Traffic Signal Coordination

1
Traffic Signal Coordination
2
Why?

• When the traffic signals are placed close enough,
  it is often seen that the cars waiting in a queue
  in the upstream signal looses green time at a
  downstream signal, only to arrive there as the
  signal turns red. Hence, coordination is needed
  to establish efficient network-wide signal
  systems.
• Coordination attempts to achieve some combination
  of the following objectives
• Minimise fuel consumption
• Minimise pollution emission
• Minimise stops
• Minimise delay
• Maximise smooth flow
• Maximise capacity
• Minimise queue length
• Minimise arrival of platoons at red lights
• The signals less than 800m apart are coordinated
  as common practise.
• All but most complex coordination scheme requires
  same cycle length for all signals.

3
Factors influencing coordination

• Benefits
• The conservation of energy and the preservation
  of environment.
• Maintenance of preferred speed
• More cost-effective transport system
• Factors lessening benefits
• Inadequate roadway capacity
• Existence of parking, loading space etc.
• Complicated intersections, involving multiphase
  control
• Wide variability of traffic speeds
• Very short signal phasing
• Heavy turning volumes
• Exception
• Some intersections with heavy traffic volume
  might not be included in the network-wide
  coordination scheme and might be designed for a
  different (mostly longer) cycle length. This type
  of intersections need special consideration while
  modelling.

4
Terminology

• OFFSET
• Offset the difference between the green
  initiation times at two adjacent intersections.
•  
• Usually expressed as a positive number between
  zero and the cycle length. Sometimes convenient
  to think of it as a negative number, usually no
  more than one half a cycle length.
•  
• Ideal offsets
• The ideal offset is a value such that the first
  vehicle of a platoon just arrives at the
  downstream signal, the downstream signal turns
  green.
• t(ideal) L/S where, L block length, S
  vehicle speed
• BANDWIDTH
• The amount of green time used by a continuously
  moving platoon of vehicles through a group of
  intersections (time difference between the first
  and the last vehicle moving through the system
  without stopping)

5
Terminology

• TIME-SPACE DIAGRAM
• Plot of signal indications as a function
• of time
• Trajectory Path vehicle travels in time
• (X-axis)
• Signal Layout according to distance
• (Y-axis)

6
Signal Progression in 1-way street

• Assuming no vehicles are queued at the signals,
  the ideal offsets can be determined assuming a
  desired platoon speed (say,60 fps)

7
Signal Progression in 1-way street

• Time space diagram
• The vertical should be scaled so as to
  accommodate the dimensions of the arterial, and
  the horizontal so as to accommodate at least
  three to four cycle lengths.
• The beginning intersection should be scaled
  first, usually with MSG initiation at t 0,
  followed by green and red phase (yellow may be
  shown for precision). See Point 1.
• The MSG of the next downstream signal should be
  located next, relative to t 0. With this point
  located (Point 2), fill in the periods of green,
  yellow, and red for this signal.
• Repeat the procedure for all other intersections,
  working 1 at a time.

8
Signal Progression in 1-way street

• Effects of vehicles queued
• The lost time is counted only at the first
  downstream intersection, at most If the
  vehicle(s) from the preceding intersection were
  themselves stationary, their start-up causes a
  shift that automatically takes care of the
  start-up at later intersections.
• The ideal offset is calculated as,
• tideal L/S (Q.h Loss)
• where,
• Q veh queued per lane,
• h discharge headway ( 1.9 s),
• Loss loss time associated with vehicles
  starting from rest at the first downstream signal
  (2 s)

9

• The time-space diagram for the example, given
  queues of 2 vehicles per lane in all links. The
  arriving vehicle platoon has a smooth flow, and
  the lead vehicle has 60 fps travel speed. The
  visual image of the "green wave" or progression
  speed is much faster. The "green wave is
  travelling at varying speeds as it moves down the
  arterial.
• However, the window available for moving a
  platoon nonstop is much smaller.

10
A note on queue estimation

• It is a difficult and expensive task to
  estimate the queue size from cycle to cycle .
• Sources of queued vehicles may include
• Vehicles turning in from upstream side streets
  during their green (which is the main-street red)
  
• Vehicles leaving parking garages
• Part of a previous platoon truncated by
  insufficient green

11
Signal Progression in 2-way street

• Determining ideal offset on a 2-way street
• If any offset is changed to accommodate the
  southbound vehicle then the northbound movement
  will suffer.
• Hence, actual offsets in both directions are
  interrelated
• tNB,I tSB,I nC
• Actual offsets are ideal offsets some error
  term.

12
Signal Progression in 2-way street

1. Offsets in a two-way arterial are not
   independent.
2. A typical cycle yields the obvious conclusion
   that the offsets in two directions adds up to the
   cycle length.
3. However, for longer block lengths the offsets
   might add to two (or more) cycle lengths. Indeed,
   when queue clearances are taken into account, the
   offsets might add to zero cycle lengths.
4. In signal optimization programs the objective is
   to minimise the total discrepancies between
   actual and ideal offsets for all the designed
   links.

13
Signal Progression in Grid
The relative difficulty of finding progressions
on a two-way street, compared to on a one-way
street, might lead one to conclude that the best
approach is to establish a system of one-way
streets, to avoid the problem. A one-way street
system has a number of advantages, not the least
of which is elimination of right turns against
opposing traffic. However, the total elimination
of the constraints imposed by the "closure" of
loops within the network or grid is not
possible. In the figure If the cycle length,
splits, and three offsets are specified, the
offset in the fourth link-denoted Link D in this
Illustration is determined and cannot be
independently specified. In a grid of
multiple one-way streets, where the offsets in
all the NS links are independently specified,
then the specification of a single EW offset will
lock all other EW offsets.
14
Signal Progression in Grid
This extends to a grid of one-way streets, in
which all of the north-south streets are
independently specified. The specification of one
east-west street then "locks in" all other
east-west offsets. Note that the key feature is
that an open tree of one-way links can be
completely independently set, and that it is the
closing or "closure" of the open tree which
presents constraints on some of the links.
15
Signal Progression in Grid
16

• After reaching intersection 1, we are back at the
  starting point and now t0 or a multiple of cycle
  length. And we can write,
• The g values should include the change and
  clearance intervals.
• Fundamental results if you set the offsets in
  one direction on a 2-way street, then you also
  set them in the other direction. In a network,
  you can set any" open tree" of links, but links
  that close the tree already have their offsets
  specified".

17
Signal Progression in Grid
18
Signal Progression in Grid
While it is sometimes necessary to consider
networks in their entirety, it is common traffic
engineering practice to decompose networks into
non-interlocking arterials whenever possible.
19
Bandwidth

• This concept is very popular as the windows of
  green are easy visual image for professionals and
  public perception and good solutions can be often
  obtained manually.
• But, when substantial internal queues exist, this
  solution can be misleading and erroneous.
• Efficiency of a bandwidth is defined as
• Efficiency(seconds) (bandwidth/cycle length) X
  100
• An efficiency of 40-55 is considered as good.
  The bandwidth is limited by the min green in the
  direction of interest.
• Nonstop volume 3600 (BW) (L)/(h)(C) vph
• Where BW measured or computed bandwidth (s) L
  number of through lanes h headway
  (sec/veh), C cycle length (s)
• Generally bandwidth in the two opposing
  dierctions are designed to be in the same ratio
  as the flows in the two directions.

20
Bandwidth
21
Signal Progression