Speed surveys and the 85th percentile in access design

Design a visibility splay to the posted speed limit and a competent highway officer will send it back. The input for an access design is the measured 85th percentile speed of traffic on the road, because that is what Manual for Streets asks for and what the authority's own checkers use. On a 30mph road where traffic actually runs at 36mph, the required y distance grows, and the splay may no longer fit within land you control.

So the 85th percentile speed survey is not a formality at the end of the transport work. It is an input that should be taken before the access position is fixed.

Why the posted limit is not the design input

Speed limits describe what drivers are supposed to do. Stopping sight distance describes what they need if they are to stop, and that depends on the speed they actually travel.

Manual for Streets and its successor set out the derivation: a reaction time component, a braking component that varies with the square of speed, and an adjustment for gradient. Authorities tabulate the resulting y distances against speed. Feed in a higher speed and the distance rises sharply, because of the squared term.

Two consequences follow for site appraisal. A road where measured speeds exceed the limit imposes a heavier requirement than the signs suggest. And a road where measured speeds fall well below the limit, which is common on narrow residential streets with parking, allows a smaller splay than the limit implies, which can rescue an access that looked unworkable.

What the 85th percentile represents

It is the speed at or below which 85 out of every 100 vehicles travel. It is a long standing convention in highway engineering, used because it captures the behaviour of the great majority of drivers while excluding the small tail travelling unreasonably fast.

It is a per direction figure. Traffic approaching from the left and from the right will often differ, particularly where there is a bend, a gradient or a junction upstream on one side. Report both, and design each arm of the splay against the speed of the traffic it is protecting against.

Counters, placement and survey duration

Automatic traffic counters, whether pneumatic tube or radar, record speed and classification continuously. A full seven day survey is the norm, because a weekday only count misses weekend behaviour and a two day count is vulnerable to one wet afternoon.

Placement decides whether the data is usable at all.

• Site the counter on a straight section where vehicles are travelling at their normal running speed, away from the influence of junctions, bends, crossings and traffic calming.
• Keep clear of parked vehicles, which shield the equipment and distort counts.
• Avoid locations where a queue forms in the peak, since queued traffic drags the distribution down and makes the result unusable.
• Check for roadworks, temporary signals or a diversion route in operation during the survey period, and record the check.
• Survey in a neutral month, avoiding school holidays and the periods either side of Christmas.

Record the calibration status of the equipment and the exact location on a plan with a photograph. Authorities ask, and a survey you cannot locate precisely is a survey you cannot defend.

Wet weather speeds and when they are asked for

The guidance derives stopping sight distance using wet weather conditions, on the basis that braking performance is lower on a wet road. Drivers also travel more slowly in the wet, so the wet weather 85th percentile is usually below the dry figure.

That produces a choice. Designing to the dry speed is conservative, because it demands a longer sight line than the wet condition requires. Some authorities accept that and ask for nothing more. Others, particularly where the applicant is arguing for a reduced splay on a constrained site, ask specifically for wet weather speeds, because that is the figure the derivation assumes.

Agree the point with the highway officer before commissioning. A wet weather survey means waiting for rain during the survey window and filtering the data by recorded rainfall, which adds cost and programme, and there is no point incurring either if the dry figure already works.

Where to measure relative to the access

Measure where the traffic passes the proposed access, or at the nearest representative point with the same road character. Do not measure two hundred metres away across a junction and assume the speeds transfer.

Presenting the data so it is not argued with

Give the authority the raw output as well as the summary. A table of 85th percentile speeds by direction and by day, the survey dates, the location plan, the equipment type and the weather record is the minimum. Add the full speed distribution as an appendix.

State the derivation you have applied and the source you have taken it from, then show the required y distance and the splay drawn on a plan at a stated scale, with the x distance marked and the land ownership shown. Where the splay crosses third party land, say so plainly and explain what control exists, because an officer who discovers it later will treat the whole submission with more suspicion.

Illustrative example: a two dwelling access onto a 30mph road with a bend 60 metres to the north. Surveys upstream of the bend and south of the access returned 85th percentile speeds of 34mph northbound and 29mph southbound. The southbound splay fitted within the frontage. The northbound splay did not, and the access was moved 15 metres south, where the geometry worked. Had the survey been commissioned after the layout was drawn, the plot arrangement would have had to change.

When the highway authority disputes the result

Three disagreements recur. The authority holds older speed data showing a different figure. It considers the survey location unrepresentative. Or it argues that the posted limit should be used regardless.

Deal with the first by comparing methodologies and dates rather than arguing about the numbers, and by offering to repeat the survey at a location the authority nominates. Deal with the second by agreeing the location in advance, which removes the argument entirely. The third is the easiest, because the guidance is clear that design should reflect actual speeds, and a note quoting the relevant passage usually settles it.

Where the dispute persists and the scheme is otherwise sound, a joint survey with the authority present at set up is disproportionate on a small scheme and worth it on a large one. The wider evidence base, and when a full assessment rather than a statement is needed, is covered in our note on transport assessment thresholds and trip generation, and the geometry of the drive beyond the splay is dealt with in private drive geometry, gradients, widths and turning.

The speed data and the splay drawing belong in the Transport Statement for a small scheme, or in the Transport Assessment where one is required.

This is general guidance, not legal advice; check the position with your local highway 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.