Split Cycle and Offset Optimisation Technique

World famous adaptive urban traffic signal control system

SCOOT® is a real-time method of signal control that puts the road users’ experience first and centre. Where traffic signals sit close together, coordinating adjacent signals can offer a dramatically improved experience. Originally developed by TRL in the 1970s, SCOOT® alongside traffic control in general, has been subject to constant development through to the present day.

Unique to TRL Software, SCOOT® 7 embraces future mobility. As our networks become increasingly multi-modal – whether that is cycling, public transport, walking, e-scooters or even connected vehicles – SCOOT® 7 is able to adapt to meet the demands of this ever-evolving environment. Not only does it adapt to the changes in mobility, but also opens increasingly readily available data set for Road Authorities to use for research and development. Exploiting your existing infrastructure, SCOOT® 7 is ready to enable future mobility now.

SCOOT® Version 7.0 is the first TRL Software only release of SCOOT® and introduces new features which broaden the toolbox for adaptive traffic control and the changing needs of mobility, specifically:

  • GLOSA, or cooperative signals data to provide road users with information that can help with their journey
  • Multiple split optimisation allows the optimiser to make larger changes without the compromises
  • Pedestrian SCOOT functionality for green man period optimised to number of pedestrians
  • Modelling link departures to help with optimisation in general
  • Loop failure logic to reduce detection requirements

Finally, unlike other at market UTC systems running SCOOT®, up to v.6.1.10, TRL Software have moved away from the previous reliance on OpenVMS with a cloud first approach, this brings the additional benefits of modern technology stack which required no specialist or on premises infrastructure being required.

SCOOT® Version 6.1-11 MMX, 2016 was the final maintenance release of SCOOT® available through the previous distributors, Siemens and Dynniq, which addressed bug fixes only. For information the fixes provided are:


Bit fields Test results show differences between Debug build and Release build. the Microsoft Visual Studio compiler in Release build does not correctly handle a particular expression containing differently sized bit field variables
Bus priority max It is possible for a stage to exceed its maximum length following bus priority recovery.
Offset min/max The offset optimiser can cause the minimum or maximum stage lengths to be violated.
Split reversions min/max During multiple reversion, the reversion modifier is set to 0.
LS recovery max The split optimiser can cause a stage maximum to be violated after long stage (LS) recovery
Lopsided offset authorities Offset authorities can appear lopsided
Conflicting local variables The following local variables in b21c.c could conflict with database variables, e.g. model_eb with Model_eb(node), depending on how the software is built.
Ghost reversions The stage change times can get out of order following a split optimisation and the appearance of a ghost stage
Ghost M17 message The M17 message does not include time when time now is frozen due to a ghost stage
T04 out by 1 The T04 message shows the current stage time out by 1 second.
Split advance min/max Sometimes a split advance or retard can cause violation of a minimum stage length.
Rev mod cleared during multiple reversion During multiple reversion, the reversion modifier (Rev_mod) is set to zero.
Summary of bus link corrupted The timer and other values in the B19 message can become corrupted.

SCOOT® Version 6.1 MMX Service Pack 1 was originally developed for Transport for London (TfL), as JTR – GOLD (Journey Time Reliability – Games Operation Led Development) for the London 2012 Olympic Games. SCOOT® MMX Service Pack 1 provided a new facility to assist with improving the Journey Time Reliability (JTR) along key routes.

SCOOT® Version 6.1 MMX provided new facilities to prioritise pedestrians at junctions with a variable invitation to cross period, a significant update and enhancement of emissions model estimates in addition to features to improved traffic flow and operation during low flow periods.

Improved traffic flow and operation during low flow periods is achieved through:

  • Node and sub-region cycle time independence which enables the ability to run a single node or a sub-set of nodes in a region to run at a different (lower) cycle time than the SCOOT® Region.
  • Ghost Staging – Reduced cycle time at quiet times using a Ghosted Stage
  • Zero Demand Queue

SCOOT® Version 5.1 MC3 was enhanced through service pack 1, which incorporated the results of research undertaken into pedestrian priority. Additionally the release included  online modelling capabilities through a SCOOT® interface to VISSIM, enhanced Split optimisation for stages on minimum and Maximum, Inter Green feedback with negative processing lag, partial exit blocking detection and queue clear time on long links.

SCOOT® Version 5.0 MC3 introduced timestamped data to enable new communications between the in and outstations in a transparent way to the system user.

Building on this, the addition of the MONACO congestion supervisor provided continual monitoring of the SCOOT® network looking for unusual, regularly occurring SCOOT® data which might indicate an inefficiency in performance. Additionally, improved control at puffin crossings and junctions with puffin pedestrian facilities and variable intergreens were introduced.

SCOOT® Version 4.5 introduced a Simplified Validation approach where SCOOT® outputs the estimated queue clear time for a range of saturation occupancies, reducing the time taken in validation / revalidation of the system. The introduction of supplementary SCOOT® detection inputs enabled improved modelling of complex links, where there are significant, irregular sources and/or sinks of traffic such that the on-street queue is often different to that modelled by SCOOT® when using standard detection locations.

Enhancements were made to bus priority where an increasing priority level will enable SCOOT® to give increasingly aggressive priority dependent on the importance, finally, continued improvements were made to cycle time optimisation.

SCOOT® Version 4.4 provided emissions optimization, new gating logic and enhanced public transport priority, and on the benefits obtained in testing these facilities.

  • Enhanced bus priority
  • Enhanced Gating
  • Emissions optimisation
  • Supplementary detectors
  • Double cycling
  • Filter links in over saturation
  • Automatic selection of alternative stage

SCOOT® Version 4.2 built on the previous release and provided the ability to model Flared Links which have additional lane capacity downstream. Parameters are established to allow for the extra lanes in terms of the number and length. Recovery from Absolute Priority where a controller “hurry call” is used to provide priority to an emergency or light rapid transit (LRT) vehicle and finally Reduced detection modelling.

SCOOT® Version 4.0 introduced the online estimation of vehicle emissions of pollutants, namely CO2, CO, NOx, VOC and PM. The emission estimates are based on a model developed under contract to the UK Department of the Environment, Transport and the Regions (DETR). The ASTRID database can be used to store these estimates and allow historic patterns of emissions to be analysed.

Several enhancements were made to the cycle time optimiser to increase its responsiveness to traffic and to network manager commands, finally, new logic was added to use information from existing stop line loops where it is was not possible  to install SCOOT® detectors in the normal position.

SCOOT® Version 3.1 was extended to include active bus priority for the first time, making use of Active Vehicle Location (AVL) and Selective Vehicle Detection (SVD), the automatic SCOOT® traffic information data base (ASTRID) system, and the INGRID incident detection system and has been given added flexibility, particularly for use in incident condition.

As well as providing current and historical information to traffic engineers, ASTRID now can feed historic information back into SCOOT®, providing a substitute cyclic flow profile that can be used for optimization when there are faulty detectors. The INGRID incident detection system contains two algorithms to provide an indication of an incident; taking current information directly from SCOOT®, INGRID detects abnormal changes in flow and occupancy, and comparing current information with historic information stored in the ASTRID data base, INGRID detects abnormal patterns in these parameters. The SCOOT® optimizers have been made more flexible and can now make larger changes to the signal timings if required.

Related Products