When a council asks for junction capacity modelling

A junction capacity assessment planning request usually arrives as one line in a highway authority consultation response: model the site access and the nearest signalised junction in the opening year and a future year, with and without development. That single sentence commits a scheme to traffic surveys, growth factors, agreed assumptions and, often, a second round when the first set of assumptions is rejected.

Agree the scope before the modelling starts. Almost every dispute about a capacity model is a dispute about its inputs.

What the model is being asked to show

The question is narrow. Does the junction operate within capacity in the assessed years with the development traffic added, and if it does not, can the deterioration be mitigated. Nothing in a capacity model tells you whether a junction is safe, whether pedestrians can cross it, or whether the scheme is acceptable in planning terms.

Three tool families do the work. Priority junctions and roundabouts are handled by empirical capacity models. Signalised junctions use signal analysis packages that calculate degree of saturation from green times and flows. Networks where queues from one junction block another need microsimulation, which is more expensive and more sensitive to the assumptions behind it.

Choose the tool that matches the problem. Modelling a signalised junction in isolation when it sits 40 m from a roundabout that floods it every peak produces a result nobody should rely on.

Choosing the assessment years

Standard practice runs three sets of figures. A base year matching the survey data. An opening year, when the development is assumed to be complete and occupied. And a future assessment year, commonly five years after opening for local highway authority work, with longer horizons used on the strategic road network under the Design Manual for Roads and Bridges.

Phased schemes complicate this. A site building out over eight years needs either an interim year at a defined trigger or an agreed assumption that the full quantum is tested at opening, which is conservative and often simpler to defend.

Settle the years in writing with the highway authority. A model run for the wrong future year is a model run twice.

Base flows and how they are collected

Base flows come from classified turning counts at the junction, usually covering weekday morning and evening peak periods, sometimes with a Saturday peak where the surrounding uses call for it. Automatic traffic counts on the links give context and a check.

Survey timing decides whether the data is accepted. Counts taken in a school holiday, in the week before Christmas or during roadworks on a parallel route will be challenged, which is why neutral month survey timing is worth planning around rather than fitting to a programme.

Older data can be used with an agreed uplift where recent counts are unavailable, but the older it is the more argument it attracts. Where queue lengths are part of the case, record observed queues during the survey, because a model calibrated against observed queues carries far more weight than one calibrated against nothing.

Collision history sits alongside this. It does not feed the capacity model, but a request for capacity modelling is often accompanied by a safety question, and STATS19 collision data is how that is answered.

Committed development and background growth

Future year flows are base flows plus background growth plus committed development. Both additions are negotiated rather than calculated, and both are where models diverge.

Background growth normally comes from the Department for Transport's national trip end dataset, applied for the relevant area and time period. Committed development means extant planning permissions not yet built out, and sometimes allocations in the adopted local plan. The highway authority will usually supply a list, and it is worth asking for it in writing early, because discovering a 400 unit commitment after the model is built is expensive.

Watch for double counting. Background growth factors already contain an allowance for development in the area, so adding every committed scheme on top of an unadjusted growth factor overstates the future flows. Say in the report how you have handled that, and be ready to run a sensitivity test both ways.

Trip generation for the development itself normally comes from a recognised trip rate database, with site selection criteria stated and the selected surveys listed. Reviewers check those criteria first.

Reading the capacity output

For priority junctions and roundabouts, the output is a ratio of flow to capacity for each arm. A value of 0.85 is widely treated as practical capacity, since queueing and delay rise steeply beyond it. For signalised junctions, degree of saturation is the equivalent measure, with 90 per cent commonly taken as the practical limit.

Read the queue and delay figures alongside the ratio. An arm at 0.87 with a mean queue of three vehicles is a different proposition from an arm at 0.87 with a queue that blocks the junction upstream, and a table of ratios alone hides that distinction.

Also read the without development case. If a junction is already over capacity in the base year, the question shifts from whether the development causes a problem to how much it adds to an existing one, which is a much better argument for an applicant than most realise.

Mitigation that is deliverable

Mitigation that cannot be built is worse than no mitigation. Before offering a junction improvement, check that the land is within the public highway or under the applicant's control, that the geometry works with the existing kerb lines and statutory undertakers' apparatus, that it survives a road safety audit, and that it does not remove the footway or the crossing that the same authority asked for.

The delivery mechanism is a highway works agreement under section 278 of the Highways Act 1980, with a design, a bond and a commuted sum in most cases. Build the programme and the cost into the appraisal, because the section 278 process routinely takes longer than the planning application that triggered it.

Signal timing changes are the least costly mitigation and the most often refused, since the authority controls the network and may already be running the junction at the settings it wants. Ask before assuming.

Arguing proportionality on a small scheme

For a scheme generating a handful of movements in the peak hour, capacity modelling is sometimes disproportionate, and the Framework's transport policy supports that position. Development should only be prevented or 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. Verify the wording and policy reference in the current edition of the Framework before quoting it.

The practical answer is usually a scaled response. Offer a Transport Statement with trip generation, a distribution assumption and a link capacity check, and explain why a full junction model would not change the conclusion. Where the authority still insists, a single junction modelled in the opening year only is a reasonable compromise to propose.

Where the scheme is large enough for a full Transport Assessment, the modelling is only one chapter, and it should sit alongside the access design, the active travel provision and the servicing strategy rather than carrying the whole case.

This is general guidance, not legal advice. Check the scope with your local highway authority before commissioning surveys or modelling.

Enter your postcode on PlanScope to see which transport documents your scheme is likely to need. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.