How Long Will the Queue Be? Estimating Delays From a Lane Closure

Lane closure queue length depends on one comparison: how many vehicles want to pass the works each hour versus how many the remaining lanes can carry. When demand is higher, the queue grows by the difference every hour, and it only shrinks once demand falls below capacity. A simple hour-by-hour table gives a useful planning estimate.

This guide is about continuous-flow lane closures on multilane roads, where traffic keeps moving past the works in fewer lanes. Stop/slow and portable signal sites work differently, because queues build while traffic is held; our stop/slow traffic control guide covers that method. The numbers below are illustrative estimates for planning, not design values. Traffic management plans and TGSs must be prepared by suitably qualified people using data and methods accepted by the road authority.

Why lane closures create queues

A road works like a pipe. On a normal day, a multilane road carries the morning peak because it has enough lanes. Close one lane and the pipe narrows at the works. If more vehicles arrive than can pass the narrowest point, the surplus has to wait upstream.

Three things decide how bad it gets:

• Demand: the vehicles arriving each hour in the direction affected, at the times the closure is in place.
• Capacity: how many vehicles per hour the remaining lane or lanes actually discharge past the works.
• Duration: how long demand stays above capacity.

The queue does not care about the daily total. A road carrying a modest daily volume can still queue badly for an hour if the closure sits across a sharp peak.

Estimating the queue step by step

The method below is a simple demand-versus-capacity (sometimes called input-output) approach. It ignores many real-world details, which is exactly why it is useful: everyone can see the assumptions.

Step 1: Get demand by hour and direction

You need hourly volumes for the affected direction across the whole closure window, plus the share of heavy vehicles. A two-way daily average hides exactly what matters: the peak and its direction. Sources include road authority count stations, council counts and commissioned counts. If the data is more than a year or so old, growth and network changes may need to be considered. The designer and road authority decide what data is acceptable.

Step 2: Choose a capacity, and admit it is uncertain

Capacity is the hard part. The flow one lane can carry past a work zone varies with:

• the share of heavy vehicles, especially on grades
• lane width and lateral clearance to the works
• how and where drivers merge
• on and off ramps or intersections near the merge
• visible work activity, plant movement and trucks entering or leaving the work area
• weather, lighting and the speed limit

Published research, mostly from North American freeways, reports single-lane work zone capacities across a wide range, and several studies find that discharge falls once a queue has formed. There is no universal Australian number to copy. The designer should adopt a value appropriate to the road and the authority's method, and that value should be checked against observed flows on the day.

Step 3: Build the hour-by-hour table

For each hour: queue at end of hour = queue at start + demand − capacity (but never less than zero, and never discharging more than capacity).

Worked example. Two lanes in one direction are reduced to one from 6 am to 10 am. For illustration only, assume the open lane discharges 1,300 vehicles per hour.

Points to notice:

• The peak lasted one hour, but delay ran from 7 am to about 9:15 am because demand after the peak was only a little below capacity.
• The longest wait is for the driver arriving at 8 am: 300 vehicles ahead, discharging at 1,300 per hour, is about 14 minutes of extra delay.
• Total extra delay is the area under the queue over time, about 360 vehicle-hours across the morning in this example.

You can run your own scenarios in our queue length and delay estimator.

Step 4: Convert vehicles into metres

Queued vehicles become a physical length using an average space per vehicle, including the gap to the vehicle in front. Cars in a stopped queue take roughly 7 to 8.5 m each in common planning allowances, and heavy vehicles much more. Mixed traffic with a high truck share needs a larger average.

In the example, 300 vehicles at an assumed 7.5 m each is about 2.25 km of vehicles. Whether that becomes a 2.25 km queue or a 1.1 km queue depends on whether drivers use one lane or two upstream of the merge. In practice the split is rarely even. Queensland's supplement to the national guide notes that drivers often favour one lane and many merge earlier than necessary, which pushes the end of the queue further back than an even split suggests. Plan for the worse case.

How sensitive is the answer?

Very. Run the same example with capacity at 1,200 rather than 1,300 vehicles per hour, a difference of under 8%:

A small change in capacity roughly doubles total delay and adds 45 minutes of queueing. That is why queue estimates should be presented as a range, why permitted hours are set conservatively, and why the real queue must be monitored from the first morning. Our guide to roadworks queue monitoring covers what to record.

Check the estimate on day one

The first closure is your best data source. While a queue is present, have someone count the vehicles passing the end of the taper in five-minute blocks and multiply by 12. That is the capacity the site is really achieving, including the effect of trucks, merging and the works. Note the time the queue first forms, the furthest point it reaches (using fixed landmarks or chainage markers) and when it clears. If the measured discharge is well below the planned figure, tell the supervisor and the designer before the next shift, not after a week of complaints. The same applies if demand is visibly higher than the count suggested.

What makes real queues longer than the estimate

When the queue on the day is worse than the plan, it is usually one of these:

1. Demand was higher than the data. School holidays ended, an event was on, or a nearby closure diverted traffic onto your road.
2. Capacity was lower. A truck-heavy stream on an upgrade, a narrow lane, or plant movements interrupting flow.
3. Merge behaviour. Early mergers lengthen the queue; late mergers forced in at the taper cause stop-start flow.
4. Rubbernecking. Drivers slow to look at the works, especially with bright lights or unusual plant.
5. Incidents. A breakdown or minor crash in a single open lane can stop flow completely.
6. Upstream bottlenecks. Traffic signals or a ramp near the end of the queue interact with it.

Managing the end of the queue

The queue is not just a delay issue; its tail is a crash risk. Drivers on multilane roads are often not expecting to stop, and the end of the queue may sit over a crest, around a curve or past the advance warning signs. In NSW, TS 05492:1.0, which replaced the TCAWS manual on 25 June 2026, added requirements on end-of-queue management; check the current requirements with Transport for NSW. National guidance from Austroads and state supplements cover end-of-queue warning. In Victoria, check current requirements with VicRoads and the project's road manager.

Practical measures the plan might include:

• advance VMS messages warning of delays or queued traffic, which our VMS board hire guide covers
• extending warning signs to beyond the predicted end of queue
• monitoring the queue end and a trigger to act when it passes a set point
• a contingency plan agreed in advance, such as reopening the lane or pausing work

Reducing the queue before it forms

The most effective controls are planning decisions:

• Timing. In the example, starting at 9:30 am instead of 6 am removes the queue entirely. This is why authorities set roadworks hours restrictions on busy roads.
• Keep more lanes open. Narrow lanes or use the shoulder, where the design allows it.
• Shorten the closure. Fewer metres of closure can mean fewer interruptions from plant and trucks.
• Stage the work. Split a long job into shorter off-peak closures.
• Inform drivers. Advance notice lets some drivers change time or route.

All of these change the traffic management plan, so they must be assessed and approved under the project's process. Our guide to traffic management plan review triggers explains when a change needs a formal review.

Next steps

Before you book a peak-hour lane closure, ask for the hourly demand, the capacity assumed and the predicted queue range in writing. If you run a traffic control company, TrafWise helps you roster the right crew for the permitted hours and record queue observations against each shift.