Sydney Coordinated Adaptive Traffic System

"SCATS" redirects here. For other uses, see SCATS (disambiguation).

The Sydney Coordinated Adaptive Traffic System, abbreviated SCATS, is an intelligent transportation system platform for monitoring, controlling, and optimising the movement of people and goods in cities through the operation of traffic signals.^[1]: 2

SCATS manages the dynamic (on-line, real-time) timing of signal phases at traffic signals, meaning that it tries to find the most efficient phasing (i.e. cycle times, phase splits and offsets) for a traffic situation (for individual intersections as well as for the whole network). SCATS is based on the automatic plan selection from a library in response to the data derived from loop detectors or other road traffic sensors.

SCATS typically uses detectors at each traffic signal to detect vehicle presence in each lane^[2]: 15 and push-buttons for pedestrians waiting to cross.^[2] The vehicle detectors are generally inductive loops installed within the road pavement, however microwave (radar) detectors can be used as a temporary measure when loop detectors fail.^{[2]: 15 [3]: 7} Similar loop detectors can detect metal bicycles or metal parts on bicycles.^{[3]: 11 [4]} Push button detectors are usually provided for pedestrians.^[3]: 7 Various other types of sensors can be used for vehicle presence detection,^[3]: 7 provided that a similar and consistent output is achieved. Information collected from the vehicle sensors allows SCATS to calculate and adapt the timing of traffic signals in the network.

SCATS is installed at more than 63,000 intersections in over 216^[5] cities in 33 countries.^[6] In Australia, where the system was first developed, the majority of signalised intersections are SCATS operated (around 11,000^{[citation needed]}).

The SCATS system is owned by the Australian state of New South Wales, whose state capital is Sydney. Transport for NSW (TfNSW) is the transport and road agency in New South Wales. In December 2019 TfNSW began to look into commercialising the SCATS system,^[7] however these plans were abandoned in 2021.^[8]

Features

Default operation

The architecture of SCATS is at two basic levels, Tactical control (local) and Strategic control (regional, known as MASTER^{[citation needed]}).^[9]: 182 The on-site local controller (cabinet at the roadside) processes of traffic information deduced from the vehicle detectors and allocates green time, extending, terminating early, or skipping demand-dependent phases.^[9]: 182

Strategic control (regional) is a regional computer which target cycle length, splits, and offsets for subsystems based on detector data.^[9]: 181 It provides area based traffic control, i.e. area traffic control (ATC) or urban traffic control (UTC).^{[citation needed]} Detailed traffic signal and hardware diagnostics are passed from the LOCAL to the MASTER, with the ability to notify staff when a traffic signal has a fault.

Typical operating modes for SCATS intersections include SCATS Isolated, Flexilink, Masterlink and Non-SCATS sites.^[9]: 182

SCATS is able to operate over PAPL, ADSL, PSTN and 3G IP network connections to each intersection.^{[10][11][12][13][circular reference]} SCATS can also operate on a network of private cables not requiring third party telecommunications support and large parts of inner Sydney previoulsy operated this way.^[11][12]

Public vehicle priority

Public vehicle priority in SCATS (using data provided from PTIPS) caters for both buses and trams. SCATS has a facility to provide three levels of priority:

• High – In the high priority mode the "hurry call" facility is used. (i.e. the phase needed by a bus, tram or emergency vehicle is called immediately, skipping other phases if necessary)
• Medium (Flexible window) – Phases can be shortened to allow the bus/tram phase to be brought in early. The bus/tram phase can occur at more than one place in the cycle.
• Low – takes its turn.

Trams would normally be given high priority, the aim of which is to get the tram through without it stopping. Buses would normally expect to receive a medium level of priority.

Pedestrian and Bicycle Priority

Pedestrian priority can be achieved by using lower cycle times, double cycling, 'Walk for Green', automatic introduction of pedestrian phases, countdown timers, and 'delinking' or 'divorcing'.^[14] Lower cycle times reduce the wait time and delay for pedestrians.^[14]: 10 Other factors impacting pedestrian access include the number of opportunities in the cycle when pedestrian can cross (often one), and pedestrian demands not being recorded in time for action by the signal controller.^[15]: 8 SCATS records pedestrian demands and this can be accessed from event logs in SCATS History.^[16][17]

The SIDRA User Guide states when pedestrians have to wait 20 or 30 seconds (average delay per pedestrian) the delay is noticeable. When the delay is 30-40 seconds there is an increased likelihood of risk taking as the delay is irritating. Risk taking behaviour is likely with a delay of 40-50 seconds and above 60 seconds there is a high likelihood of risk taking as the delay exceeds tolerance level.^{[18][15]: 8}

