Observed queue lengths, cycle by cycle

Measured at the point in the cycle that matters

A queue length figure means nothing without the moment it was taken. We measure at the start of green — the standing queue that has accumulated during red — and record the maximum extent reached during each cycle. Both matter: the standing queue drives green time allocation, the maximum extent tells you whether the queue reached the upstream junction.

Measurements are reported in whichever unit your analysis needs: number of vehicles, distance in metres or feet, or passenger car units where a heavy-vehicle mix would otherwise distort the figure.

What gets recorded

• Standing queue at the start of green, by lane and by cycle
• Maximum queue extent within each cycle
• Average and 95th percentile queue across the survey period
• Blocking-back events where a queue reaches the upstream junction or a driveway
• Residual queues that survive a full green phase — the clearest sign of oversaturation
• Starting and stopping delay, and total approach delay where required
• Cycle-by-cycle time stamps so results can be aligned with signal logs

Calibrating a model with observed queues

Modelled queues and real queues diverge, and the gap is where credibility is won or lost with a reviewer. Observed queue data lets you calibrate saturation flow, adjust the start-up lost time, and demonstrate that the model reproduces conditions on the ground before it is used to predict conditions that do not exist yet. Increasingly, agencies ask for exactly this evidence.

Frequently asked