Table of Contents

6.11.2 Best Practices in Active TSP Provision

TSP requires a balance between benefits to transit and impacts on other road users. The aim is to maximise transit performance while minimising the impacts on others. Table 6.11 outlines best practice in this optimisation.

Table 6.11: Optimising TSP performance – best practice

Good practiceWhy
Avoid excessive priorityBenefits of TSP systems to provide effective transit decline at high degrees of intersection saturation. At road saturations above 1.0 there is often no benefit at all. This is because buses (and trams) share road space (and time) with traffic. If the traffic system is seriously disrupted, as a result of excessive green time in providing TSP, the traffic queues that result can act to disrupt public transport as well as general traffic.

Implement conditional priority:

  • add constraints to maximum and minimum green times
  • use weighting systems which encourage some phase retention to reduce delays to non-transit vehicles
  • initiate priority only when buses are late.

Constraining the time available for extension or early provision of green phases is a common approach to balance time inputs for all road users. Some examples are:

  • The degree of saturation at each intersection is used to compute a ‘spare green time’ value. This is the difference between the actual degree of saturation occurring and the target saturation which traffic engineers have agreed for each intersection.
  • Many TSP systems also tend to permit ‘priority free’ running of the traffic signal system after priority has finished enabling traffic flows to be re-balanced. This includes limits to the number of consecutive signal cycles where priority can be called.
  • TSP should be targeted to only buses (or trams) which are running late.
Consider need for indirect priority/traffic meteringIndirect priority approaches, where traffic signals far ahead of the public transport vehicle are adjusted to clear downstream traffic well ahead of arrival at intersections are also considered good practice in TSP. This approach is related to ‘green wave’ measures adopted for area traffic control systems but applied to TSP. A variation of this approach is to ‘meter’ the number of vehicles within the road network thus limiting congestion and associated queues. This approach is considered particularly effective for systems prone to high congestion levels (and hence ineffective TSP).
Consider need for active pedestrian detection

Safety for pedestrians (and other road users in general) is considered one of the most important principles in designing traffic signals. Warrants for the design of signals identify minimum pedestrian clearance times to ensure slower pedestrians can clear crossings safely. This acts as a timing constraint which limits how signals can be ‘closed down’ to permit TSP.

Pedestrian detection devices ensure don’t walk phases can be run only when demanded, freeing up time in the signal system to provide green time for all intersection users.

Phase compensationPhase compensation occurs when green time removed for transit vehicles is replaced in following cycles. It is a strategy which permits traffic queues built up as a result of priority to be dissipated soon after priority events. As a result it acts to balance demands for time between road users. However in dissipating queues it can act to assist the progress of transit vehicles.

Source: Austroads (2007f), summarised.

Accuracy in estimating the arrival time of buses at signals is a major requirement for successful TSP design. If buses arrive before predicted arrival times, signal priority (such as a green extension) may be called and not used wasting time for other road users. If buses arrive after predicted times, they may not receive priority even though extra green time has been provided. Several factors can act to improve the accuracy of bus arrival time estimates, as outlined in Table 6.12.

Table 6.12: Ensuring accurate arrival time estimates

Good practiceWhy
Continuous vehicle detection and automatic vehicle location systems

Short range detection of buses causes ‘abrupt’ needs for priority which can often not be catered for within the safety requirements of signals. A better approach is to gradually adjust cycles by a smaller amount using longer range advance detection (an indirect priority approach). Where detection is close to signals abrupt ‘aggressive’ signal changes are required which can seriously disrupt traffic.

A range of criticisms of loop detection systems have been made:

  • Loop-based systems can be vulnerable to failures in detectors. Clearly vehicle tracking is not possible in a loop-based system without working detectors and hence it is less accurate.
  • Loop-based systems are also imperfect tracking systems in that they only detect vehicles at certain points. If vehicles are delayed between the detector point and the stop line then the provision of signal priority may be wasted.

The use of automatic vehicle location (AVL) systems to provide more continuous monitoring of vehicle locations can assist in estimating the arrival time of vehicles more accurately. This strategy can also be combined with techniques such as ‘green wave’ or ‘rolling horizon’ to provide better estimates of arrival times at upstream stops and adjustments to these stops over a longer period (an indirect strategy).

Removing approach stops

An important factor affecting the accuracy of vehicle arrival prediction is the location of nearside or approach stops at intersections. When stops are located near to signals there is much variation in the amount of time it will take for the bus to reach the stop line. Travel time will be much slower if passengers board at the stop than if the bus proceeds without stopping. This means the TSP system must estimate an average time considering both alternative outcomes.

An alternative approach is to ensure that priority is not called when priority vehicles have to board/alight passengers at approach stops (e.g. link the activation of the priority to bus doors).

Relocation of approach-side stop to far-side locations is considered more compatible with effective priority provision. Approach stops cause more delays in TSP applications than equivalent departure stops. Removal of approach stops and their replacement to far-side locations was shown to reduce delays to buses by almost 50%.

Phase compensationPhase compensation occurs when green time removed for transit vehicles is replaced in following cycles. It is a strategy which permits traffic queues built up as a result of priority to be dissipated soon after priority events. As a result it acts to balance demands for time between road users. However in dissipating queues it can act to assist the progress of transit vehicles.

Source: Austroads (2007f), summarised.