Cycle times

In SCATS the nominal cycle length (nCL) is a reference in used to set initial splits and coordination. The running cycle length (rCL) varies from cycle to cycle because phase times vary with demand. SCATS can change nCL dynamically.^[9]: 183 An example diagram of nCL vs rCL, including skipped phases, shows actual cycle times (rCL) of between 108 s to 176 s for an intersection with nominal cycle time (nCL) of 140 s.^[9]: 183

A graphical program called SCATS History Viewer has export functionality for actual phase and cycle statistics.^[9]: 185 The program facilitates the extraction of this data into commonly used formats that can be easily consumed by other systems.^[19]

Longer or more frequent green signals for pedestrians can reduce unsafe crossing by 34%.^[20][6]

Instant fault detection and quick repair

The ATC system is equipped with the function of fault detection and logging the fault detected in order to facilitate repair and maintenance. Should there be a telecommunication breakdown, the ATC junction controller concerned will switch to standalone mode and continue to function.

Traffic Adaptive Operation

ATC systems provide advanced method of traffic signal control called Traffic Adaptive Control where the operational timing plans including cycle length, splits and offsets are continuously reviewed and modified in small increment, almost on a cycle-by-cycle basis, to match with the prevailing demand measured by the detectors connected to the on-street traffic controllers.

SCATS Ramp Metering System

The SCATS Ramp Metering System (SRMS)^[21] is a SCATS subsystem and controls traffic signals at the entries of motorways and integrates with SCATS intersection control for promoting integrated real-time management of the traffic corridor as a whole. The objective of SRMS, based on current traffic conditions, is to efficiently determine:

• When ramp metering signals start and end ramp metering operation
• The metering flow rates of the operating ramp metering signals
• Which actions shall be taken to signalised intersections of the corridor to promote network-wide benefits.

SRMS achieves these objectives by implementing a collection of pre-configured adaptive intelligent strategies either automatically or manually. In manual mode, the SRMS operator can create new or manipulate existing rules in order to adjust the ramp metering system for effective operation during any planned or unplanned events (e.g. incidents). SRMS is a distributed control system that operates on a central control server and road-side traffic controllers. The central control server is a component of SCATS and inherently provide integrated motorway and arterial real-time management. The road-side controllers are installed on motorway on-ramps and are used to:

• Set the traffic signal times
• Set the state of on-ramp changeable signs
• Manage the sequences start and end ramp metering operation; and
• Measure traffic states using vehicle detectors.

Metering rates are determined by the local traffic signal controller or by the central control server. Metering rates can be determined in two ways:

• adaptive operation, or
• time-of-day-based operation typically when a communications failure or critical vehicle detector failures take place

The adaptive operation optimises mainline traffic state by using real-time data from vehicle detector stations installed at several mainline locations, ramps and optionally at arterial roads. The adaptive operation determines control actions at 10 seconds intervals and applies some or all of the following strategies simultaneously:

• Coordinated ramp metering
• Ramp queue management
• Automatic begin and end of ramp metering operation
• Variation routines for integration with SCATS intersection control
• Variation routines for automated incident responses and unusual circumstances
• Manual controls for incident responses and unusual circumstances
• Critical lane occupancy calibration
• Fault-tolerant strategies
• Data logging for performance reporting and off-line analysis

SRMS is currently used as the Auckland ramp metering system.

Simulation

SCATS can be simulated in-the-loop (SCATSIM) using third party traffic simulation tools. SCATSIM offers an interface supported by Aimsun, PTV VISSIM, Quadstone Paramics and Commuter. SCATSIM offers kerb-side hardware and firmware emulation that interfaces seamless to the SCATS Region and Central Manager offering the same control strategies used in field deployments for both intersections and ramp metering (SRMS). The configuration files prepared by authorities for the Central Manager, Region, SRMS and kerb-side controllers can be re-used without modification by SCATSIM.

When Commuter software was acquired by Autodesk,^[22] Azalient Ltd support for the Commuter interface was deprecated. Azalient Ltd also developed a plugin that enabled the Quadstone Paramics interface to SCATSIM. This plugin is also deprecated.

