A queue extending beyond the stop line is not simply a visible sign of congestion. It can block a pedestrian crossing, prevent a bus leaving a stop, restrict access to a side road or create a safety risk for cyclists filtering through slow-moving traffic. Junction queue analysis turns these observations into measurable evidence, allowing highways teams to identify where capacity is being lost and what action is most likely to improve conditions.
For authorities and signal engineers, the value lies in understanding the full pattern of demand rather than relying on isolated peak-hour counts. The critical questions are how far queues extend, how long they persist, whether they clear each cycle and where they create conflict or network-wide disruption.
What junction queue analysis measures
Queue analysis assesses the build-up and discharge of traffic at a junction approach. A useful study records more than the maximum queue length. It establishes the time a queue begins to form, the number of cycles required to clear it, the rate at which vehicles depart during green time and whether residual queues remain after a signal stage ends.
These measures reveal different operational problems. A short queue that clears every cycle may be acceptable, even if it appears busy. A modest queue that remains for several cycles is more concerning because it indicates that available green time, saturation flow or downstream capacity is insufficient for demand. Where queues regularly reach upstream junctions, accesses or pedestrian facilities, the problem becomes a network and safety issue rather than a single-junction issue.
Queue length can be expressed in vehicles or metres. Both have a purpose. Vehicle counts are helpful for comparing demand by lane or movement, while metres show the real road-space impact and make it easier to assess spillback. For mixed traffic conditions, it is often sensible to consider both, particularly where buses, lorries and cycles affect how quickly a queue forms and discharges.
Why counts alone do not explain delay
Traffic counts remain essential, but they do not show how a junction behaves under pressure. Two approaches may carry the same number of vehicles in an hour while producing very different user experiences. One may operate with steady movement and brief queues; the other may suffer recurring oversaturation, with drivers waiting through more than one signal cycle.
This is why junction queue analysis should be considered alongside turning counts, signal timings, stage sequence, pedestrian demand, bus movements and downstream constraints. The combined evidence helps engineers distinguish between a timing issue and a capacity issue.
For example, a right-turn lane may appear underused in an hourly count but still generate an intermittent queue that blocks the ahead lane whenever several vehicles arrive together. Equally, extending green time for a congested approach may offer little benefit if vehicles cannot exit because the next junction is blocked. Without observing queue formation and discharge, changes can simply transfer delay elsewhere on the network.
Collecting data without disruptive roadworks
Legacy inductive loops can provide useful detection, but installation and replacement can require carriageway works, traffic management and ongoing maintenance exposure. Above-ground detection gives traffic teams another route to detailed junction intelligence without cutting into the road surface.
AI-powered video detection can identify vehicles, cyclists and pedestrians within defined detection zones. It can monitor approach lanes, stop lines, queue extents and turning movements, while providing time-based data for analysis. Radar detection is particularly valuable where reliable vehicle presence, speed and range measurement are required in difficult lighting, rain or low-visibility conditions. In many locations, combining complementary detection methods provides the clearest operational picture.
The important specification question is not simply which sensor is available. It is whether the detection layout will capture the movements and queue boundaries that matter to the scheme. A detector positioned only at the stop line may support actuation, for example, but it will not quantify the risk of a queue extending 80 metres upstream. Queue monitoring needs appropriately defined zones or tracking coverage along the approach.
Temporary deployment is also useful before committing to a permanent intervention. A targeted survey can validate concerns around school peaks, event traffic, weekend retail demand or seasonal variation. This reduces the risk of optimising a junction for a condition that occurs only occasionally.
A practical method for analysing junction queues
The most useful studies begin with a clear operational question. That may be whether a pedestrian stage is causing excessive vehicle delay, whether a new development is affecting a nearby signalised junction, or whether a bus priority measure is creating unacceptable queues in an adjacent lane.
Define the study period and thresholds
Select periods that represent the problem, not merely the busiest hour. Morning and evening peaks are often relevant, but the worst queue may occur at school closing time, after a sporting event or during a constrained weekend period. Where feasible, collect several days of data to separate recurring conditions from anomalies caused by incidents, weather or temporary works.
Set meaningful thresholds before reviewing the results. These might include queues reaching a crossing, queues that fail to clear within one cycle, or an approach occupancy level that signals likely spillback. A threshold should relate to an operational consequence, not an arbitrary number.
Map every movement and conflict point
Analyse each approach by lane and turning movement. A shared ahead-and-left lane behaves differently from a dedicated lane, and a short flare may store only a limited number of vehicles before affecting the main carriageway. Include bus stops, cycle facilities, pedestrian crossings, accesses and upstream stop lines in the mapping.
This step often exposes the actual constraint. A junction may have enough nominal green time, yet its effective capacity is reduced by late lane changes, blocked exits, loading activity or uncertainty around cyclist movements. Engineers need to see where traffic is physically held, not only what the controller plan intends to deliver.
Compare demand with discharge
For each signal cycle, compare the traffic arriving on an approach with the traffic discharged during green. If more vehicles arrive than leave over repeated cycles, a residual queue will grow. If the first few seconds of green are consistently lost because the exit is blocked, increasing green time may not solve the issue.
Discharge patterns can also reveal whether signal settings are appropriate. A lane that empties early on every cycle may have spare green time, while a competing movement remains saturated. However, reallocation must be tested carefully. Pedestrian clearance requirements, intergreens, public transport reliability and side-road access all shape what is feasible.
Identify the cause before selecting the remedy
Not every queue needs a signal timing change. The cause may be a downstream bottleneck, poor lane discipline, an unsuitable detector position, a development access, loading restrictions or an incident pattern. It may also be a genuine capacity constraint that cannot be resolved through optimisation alone.
This is where data-led review prevents wasted investment. A short additional green extension may be effective where demand is variable and an approach is genuinely under-served. In contrast, adaptive operation using reliable real-time detection may be better where demand changes significantly through the day. Where downstream blocking is the cause, linked junction operation or targeted enforcement may offer more benefit than extra green time.
Turning analysis into safer, more efficient operation
The strongest outcomes come from using queue data as part of an ongoing operational process, rather than a one-off report. Before-and-after measurement should confirm whether a signal change has reduced residual queues, improved journey reliability or prevented blocking of a crossing. It should also identify any unintended effect on other approaches.
For a busy urban junction, even small improvements can matter. Reducing repeated stop-start conditions can lower delay, improve bus punctuality and reduce unnecessary emissions. Preventing queues from obstructing crossings and cycle routes can make an immediate difference to perceived and actual safety.
C & T Technology supports this approach with above-ground radar and AI video detection solutions that provide the timely, location-specific information needed for practical traffic management decisions. The objective is not to collect data for its own sake, but to give engineers confidence in what is happening at the kerbside and within the junction.
A well-designed junction queue analysis should leave the project team with a clear next action: retain the existing operation because it performs acceptably, adjust control parameters, improve detection coverage or investigate a constraint beyond the stop line. That clarity is what turns observation into safer roads, reduced congestion and better use of the existing network.