MASTERLINK Adaptive MODE

1
MASTERLINK (Adaptive) MODE

• Cycle time, stage splits and offsets are
  controlled by Regional Control Computer
• Traffic adaptive control providedIn Masterlink
  mode SCATS issues call commands to local
  controllers
• These commands terminate the running stage and
  define the next stage that is to run

2
SIGNAL CONTROL(Masterlink)

• However, SCATS cannot force a termination of a
  stage if the controller has any safety intervals
  active such as
• stage minimums
• signal group minimums
• pedestrian Walks and Clearances

3
SCATS operation

• Based on Degree of Saturation calculation
• Junction optimisation and linking
• Loop per strategic lane
• Stop line loop for accurate determination of flow
  during green time
• Self calibrating
• Concept of space time
• Gap headway waste
• Principle of efficient use of road space.

4
How to Control Traffic ?
Detectors
At critical (major) intersections, detectors
are required in all lanes. Usually located within
5 m of the stop line
5
4
6
7
8
3
2
1
9
10
5
SCATS FUNCTIONS

• SCATS has four main functions

• SIGNAL CONTROL
• MONITORING
• DATA COLLECTION
• STRATEGIC MANAGEMENT

6
RELATIONSHIP of DELAY to CYCLE TIME

• For a single signalised intersection -

Delay
Rapid increase in delay. Very long queues develop.
Cycle Time
Co
7
SUBSYTEM COMMON ELEMENTS

• Split, cycle and offset are selected on a
  subsystem basis.
• All intersections in a subsystem operate on the
  same
• Split Plan Number
• Cycle Time
• Offset Plan Number

8
How to Control Traffic ?
Detectors
At minor intersections, detectors are not
essential but it is useful to have detection on
the side road. Also SCATS backup (queue)
detectors are often useful on the approaches to a
major intersection
Backup detectors
2
1
9
9
SCATS Detectors

• Degree of Saturation (DS) in SCATS calculations
  rely on a 2.5 to 4.0 meter long sensor zone
• Can be loop or video
• Place sensors in all important approach lanes at
  or near the stop line.

10
SCATS DS Usage

• Possible split percentages examined each cycle
  to determine the most equisat timings for the
  next cycle, i.e.minimal delay.
• Equisat DS on critical approaches equal,.
• Maximum projected DS for each possible
  combination calculated (using last cycle DS
  values).
• Plan with the lowest maximum selected. Projected
  DS DS (old split/new split)
• DS also can increase/decrease cycle length.

11
Best Measure

• The measure used in SCATS is Degree of
  Saturation(DS)
• Volume alone is misleading as it takes no account
  of road capacity.
• Another term for Degree of Saturation is traffic
  density
• Degree of Saturation is usually expressed in
  percent ()whereas traffic density tends to be
  referred to as vehicles per lane per kilometre
  which takes no account of capacity.

12
In the most simplistic terms, the Degree of
Saturation of a section of road can be expressed
as follows

• x q / Q
• where x - Degree of Saturation
• q - actual flow rate ( in any appropriate units)
  (usually vehicles per hour)
• Q - maximum achievable flow

13
Flow at Signals

• Each lane of a road has a fairly constant
  maximum flow capability, i.e. there is a
  relatively predicable "saturation flow" level at
  which maximum throughput occurs. At traffic
  signals, this saturation flow can obviously
  only occur while the signal facing the particular
  approach is green.
• The fraction of the time that vehicles can flow
  is given by the ratio g/C
• where g green time for an approach
• C cycle length

14
Flow at Signals

• Therefore the capacity Q of an approach is
• Q sg/C
• Where ssat flow
• Substituting this equation for Q above gives a DS
  at traffic signals
• x qC/sg

15
Remember
DS qC/sg

• The saturation flow value cannot normally be
  altered by a traffic signal engineer. It is an
  intrinsic characteristic determined by roadway
  width, grade and other environmental factors.

16
Measurement of Flow

• An approach to signals has a stream of 8 cars in
  60 secs
• So flow, q 8 veh / min 480 v p h
• If cycle length, C 90 secs
• green time, g 25 secs
• Saturation flow, s 1800 vph
• Then DS q C 480 90 0.96 96
• s g 1800 25

17
Measurement of Flow

• An approach to signals has a stream of 6 vehicles
  in 60 secs
• Now flow, q 6 veh / min 360 v p h
• If cycle length, C 90 secs
• green time, g 25 secs
• Saturation flow, s 1800 vph
• Then DS q C 360 90 0.72 72
• s g 1800 25

18

• In the last example, the Traffic Density (Degree
  of Saturation) is still quite high (100) but
  our calculation gives only a reading of 72.
• It is apparent that the simple equation
• is NOT suitable for real-time variable mix
  traffic flows.