The Sydney Strategic Travel Model was designed by Hague Consulting Group in 1997.^[23]: iii

History

SCATS was developed in Sydney, Australia by the Department of Main Roads (a predecessor of Transport for NSW) in 1975.^[5] It began to be used in Melbourne in 1982,^[24] Adelaide, South Australia in 1982 and Western Australia in 1983.^[25]

It is also used in New Zealand, Hong Kong, Shanghai, Guangzhou, Amman, Tehran, Dublin, Rzeszów, Gdynia, Central New Jersey,^[26] in part of Metro Atlanta,^[27] and Cebu City,^[28] among several other places. In Hong Kong, SCATS is currently adopted in the area traffic control systems at Hong Kong Island, Kowloon, Tsuen Wan and Shatin.

The system may be referred to by an alternative name in a specific installation. However, since deployment outside Australia, New Zealand and Singapore, localised names do not appear to be commonly used. The following are some local alternative names that have been or are in use:

• Canberra "CATSS" (Canberra Automated Traffic Signal System)
• Melbourne "SCRAM" (Signal Co-ordination for Regional Areas of Melbourne)
• Adelaide "ACTS" (Adelaide Co-ordinated Traffic Signals)
• Perth "PCATS" (Pedestrian Countdown at Traffic Signals)
• Singapore "GLIDE" (Green Link Determining System)
• Northern Territory "DARTS" (Dynamic Arterial Responsive Traffic System)

SCATS Traffic Signal Operation

In New South Wales, Australia

Transport for NSW (TfNSW) is responsible for controlling signals in New South Wales, the same agency which develops the SCATS software. As of November 2025, there were 4860 traffic signals across NSW.^[29]

In NSW from 2022 to 2024, road crashes at signalised intersections injured 666 pedestrians and resulted in the deaths of 19 people. In 2025, at least 8 pedestrians were killed at traffic lights.^[30]

Since 1994 automated automated pedestrian phases were enabled from 7am to 7pm on Monday to Thursday, and from 7am to 9pm on Friday at all intersections across the Sydney CBD.^[31][32]

Simple signal changes can give more time or priority for people to safely cross the road.^[33]: 10 In January 2018, Transport for NSW reduced the cycle time for a subset of the Sydney CBD from 110^[34] to 90 seconds.^[35][36] The Lord Mayor of the City of Sydney council wrote to the Roads Minister to request a broader rollout of 90 second cycle times on the 8th of November 2018.^[35][36] The Roads Minister declined to comment on the request.^[36]

Transport for NSW commenced a broader roll out of automatic pedestrian signals on 23 March 2020.^[37][38] During COVID, pedestrian crossings at traffic lights were automatically activated 24/7 across the CBD and areas of the inner city, including Darlinghurst, Surry Hills, Pyrmont and Ultimo. Automated pedestrian crossings were expanded to key health precincts were also gradually rolled out from March 2020. Automation was not implemented state wide as it "would unnecessarily impact on traffic flow".^[39]

In 2022 TfNSW began to remove the pedestrian button covers in Sydney. Crossings around health precincts were no longer automated, however automation remained during daylight hours in the CBD.^[40] In the core of the CBD of Sydney, pedestrian signals remain automated during a portion of the day as of 2022.^[41]

Due to noise complaints caused by the audio warning when the green man is first displayed, the hours of operation were reduced between November 2022 to January 2023. As of July 2023, in the CBD of Sydney, the pedestrian auto-call feature runs from 6am to 10pm.^[32]

Automated pedestrian crossings also run on King Street Newtown and the Parramatta CBD. On some key transport routes in the Inner Sydney areas (such as the light rail corridor in Devonshire Street), automated pedestrian crossings do not operate as light rail is the priority mode.^[32]

The NSW Pedestrian Protection Program was launched in August 2015. This involved upgrading SCATS intersections with red turn arrows for vehicles or leading pedestrian intervals (a head start) for pedestrians to increase safety. An conducted from October 2021 to December 2022 found significant reduction in Fatal and Serious Injury pedestrian-involved crashes (between 43% to 47%), as well as a reduction of 20% (not statistically significant) to 38% (statistically significant) in overall pedestrian-involved crashes. In 2023 Transport for NSW published a summary report on the evaluation of the program.^[42] Stakeholders identified that despite initial reservations by some that the implementation may lead to greater vehicular congestion, there was no evidence to suggest this occurred.^[42]: 3

The Monash University Accident Research Centre assisted with the statistical analysis for the project.^[43] The program was discussed at conferences in 2022^[44] and 2023.^[45] The program was highlighted in a study conducted under the United States Department of Transportation Federal Highway Administration Global Benchmarking Program.^[46]

