Saturation flow rate sits at the core of every signal timing calculation. It defines how many vehicles can cross a stop line per hour if the green phase ran continuously without interruption. Get it wrong, and every timing plan built on top of it will be slightly off: too little green time in one direction, unnecessary delay in another, compounding across a network. In smart signal environments, where timing adjusts continuously in response to live conditions, a poorly calibrated saturation flow rate creates systematic error that adaptive algorithms can't fully correct.
What saturation flow rate actually measures
The saturation flow rate describes the throughput capacity of a lane or lane group under ideal conditions, expressed in passenger car equivalents (PCE) per hour of green. In Australia, practitioners typically work from the Austroads guidelines, which set a base saturation flow rate of 1,850 PCE per hour of green for a standard through lane. That number assumes a flat approach, no turning movements, standard lane width, and a composition of entirely passenger vehicles.
In practice, none of those conditions hold everywhere. Heavy vehicles reduce effective capacity because they take longer to accelerate and occupy more stop-line space. Right-turning traffic conflicts with oncoming flows, which reduces usable green time. Narrow lanes force drivers to leave larger lateral gaps. Bus stops near the stop line interrupt the discharge stream. Each of these factors is captured by an adjustment to the base rate, and the adjustments multiply quickly.
A right-turn lane at a busy urban intersection might carry a saturation flow rate of 1,200 PCE per hour of green after accounting for opposing traffic, pedestrian crossings, and the occasional oversized vehicle. Using the base figure of 1,850 in that context would produce a timing plan that allocates too little green time to a movement that needs significantly more.
How saturation flow rate is measured in the field
The standard field method involves recording queue discharge at a signalised intersection. An observer counts the vehicles that cross the stop line during the green phase, starting after the first four vehicles (the initial queue discharge period, which includes drivers reacting to the signal change). The remaining vehicles crossing under steady-state flow give the saturation flow rate for that movement.
At least 10 to 15 signal cycles are typically required to get a reliable average. Each cycle provides one data point, and saturation flow varies cycle-to-cycle with vehicle mix, approach speed, and lane discipline. Raw counts are then converted to PCE values by applying heavy vehicle equivalency factors, usually in the range of 2.0 to 2.5 PCE for an articulated vehicle on a flat approach.
Some agencies use video-based methods. Fixed cameras capture a lane cross-section and software extracts headway data from the recorded footage. This approach is slower to set up but removes observer fatigue from the equation, and the footage can be reviewed if a count looks anomalous. Bluetooth and radar sensors at the stop line can also measure inter-vehicle headways directly, which is a more precise input to saturation flow calculations than raw vehicle counts alone.
Why it matters for adaptive signal control
Fixed-time signal plans use a saturation flow rate that's calibrated once and then held constant until the next plan review, which might be every 3 to 5 years. Adaptive systems need a more current picture. Adaptive signal control platforms such as SCATS and SCOOT estimate saturation flow dynamically by monitoring detector occupancy and flow data from each signal cycle. The system infers how much of the green phase is being used at capacity and adjusts its internal saturation flow estimate accordingly.
That continuous recalibration is one of the core reasons adaptive control outperforms fixed-time plans in variable traffic conditions. If a new bus route increases heavy vehicle proportions on an approach, or a lane is temporarily narrowed by road works, the adaptive system picks up the change through the detector data and adjusts its timing calculations within a few cycles. A fixed-time plan would require a manual update.
The detector infrastructure is the critical link. Poor detector performance, missed counts, or misclassified vehicles feed corrupt data into the saturation flow estimate and degrade timing output. This is worth understanding alongside queue detection at signalised intersections, since both capabilities rely on the same sensing layer and their accuracy is interdependent.
Common mistakes in saturation flow estimation
Three mistakes appear repeatedly in saturation flow studies conducted for signal design projects.
- Using a single survey period. Saturation flow varies by time of day, day of week, and season. A survey conducted only during the AM peak will miss the different vehicle mix and turning proportions in the PM peak or off-peak periods.
- Ignoring geometry changes. Road works, kerb extensions, and parking changes all alter effective lane width and the sight lines that influence driver behaviour at the stop line. An out-of-date saturation flow figure is worse than a carefully measured one.
- Applying a national default without checking local conditions. The Austroads base rate of 1,850 PCE per hour is a reasonable starting point, not a finished value. Urban arterials with high pedestrian conflict, frequent bus movements, or steep grades all warrant field-measured adjustments.
Saturation flow in network-wide planning
At the network level, saturation flow rates feed into degree-of-saturation calculations for each movement at each intersection. Degree of saturation is the ratio of demand flow rate to capacity, and it determines how close a movement is to failing (exceeding 1.0 means the movement can't discharge all waiting vehicles within the allocated green). Transport authorities use these values to identify intersections approaching capacity before congestion becomes visible in travel time data.
The Austroads Guide to Traffic Management Part 6 covers the methodology in detail, and most Australian state road authorities publish supplementary guidance aligned to local conditions. For projects involving detailed intersection modelling, software tools including SIDRA and LinSig accept field-measured saturation flow rates as direct inputs, which produce more accurate capacity assessments than relying on default values alone.
Accurate saturation flow data also matters for transit signal priority systems. When a bus or tram triggers a green extension or early green at an intersection, the system needs to know how much green time the adjacent competing movements will lose and whether those movements can absorb the reduction within the same cycle. That calculation depends directly on saturation flow rate for each affected lane group.
Practical implications for signal engineers and transport authorities
For councils and transport authorities commissioning signal timing reviews, it's worth specifying field-measured saturation flow rates rather than accepting default values. The additional survey effort, typically 2 to 4 hours of observation per intersection movement, adds a small amount to project cost but substantially improves the reliability of the timing plans that follow.
For adaptive systems, the requirement shifts from periodic field surveys to maintaining the quality of the detector data the system uses to estimate saturation flow continuously. Detector faults, whether from damaged loops, misaligned radar heads, or communication drop-outs, degrade the saturation flow estimates the system builds from real-time traffic. Preventive maintenance on detector infrastructure is therefore directly connected to the performance of adaptive signal control, not just to counting accuracy.
Bob Panich Traffic Signals designs and delivers traffic signal systems that account for local saturation flow conditions from the planning stage through to commissioning, ensuring that adaptive and fixed-time installations operate with accurate capacity inputs from day one.