Review of Traffic Signals on High-speed Roads

Review of Traffic Signals on High-speed Roads


Alastair Maxwell, Keith Wood, TRL, UK


This paper reviews the ?dilemma zone? (whether to brake hard or risk red running) on higher speed approaches to traffic signals, policy and practice used to mitigate the safety consequences of this, and the classification of a ?high-speed? approach.



At the end of green, drivers approaching traffic signals can be faced with a decision of whether to stop or continue. As speeds increase the decision becomes more difficult and the safety consequences of making the wrong decision greater. Some drivers, depending upon their speed and distance from the stopline, can be faced with a choice of either braking hard or risk running the red, both of which have safety implications.

The distance from the intersection over which there can be some uncertainty of the appropriate action ? whether to stop or continue - is called the ?dilemma zone?.

Various strategies are used in Europe, and around the world, with the aim of reducing the instance of drivers being caught in the ?dilemma zone?, and the safety consequences if caught in it.

In the UK, dilemma zone protection traffic signal control strategies are required on approaches where the 85th percentile speed exceeds 35 mph (where 15% of free-flow approach speeds exceed 35 mph (56 km/h)).
The UK Department for Transport commissioned TRL to reassess the size and position of the dilemma zone, the speed at which traffic signal approaches should be classified as ?high-speed?, and to review policy and practice used abroad.

Note that, any views expressed are not necessarily those of the Department for Transport.


The main strands of the research were:

? An international literature review on traffic signals on high-speed roads, covering: research into the size and position of the dilemma zone; driver reaction times; vehicle braking capabilities and ?acceptable? braking; accidents; policies; and traffic signal control strategies and studies of their effectiveness.
? Site studies using video recordings at five UK traffic signal junctions on high-speed roads, with regards to driver behaviour after the onset of amber.

The results included: driver decisions, whether to continue or stop, by speed and distance to the stopline; red-running; brake reaction times; and decelerations used.

The findings were used to make an assessment as to whether, with current vehicle technology and driver behaviour, the size and the position of dilemma zone has changed since the last two major pieces of research, 'Webster and Ellson' (1965) in the UK, and 'Zeeger' (1971) in the US, and if it has, whether the high-speed road criterion of 35 mph could be raised, and/or traffic signal control requirements amended.

Summary of main results

The 'Webster and Ellson' test track trials showed that at traffic signals acceptable stopping distances for the majority of drivers (90%) were found to approximately equate to a one second brake reaction time and an average deceleration of 3.6 m/s².

The video studies in this project showed that most drivers were not prepared to use decelerations as high as shown by Webster and Ellson, and drivers stop-continue decisions were found to be largely based upon their journey time to the stopline at the onset of amber, not upon braking ability. The results aligned with the 'Zeeger' on-street findings. The propensity to red-run was found to be very high, with 90% of drivers continuing if 2.5 seconds from the stopline and 10% continuing if 4.5 seconds from the stopline (for all speeds between 35 mph and 55 mph (89 km/h)) ? in the UK the stopping amber is three seconds.

In-service Large Goods Vehicle (LGV) braking performance is much lower than cars, with some unable to achieve the stopping distances derived by Webster and Ellson. LGVs were found to be about one-and-a-half times more likely to run-the-red than other vehicle types, and run-the-red later.

Studies indicate that over 40% of personal injury accidents at traffic signals on high-speed roads are rear-end shunts, and between around one fifth and one quarter of personal injury accidents at traffic signals in the UK are recorded as involving a red-running vehicle. The incidence of red-running tends to increase with approach speed. Dilemma zone events account for a notable proportion of red-light violations at higher speed sites, and appear to account for a significant proportion of rear-end shunts.

The main policy in the UK for traffic signals on high-speed roads has been to limit the number of dilemma zone events through specific traffic control strategies. Similar ?green extension? strategies used in the US, have shown accident rate reductions between 35 and 55% when compared to strategies without dilemma protection ? particularly significant was the reduction in rear-end shunts. However the effectiveness of these strategies is reduced at higher flows where there is a tendency for the green to run to maximum.

Recent developments in traffic control include ?optimisation? strategies, which allow green to end when there are no closely following vehicles in the dilemma zone, i.e. negating the chance of a rear-end collision, but not red-running; and specifically providing extensions for large vehicles.

The employment of dilemma zone strategies with linked signals has so far been limited. Research indicates potential safety benefits, without significantly modifying the network timings.

Alternative methods to address the dilemma zone issue include: speed reduction; longer amber periods; advanced warning of the amber period; and red-light running cameras. The effectiveness of each of these strategies is reviewed, and an assessment made as to their applicability.

Preliminary conclusions

While braking performance has improved over the last 40 years, there is no indication that the braking accepted by drivers at traffic signals has increased, or driver brake reaction times lowered. If anything the Webster and Ellson acceptable stopping distances are at the shorter end of observed research, especially for Large Goods Vehicles (LGVs).

Given driver behaviour and some low in-service LGV braking levels, there appears to be no warrant to relax the current UK 'high-speed' road criteria.


Association for European Transport