Trip rates and the TRICS database

A trip rate is the number of trips a development generates per unit of development in a given time period: vehicle trips per dwelling in the morning peak hour, or person trips per 100 square metres of floorspace across a 12-hour day. TRICS is the database of surveyed trip rates that UK transport assessments draw those figures from, and the number you end up quoting depends almost entirely on which surveyed sites you chose to average.

That is why highway authorities test the site selection rather than the arithmetic. The arithmetic is never wrong.

What a trip rate expresses

Three components: a count, a unit and a period.

The count is trips, split into arrivals and departures, with the two-way total usually quoted. A trip is a one-way movement, so a resident driving to work and home again generates two vehicle trips.

The unit is whatever scales the development: dwellings for residential, 100 square metres of gross floor area for most commercial uses, bedrooms for hotels, pupils for schools, treatment rooms for health uses. Getting the unit right matters more than it sounds, because a rate expressed per 100 square metres applied to a gross external area rather than gross floor area introduces an error nobody notices.

The period is either a specific hour, usually the network peak, or a longer window such as 07:00 to 19:00. A daily rate and a peak hour rate answer different questions, and junction capacity work needs the peak.

Multiply rate by quantum and you have the predicted trips. Everything contentious happens before that multiplication.

How the database is built

TRICS holds surveys commissioned and submitted by its consortium members and contributors, carried out to a consistent method: manual classified counts at every access to a development, usually across a 12-hour day on a neutral weekday, with the site's characteristics recorded alongside.

The site record is what makes selection possible. Each survey carries the land use category and subcategory, gross floor area or unit count, parking provision, location type (town centre, edge of town centre, suburban area, neighbourhood centre, edge of town, free standing), region, survey date, and contextual data on population and car ownership in the surrounding area.

Categories run across residential, employment, retail, health, education, leisure, hotel and food and drink, and mixed use. Within each there are subcategories, and the choice between two plausible subcategories can move the rate by a third.

Multi-modal surveys record more than vehicles: pedestrians, cyclists, public transport users and vehicle occupancy. Those surveys are fewer, which constrains what you can do with mode share.

Selecting comparable sites

The selection is a series of filters, and each one should be justified rather than accepted as a default.

1. Land use category and subcategory, matched to the proposal's actual operation rather than its use class.
2. Geographic region. Excluding Greater London is conventional for schemes outside it, because London's car ownership and public transport provision distort the average. Include it and expect to justify why.
3. Location type, which is the filter that does the most work. A suburban site and a town centre site of identical floorspace generate very different vehicle trips.
4. Size range. A band around the proposal's quantum, commonly something like half to double, rather than the whole database.
5. Survey date. Surveys more than about ten years old are usually excluded, and post-2020 surveys carry their own interpretation problems.
6. Parking provision, where the proposal's parking ratio differs materially from the surveyed sites.

Aim for a sample of at least five to ten sites after filtering. Fewer than five and the average is driven by individual sites. Thirty sites with a wide size range is not a better sample, it is a less comparable one.

State the resulting range as well as the mean. A highway authority that can see the minimum and maximum rates alongside the average is being given the information to judge, which tends to shorten the correspondence.

Time periods and peak hours

Two peaks matter and they are not always the same hour.

The network peak is when the surrounding highway network is busiest, usually within 08:00 to 09:00 and 17:00 to 18:00 on a weekday. The site peak is when the development itself generates most trips, which for a school is earlier, for a gym later, and for a supermarket on a Saturday afternoon.

Junction capacity assessment normally uses the network peak, because that is when spare capacity is scarcest. Access design and internal circulation may need the site peak. A food store or a leisure use will often need a Saturday assessment as well, and the highway authority will ask for it if you have not offered it.

Where the site peak falls outside the network peak, say so and show both. Presenting only the network peak for a use that empties its car park at 15:30 next to a primary school is the kind of omission that produces an objection on amenity rather than capacity.

Multi-modal rates and mode share

Vehicle trip rates answer the capacity question. Mode share answers the sustainability question, and the two are often assembled from different sources in the same document, which is where inconsistency creeps in.

Three routes to mode share, in descending order of reliability for a specific site:

• multi-modal TRICS surveys from comparable sites and locations, which give observed person trips by mode
• census travel-to-work data for the relevant output area, adjusted with care because the 2021 Census was taken under pandemic conditions and understates commuting travel
• a local authority's own monitoring or a nearby site's travel plan surveys

Whichever you use, reconcile it with the vehicle trip rate. If your vehicle rate comes from suburban sites with generous parking and your mode share comes from an urban census area, the two halves of the assessment describe different developments. Reviewers notice.

Where a Travel Plan commits to a mode shift target, the baseline has to be the same figure used in the trip generation work, and the monitoring method has to be capable of measuring it.

Common challenges from a highway authority

The objections repeat, and most are avoidable:

The sample is not comparable. Usually a location type or parking provision mismatch, and usually correct.

The size band is too wide. A residential rate built from sites of 15 to 400 dwellings tells you little about a scheme of 60.

Outliers have been handled inconsistently. Excluding a high-trip site as an outlier while keeping a low-trip one invites the authority to redo the exercise its own way.

Existing use trips have been netted off without evidence. Discounting the trips of a lawful existing use is legitimate, and it needs the lawful use established and the existing trips either surveyed or justified from comparable data. An assumption that a vacant building was fully occupied at typical rates will be resisted.

Pass-by and diverted trips have been claimed too generously on retail schemes, or internalisation assumed on a mixed-use scheme without a basis.

The database version is not stated, so the output cannot be reproduced.

Presenting the selection audit trail

Attach the output, not a summary of it. The appendices to a Transport Assessment should carry the selection criteria page showing every filter applied, the list of surveyed sites used with their characteristics, and the rate calculation sheets.

Then add a short narrative in the main text. State the database version and the date of the interrogation. Explain each filter in one sentence. Name any site excluded and why. Give the mean, the range and the sample size. Where the proposal differs from the surveyed sites in a way the filters cannot capture, say so and explain the direction of the likely error.

For a smaller scheme the same discipline applies in a Transport Statement, at proportionate length. Councils usually expect a suitably qualified author for transport work, and professional review is strongly advised before the figures support a capacity assessment. A parking argument runs on separate evidence, which is what a Parking Survey provides.

Reproducibility is the whole point. An officer who can rerun your selection and get your number will accept it. One who cannot will commission their own.

This is general guidance rather than 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.