.

Group what you find by pattern rather than by count: turning movements at the same location, pedestrian collisions on the same crossing desire line, overrun on the same bend, collisions clustered in the same hours. Then say plainly whether the development would add movements of that type at that place.

Where a genuine pattern exists, propose something. A relocated access, an improved crossing, a length of footway, revised road markings under a section 278 agreement. An assessment that identifies a cluster and offers nothing invites the objection it was written to avoid.

Visibility splays and measured speeds

Visibility is the second half of the safety case, and the number that governs it is not the speed limit. It is the measured speed of traffic, normally the 85th percentile wet weather speed taken from a survey at the access location.

Manual for Streets sets out stopping sight distances for lower-speed streets, with commonly used values of about 25 m at 20 mph, 33 m at 25 mph and 43 m at 30 mph, measured along the nearside kerb from an x-distance of 2.4 m back from the carriageway edge. On higher-speed roads outside built-up areas the Design Manual for Roads and Bridges applies instead, and its distances are longer.

Three points cause most of the arguments. The x-distance can reduce to 2.0 m in low-speed situations, which is a concession the authority grants rather than one you assume. Splays must be within land you control or within the highway, because a splay across a neighbour's hedge is not deliverable. And the eye height and object height used to measure the splay have to be stated, since a splay that works at 1.05 m fails at 0.6 m over a rising verge.

Survey the speeds. A road with a 30 mph limit where traffic runs at 38 mph needs the longer distance, and an applicant who quotes the limit rather than the measurement will be corrected by the highway officer at week five.

When the data supports an access that looks substandard

This is where the analysis earns its fee. Plenty of workable accesses fall short of the geometric standard, particularly on older frontages, and the standard is guidance rather than a legal requirement.

The argument has three parts. First, the existing access has operated for a stated number of years at a given intensity of use with no recorded personal injury collisions attributable to it. Second, the development changes that intensity by a quantified amount, and the change is small. Third, whatever mitigation is available has been applied: vegetation cut back and secured by condition, a bellmouth widened, signage improved, a passing place formed.

Put like that, it is an argument about whether the residual position is unacceptable rather than whether it is ideal. Officers can accept that. What they cannot accept is a bare assertion that the access has always been fine.

Keep the same honesty in the other direction. Where the data does show a problem, say so early, because a highway authority that finds the cluster you omitted will read everything else more sceptically.

Writing a conclusion the officer can agree with

Highway officers write consultation responses under time pressure. Give them the paragraph they can adopt.

That means a conclusion in plain terms: the collision record shows no pattern related to the proposed access, or shows a pattern which the proposed works address. Visibility of x metres by y metres is achievable within land the applicant controls and accords with the relevant standard for the measured speed. The development would add a stated number of movements in the peak hour. The residual impact on highway safety is not unacceptable.

Numbers, not adjectives. Where the scheme is small enough that the transport evidence sits in a Transport Statement rather than a full assessment, the safety analysis belongs there in exactly the same form. Mode share commitments in a Travel Plan support the wider case but do nothing for the safety test.

Where a road safety audit sits

A road safety audit is a separate, formal process carried out by an independent audit team to the relevant national standard, usually required where highway works are proposed under a section 278 agreement. It is not part of a transport assessment and it is not something the planning submission itself produces.

PlanScope does not generate road safety audits. Where an audit is required, that work goes to an audit team with the right competence, and the transport documents should reference it rather than attempt it. You can check what PlanScope does cover in the full list of report types.

This is general guidance, not legal advice; check the position with your local planning authority or a qualified professional before relying on it. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.