Selecting TRICS sites and the filters councils challenge

TRICS trip rate selection is audited, not accepted. A highway officer reading a transport chapter will go straight to the appendix, look at the selection criteria page, and ask why each filter was set where it was. Every filter narrows the sample, every narrowing moves the rate, and a filter with no stated reason looks like a filter chosen for its answer.

So write the selection as an argument. Land use category, sub-category, geography, location type, size range, parking, date range, days surveyed, and then the sites included and excluded with a reason against each exclusion. Do that and the trip rate stops being the disputed issue.

What the database contains and what it does not

TRICS holds multi-modal survey data from sites across the United Kingdom and Ireland, recording vehicle, person, pedestrian and cycle movements over surveyed days, with information about each site's size, location, parking provision and local population. It is the standard source for trip generation in England and it is what a highway authority expects to see.

It does not contain your site. Every rate derived from it is an inference from other places, which is why the comparability argument matters more than the arithmetic. It also does not contain every land use. Specialist uses, unusual operating models and new formats are often absent or represented by two or three surveys from the wrong decade.

And it does not tell you what will happen. It tells you what happened at surveyed sites on surveyed days. Presenting it with that honesty usually gets a better reception than presenting it as a prediction.

Choosing land use category and sub category

Start at the top of the hierarchy and record every choice. Residential splits into sub-categories including privately owned houses and privately owned flats, and the difference between them is material: flats typically generate fewer vehicle movements per unit than houses, so a mixed scheme assessed entirely as houses overstates traffic and one assessed entirely as flats understates it.

Where a scheme has a mix, derive separate rates for each component and combine them by unit count rather than picking a single sub-category. Set that calculation out in a table in the main text, not buried in an appendix.

The sub-category argument is where an officer will start if they disagree with the whole assessment, so deal with it first and deal with it explicitly. Say what the scheme is, say which sub-category it matches and why, and say what you did about the parts that do not match.

Filtering by location type, parking and population

Three filters do most of the work, and each has a trap.

Location type describes the setting of the surveyed site, with options running from town centre through edge of centre and suburban area to free standing locations. Pick the ones that match your site's accessibility rather than its administrative label. A suburban site with a station at the end of the road behaves differently from one four miles from a bus route, and the population filters are the way to capture that.

Population within one mile and within five miles filters the sample by local density. Set a range around your site's figures rather than a single value, and state both the site figures and the range used. Officers check this, because a narrow population band is the easiest way to cut a sample down to the three sites that give the answer you wanted.

Parking provision is the filter most often left at the default. If your scheme provides one space per dwelling and the retained surveys average two, the vehicle rate carries that difference into your assessment. Either filter to a comparable range or explain why the difference does not matter.

Survey date ranges and pandemic era data

Date range is the filter most argued about. Too wide and the sample includes travel behaviour from a different era. Too narrow and the sample collapses to a handful of sites.

Surveys carried out during 2020 and 2021 need explicit treatment, because lockdowns and changed working patterns make them unrepresentative of normal conditions and unrepresentative in different directions for different land uses. TRICS categorises surveys affected by those conditions, so state whether you have included or excluded them and why. Excluding them silently is the fastest way to attract a challenge.

Post-2021 residential data raises a real and unresolved question about home working and its effect on the weekday morning peak. Say what you think about it, and say what the effect would be if you are wrong. An assessment that acknowledges an honest uncertainty reads better than one that pretends the question does not exist.

Sample size and the variability it hides

An average trip rate derived from four sites tells you very little about the spread. The TRICS Good Practice Guide covers sample size and cautions against selections made to produce a preferred outcome, so check the current edition and follow it.

Report more than the mean. Give the number of sites, the range of individual site rates, and where the sample is small say so. Where the scheme is sensitive to the rate, run the junction assessment at the upper end of the observed range as a sensitivity test. That costs an hour and removes an entire line of objection.

Presenting the selection so it can be audited

Include the full output in an appendix, with the selection criteria page intact, and add a short table in the main text summarising each filter and the reason for it. List every site excluded by hand, with the reason against each one. Excluded because the survey pre-dates 2010 is a reason. Excluded because the rate was high is not, and an officer who spots that will discount the whole chapter.

Then cross-check the resulting rate against something independent, such as observed flows at a comparable local development or census travel to work data for the ward. Where the two agree, say so. Where they do not, explain the difference before somebody else does.

The rate feeds everything downstream, from the junction modelling in a Transport Assessment to the mode share targets in a Travel Plan, so an unstable rate destabilises the whole submission. Whether a full assessment is needed at all is a separate threshold question covered in our note on trip thresholds.

Where a bespoke survey beats the database

Commission your own survey in four situations: the land use is not represented in TRICS, the operating model is materially different from the surveyed sites, a directly comparable development exists within a few miles, or the scheme is large enough that the rate drives a costly mitigation package.

A local survey at a comparable site is persuasive in a way a national average is not, because the officer knows the road. Agree the survey site, the dates and the method with the highway authority before you commission it, and avoid school holidays, August and December.

Parking demand is a separate exercise again and is usually surveyed locally rather than derived from the database, which is what a Parking Survey covers. Where the scheme touches the strategic network, expect a further layer of scrutiny, set out in our note on access onto the strategic road network.

Illustrative example. A 90 unit scheme of 60 houses and 30 flats is first assessed using privately owned houses alone, across England excluding Greater London, with a 2015 to 2024 date range and no parking filter. The highway authority challenges the sub-category and the parking range. The revised assessment splits the components, filters parking to between 0.8 and 1.4 spaces per unit, states the treatment of pandemic era surveys and reports the range as well as the mean. The objection is withdrawn. The example is illustrative.

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.