A standing queue is not just a delay measurement. On a busy approach, it can block a pedestrian crossing, prevent buses leaving a stop, trap vehicles in a junction or push rat-running into residential streets. This guide to queue monitoring strategies sets out how highways teams can turn reliable detection into proportionate, operational action rather than simply producing another congestion report.

Start with the operational question

Queue monitoring should begin with the decision the data needs to support. A signal engineer may need to know whether a queue has reached the previous stop line. A network operations team may need early warning that traffic is approaching a motorway slip road or town-centre junction. A road safety team may be investigating whether recurring queues create rear-end collision risk.

These are related, but they are not the same use case. A single measure of average queue length will rarely serve all of them. Define the monitored approach, the direction of travel, the time periods that matter and the consequence of a queue extending beyond a specified point. This creates a monitoring design that can be tested against a real operational need.

For example, a queue of 12 vehicles may be manageable on a long arterial approach but unacceptable where it obstructs a side-road exit after only six vehicles. The meaningful threshold is therefore location-specific. It should be based on available storage, downstream constraints, vulnerable road user movements and the authority’s chosen network priorities.

Choose detection that sees the whole problem

The quality of a queue strategy is constrained by its detection. Traditional inductive loops can provide useful occupancy and demand information, but installation and repair require carriageway works. They also provide measurements at fixed points, which can make it difficult to establish the rear of a queue or distinguish slow-moving traffic from a stationary queue.

Above-ground radar and AI-powered video detection offer a different approach. Correctly positioned, they can observe an approach zone, track vehicles and report presence, speed, classification and queue extent without cutting the road surface. This is particularly valuable where road closures, reinstatement risk or buried services make intrusive detection undesirable.

The best technology choice still depends on the site. Radar performs well where reliable all-weather vehicle detection and range are priorities. Video analytics can add rich scene understanding, including vehicle trajectories and the movements of cyclists or pedestrians, but needs careful camera siting, commissioning and privacy-aware configuration. Wireless sensors may suit locations where rapid deployment or flexible layouts are more important than a wide field of view.

Avoid specifying a detector only by its headline range or detection accuracy. Ask what it can report at the point where the decision must be made. Can it identify a stationary vehicle at the critical queue threshold? Can it reliably separate turning traffic from through traffic? Can the output be passed into the signal controller, urban traffic control platform or traffic data platform in a usable format?

Detection zones need engineering, not guesswork

A queue monitoring zone should cover the area where queue growth becomes operationally significant. On a signalised junction, that may mean a stop-line zone, an intermediate zone for demand assessment and a remote zone that identifies potential blocking back. At a merge, the priority may be the point at which a queue approaches a live lane or compromises visibility.

Zones must reflect lane geometry and real driver behaviour. If left-turn traffic regularly bypasses the main queue, combining both movements in one zone can overstate the delay faced by through traffic. Equally, a detector that sees an adjacent entry lane may generate false queue events unless its field of view and classification logic are set correctly.

Site surveys should account for poles, signs, street furniture, gradients, tree growth and seasonal changes in lighting. These details are not peripheral. They determine whether a technically capable detector delivers dependable data over its operating life.

Measure queue behaviour, not only queue length

Queue length is a useful indicator, but it is only one part of the picture. A short queue that clears every cycle may cause little network harm. A similar queue that persists for 20 minutes, repeatedly blocks an access or grows during every cycle needs intervention.

A practical dataset should connect queue extent with time, discharge and recurrence. Four measures are particularly useful:

  • maximum queue extent, measured in metres, vehicles or defined detection zones;
  • queue duration, including how long the queue remains above a critical threshold;
  • clearance performance, showing whether the approach clears during the available green time; and
  • recurrence, identifying the days, periods and conditions in which the issue repeats.

These measures reveal whether the constraint is likely to be insufficient green time, downstream restriction, an incident, poor lane discipline or demand that exceeds practical junction capacity. They also prevent teams from responding to isolated peaks as though they were a daily network failure.

Vehicle classification adds further value. A queue dominated by HGVs may have different discharge characteristics from a queue of private cars. Bus movements may justify signal priority considerations, while the presence of cyclists at an advanced stop line must be handled safely rather than treated as residual demand. The aim is not to collect every available data point. It is to collect the evidence needed to make a defensible traffic management decision.

Set thresholds that trigger a response

A monitoring system becomes operationally useful when it has agreed thresholds and actions. These thresholds should distinguish between observation, intervention and escalation. For instance, a first threshold may flag an unusual build-up for review, while a second may initiate a signal plan change or alert the control room. A third may indicate a safety-critical condition, such as blocking back across a junction.

The response must suit the cause. Extending green time can help where demand is temporary and downstream capacity exists. It can worsen conditions on other approaches if used too freely. Holding traffic upstream may protect a constrained town centre, but only if the holding location has sufficient storage and does not create a new safety risk. For a recurring issue, revised stage sequencing, linked-junction coordination or a longer-term scheme may be more appropriate than repeated real-time intervention.

Use historical data to establish normal operating ranges before setting alerts. A threshold that generates alarms every weekday morning will soon be ignored. Conversely, a threshold set too high will detect the issue only after traffic has already blocked key movements. Review thresholds after signal timing changes, highway works, new development access or changes to traffic patterns.

Build data quality into the strategy

Queue data should be checked against what is happening on site. During commissioning, compare detector outputs with observed queue positions across different traffic conditions. Test peak flows, low flows, darkness, rain and unusual vehicle mixes where possible. A system that appears accurate during a quiet site visit may behave differently in the evening peak.

Ongoing validation matters too. Sudden changes in queue patterns may be genuine, but they can also indicate a shifted camera view, an obscured lens, a changed road layout or a communications issue. Trend dashboards should therefore show detector health alongside traffic indicators. Missing data should not be interpreted as free-flowing traffic.

For AI video systems, establish clear governance for image handling, retention and access. Where operational objectives can be met using anonymised events and metadata rather than stored footage, that approach reduces unnecessary data exposure while retaining the traffic insight required.

Integrate monitoring with network control

Queue monitoring delivers the greatest value when information moves quickly to the people and systems that can act on it. At isolated junctions, this may mean direct inputs to adaptive signal control or demand-responsive stages. Across a corridor, it may involve coordinating signal plans to avoid transferring queues from one junction to the next.

For wider networks, queue events can inform operator decisions, variable message signs, planned diversion routes and incident response. Integration should not create opaque automation. Engineers need to understand the conditions under which a strategy changes, the maximum duration of that change and the safeguards that return the network to a stable plan.

C & T Technology’s above-ground detection approach is well suited to these applications because it supports fast deployment without the disruption of road-embedded loops. The practical advantage is not simply a shorter installation programme. It is the ability to obtain targeted, maintainable detection where queue visibility is most needed.

Review outcomes and refine the intervention

A queue monitoring strategy should be assessed against outcomes that matter to road users and network operators: fewer instances of blocking back, improved junction clearance, more reliable bus movement, reduced exposure to queue-related risk and less time spent responding to unverified reports. Before-and-after analysis should compare like-for-like days and account for school terms, weather, roadworks and known incidents.

Not every queue should be eliminated. Some queues are the managed consequence of protecting pedestrians, prioritising public transport, controlling access to a constrained area or preventing a larger network failure elsewhere. The objective is to make that trade-off visible, evidence-led and safe. Well-designed detection gives highways teams the confidence to act earlier, test changes properly and keep traffic moving where it matters most.

C & T
Privacy Overview

This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.