Stop/Slow Cycle Times: How to Work Out Green Time, Clearance and Queues

A stop slow cycle time is the total time to run one full rotation: green time for one direction, clearance while the last vehicle travels through, green time for the other direction, then the second clearance. Work holds are added on top. The cycle sets how long each approach is stopped, and therefore how long each queue grows.

Getting the cycle right matters because every part of a stop/slow operation flows from it: queue lengths, where the end-of-queue warning goes, how long drivers wait and whether the site can cope at peak times. This guide shows the arithmetic with a worked example. The numbers are illustrative estimates for planning, not design values. The TGS must be prepared by a suitably qualified person using the current guide, and on site the controllers' confirmation that the section is clear always overrides any timer.

The four parts of a stop/slow cycle

The cycle time is the sum: C = G(A) + T(A) + G(B) + T(B) + any hold.

Each approach is stopped for everything except its own green: stop time for A = C − G(A). That stop time is what drives the queue at A.

Step 1: Work out clearance time

Clearance time is the time for the last vehicle released to travel the length of the single-lane section and clear the far control point.

Clearance time (s) = single-lane length (m) ÷ travel speed (m/s) + allowance

To convert km/h to m/s, divide by 3.6. A 250 m section travelled at 40 km/h (11.1 m/s) takes about 22.5 seconds. Add an allowance for vehicles that accelerate slowly from a stop, heavy vehicles, cyclists, and drivers who crawl past the work. In the example below we add 15 seconds and round to 40 seconds each way.

Clearance is pure lost capacity: no one moves through the site from either direction while it runs. That is why long single-lane sections hurt so much, and why AS 1742.3 and the AGTTM include guidance on the maximum length of a single-lane section for manual control.

Step 2: Work out the arrivals

You need the arrival rate for each direction separately, at the busiest time the site will be operating. A two-way daily figure is not good enough: peaks are often strongly directional. Convert to vehicles per second or per minute:

• 240 vehicles per hour = 4 per minute = 0.067 per second
• 180 vehicles per hour = 3 per minute = 0.05 per second

Note the proportion of heavy vehicles too, because trucks take longer to discharge and occupy more queue length.

Step 3: Work out the green time each direction needs

When a controller releases a queue, vehicles move off at a roughly steady rate once the first few are moving. Call the average gap between vehicles passing the controller the discharge headway. For planning, we assume 2.5 seconds per vehicle in mixed traffic. Real values vary with heavy vehicles, grades, driver hesitancy and how the controller works, so treat this as an assumption.

To keep queues from growing, each direction's green must discharge everyone who arrives during a full cycle:

Minimum green (s) = arrival rate (veh/s) × cycle (s) × discharge headway (s)

Because green times are part of the cycle, the minimum cycle that just balances is:

Minimum C = total clearance ÷ (1 − sum of (arrival rate × headway) for both directions)

Worked example: a 250 m single-lane section

Assumptions, all illustrative: 250 m section, 40 seconds clearance each way, 240 veh/h northbound, 180 veh/h southbound, 10% heavy vehicles, 2.5 seconds discharge headway.

• Northbound demand on green: 0.067 × 2.5 = 0.167 of the cycle
• Southbound demand on green: 0.05 × 2.5 = 0.125 of the cycle
• Minimum cycle: 80 ÷ (1 − 0.292) = about 113 seconds

At 113 seconds the site is balanced on a knife edge: every second of green is used, and any surge, slow truck or hesitant driver makes the queue grow. A practical cycle has spare green. Try 150 seconds:

Queue length is converted using allowances of about 8.5 m per car and 22 m per heavy vehicle, including the gap to the vehicle in front. One state method uses similar figures; the AGTTM provides its own method based on a five-minute count, and that is the one the designer will normally use. Our stop/slow shuttle cycle planner runs this arithmetic for other inputs.

What a work hold does to the queue

Now suppose the crew needs both directions held for 3 minutes so an excavator can cross. Northbound stop time becomes 110 + 180 = 290 seconds, almost 5 minutes. At 4 vehicles per minute that is about 20 vehicles, or roughly 200 m of queue, more than double the normal queue.

Two consequences follow:

1. The end of the queue moves upstream. If the end-of-queue warning was positioned for the normal cycle, drivers now meet stopped traffic before the warning. Holds need to be in the design, not added on the day.
2. Recovery takes several cycles. Twenty vehicles need about 50 seconds of green to discharge, more than the 40 seconds adopted. The queue shrinks only as spare green chips away at it over following cycles.

Watch for the tipping point

The danger with stop/slow is that it degrades suddenly. In the example, if northbound volume rises from 240 to 400 veh/h at school pick-up, the green needed in a 150 second cycle becomes about 42 seconds, more than the 40 seconds adopted. The queue now grows every cycle and does not recover until volume drops.

Lengthening the cycle helps a little, because clearance time becomes a smaller share of it, but it also lengthens every stop and every queue. Past a point, the right answer is a different method or time, not a longer cycle. Signs you are approaching the limit:

• greens are running long and controllers are still sending queues away unfinished
• the queue at one end is passing the end-of-queue marker
• the queue is approaching a crest, curve, intersection or rail crossing
• drivers are queue-jumping, turning around or abusing controllers

Our guide to roadworks queue monitoring covers what to record and when to escalate. Options such as portable traffic signals, a shorter single-lane section, working outside peak periods or a detour must be assessed by the designer.

Running the cycle on site

The calculation is for planning. On the day, the cycle is run by people, and the safety rule is absolute: do not release opposing traffic until the last vehicle is confirmed clear.

• Confirm the tail. The releasing controller describes the last vehicle by radio (for example "last is a white tipper"), and the receiving controller confirms when it has passed. Our guide to traffic control radio communication covers the protocol.
• Plan for lost radio contact. Agree the fallback before work starts, such as holding both ends until contact is restored.
• Watch for slow vehicles. Cyclists, tractors, oversize loads and learner drivers can take far longer than the planned clearance.
• Balance by observation. If one queue is consistently longer, adjust green times within the limits of the TGS and tell the supervisor.
• Record changes. Note when timings were changed, why and what happened to the queues.

Portable traffic control devices and portable traffic signals run programmed clearance periods. The same logic applies, but the timings are set and checked by competent people in line with the device's operating requirements. Our portable traffic lights versus stop/slow bats guide compares the options, and stop/slow traffic control covers the method itself.

Next steps

For national guidance on manual traffic control and end-of-queue management, start with Austroads; in NSW check TS 05492:1.0 with Transport for NSW, and in Queensland the supplements published by Transport and Main Roads. If you run a traffic control company, TrafWise keeps the TGS, shift records and queue observations together for each job.