The TfNSW Transport Modelling Guidelines state in general for new signals a "Nominal cycle time of 140 sec." should be applied for new signals and "Cycle shorter times than 140 sec may apply to off-peak traffic, to intersections along minor routes, and to isolated intersections."^[9]: 191

The SCATSIM modelling methodology^[9]: 187 requires SCATS data from TfNSW Network Operations, and includes two review stages by TfNSW Network Operations.^[9]: 187

A trial of a new sensor at Manly has found longer or more frequent green signals for pedestrians reduced unsafe crossing by 34%.^[47][30] The trial uses a FLIR TrafiOne camera sensor,^[6]: 11 and has four thresholds. When the pedestrian occupancy is 0-8%, a 6 second green is shown. For 9-15%, 8 seconds, for 16-27% 10 seconds, and for above 28% pedestrian occupancy, 12 seconds green.^{[6]: 12, 14}

A similar infrared camera system will be installed at the intersection of Pitt Street and the Great Western Highway, Parramatta, located near a high school and several apartment buildings. Installation was planned to begin from the week of the 1st of December 2025. The location will expand the trial scope to detect cyclists as well as pedestrians and vehicles.^[30] Five cameras will be installed at the north-east, north-west and south-west of the intersection.^[48] Changes to increase "operational efficiency" were proposed in August 2017 at this intersection.^[49][50] Modifications and to the traffic signal layout at this intersection were made in 2018.^[51] Upgrades were complete by July 2018.^[52] Signal timing was previously optimised at this intersection in May-June 2025.^[53]

In Western Australia, Australia

Main Roads Western Australia previously published a real-time websocket feed of signal timing.^[54] As of 2025 WA Main Roads publishes historical SCATS traffic signal phase data under an open source Creative Commons CC BY 4.0 license.^[55] Data is published in monthly machine-readable Parquet files.^[56] Data includes times for each phase and measured volume.^[57]

On the TrafficMap website WA Main Roads openly publish Detector Volume Data,^[58] Pavement and Signage Drawings,^[59] Traffic Signal Arrangement Drawing,^[60] Signal Data (including Phase Times, Pedestrian Phase Times, Special Times, Link and Offset Plans, and SCATS Phase History tables)^[61] and Phase Sequence Charts^[62] for every signal in the state.^[63]

In Victoria, Australia

The Department of Transport and Planning are responsible for the safety and efficiency of traffic signals throughout Victoria.^[64] SCATS was trialled in Melbourne in 1978 and adopted for use throughout Victoria in 1980. In 2018, SCATS controlled more than 4,000 sets of traffic signals across Melbourne and other Victorian rural cities such as Ballarat, Bendigo, Traralgon, Geelong and Mildura.^[65]

DataVic publish Traffic Signal Volume Data sourced from the detector loops and the SCATS system under an open-source Creative Commons CC BY 4.0 license. Volume data is available for the day two days prior to the current date.^[66] Historical data is published back to 2014.^[67] Historical Annual Average Daily Traffic Volume data is published from 2001 to 2019 in the GeoJSON format under CC BY 4.0.^[68]

DataVic also publishes Traffic Signal Configuration Data Sheets, also known as 'operation sheets' or 'op-sheets'.^[69] These operation sheets detail signal group and detector functions at each intersection along with the phasing of the site. They include detailed notes outlining the specific operation of signal groups, phases, detectors and general site operation, the traffic signal sequences (phases), and the phase and pedestrian time settings which govern how the site operates.^[70]

In South Australia, Australia

The City of Adelaide typically has green walk signals in the range of 5-8 seconds.^[15]: 11

Auto pedestrian demand was implemented in the City of Adelade before 2015. As of March 2025 it was is enabled at 49 locations between 7am and 7pm. Some sites, including North Terrace at the Railway Station, and the intersection of King William Street and Rundle Mall and Hindley Street, are enabled 24/7.^[15]: 7

A March 2025 review of signal timing along O’Connell Street in North Adelaide in found that reducing the cycle length could reduce the average pedestrian delay by 14 seconds and allow between 6-21 extra opportunities for people walking to cross during midday peak hour. Reducing signal cycle times was also found to reduce queue lengths and reduce average delays for vehicles by 38-46% in the PM and 50-54% in the midday peak.^[15]: 19

See also

• PTIPS – works together with SCATS to provide transport vehicles with priority at traffic signals

Other Intelligent Transportation Systems include:

• STREAMS
• BLISS
• Inductive loop vehicle detection