DS qC/sg
19
Some Flow Relationships

• If we plot Traffic Density v. Flow we get this
  relationship as depicted by the graph gtgtgtgtgtgtgtgtgt
• Note how the curve doubles back. For any value
  of flow there are two values of Density

20
Flow Relationships

• Once again flow is a very poor measure of traffic
  density. A flow of 600 vph could mean light flow
  or oversaturated flow (i.e. crawling traffic)

21
VEHICLE SPACE

• The relationship between traffic density and
  vehicle space is
• Almost linear
• Largely insensitive to vehicle mix (type and
  length)

DENSITY
SPACE (Distance)
22
VEHICLE SPACE

• However it is difficult to measure space distance
  in real time. It is much easy (with a loop
  detector to measure SPACE TIME (i.e. the time
  between vehicles OR the detector OFF time)

Loop
Space Time (secs)
23
VEHICLE SPACE

• However, the actual space time measured depends
  on the detection zone (loop) length.
• Note that short loops tail back (at high DS they
  can have long space times). Long detectors go
  blind (i.e. zero space values even at moderate
  densities.
• A detector size of 4.5m is normally used .

24
VEHICLE SPACE
DENSITY

• Some important features of this relationship
  between DS and Space.
• Typically a normal lane has a SATFLOW of 1800 vph
  and an optimum space time of 1.0 secs.

150 100 50
Oversaturated
Maximum flow (SATFLOW)
Undersaturated
4.5 m
1.0 2.0 3.0 4.0 5.0 secs
SPACE TIME
25
OPTIMUM SPACE

• The average space between vehicles during a
  period when maximum flow (SATFLOW) is achieved is
  called the Optimum Space Time and is typically 1
  sec for a wide range of local conditions, vehicle
  mixes and lane geometry.
• SCATS continually (every cycle) measures the flow
  and space time in each specified lane and stores
  the maximum flow and its associated optimum space
  time for each 24 hr period.

26
WASTE TIME

• All drivers leave some space between them and the
  vehicle in front - even in bumper-to-bumper
  conditions.
• In undersaturated conditions drivers (on average)
  will leave a greater space - s than the Optimum
  Space time - t between them and the vehicle in
  front.
• If we call this difference an average WASTE time.
  Then,
• wav s - t

27
WASTE

• The total WASTE for the lane is therefore
• Wt (s - t) n
• where n number of vehicles passing
• and t space time at maximum or optimal flow
• A SCATS controller actually measures the total
  Space Time ( T ) where T s n
• So the total Waste Time is now
• Wt T - tn

28
Degree of Saturation Definition

• In SCATS we define Degree of Saturation as
• The ratio of the NON-WASTED time to the TOTAL
  time available for a lane.
• If there is no WASTED time (i.e. maximum
  efficiency), then the lane must be running at
  SATFLOW

29
DS

• The NON-WASTED time is simply the total green
  time available - g less the actual WASTED time
• U (g - Wt)
• Therefore
• DS U / g
• ( g - Wt) / g

30
DS

• But remember
• So
• DS g - (T - tn) / g
• This formula drives SCATS

Wt T - tn
31
DS
DS g - (T - tn) / g

• Just check this formula
• t - Optimum space time, is a constant (for
  each 24 hr period)
• If T total space time increases, DS decreases
• If T (t n), then DS g / g 1.0 100
• If T gt (t n), then DS lt 100, i.e.
  undersaturated
• If T lt (t n), then DS gt 100, i.e.
  oversaturated

32
DS

• Actual DS formula in SCATS 6.4

DS NFg-(T-t.n)/(gr) NF Bias factor
nnumber of spaces counted 1 T Total non
occupancy t standard space time _at_ max flow
gphase time r remaining phase time
33
Two Methods Of Selection

• SCATS can select the best set of splits for each
  subsystem by two different methodsDISCRETE
  SPLIT SELECTION
• INCREMENTAL SPLIT SELECTION

34
Discrete Split Selection

• In DSS operation, SCATS automatically selects
  cycle by cycle, the most appropriate plan from a
  table (suite) of plans pre-entered by a SCATS
  traffic engineer.

35
Discrete Split Selection
Under DSS operation, SCATS selects the most
appropriate split plan for the current traffic
conditions from an suite of plans that as been
previously entered. A example of a suite of 4
plans
36
Discrete Split Selection

• How does SCATS select the best split plan from
  this suite of plans?
• By looking at the performance of the currently
  selected split plan compared with the expected
  performance of all other plans in the suite.

37
Discrete Split Selection
A B C

• Assume that SCATS measured the DS for the busiest
  lane on each approach while running SP 2 as
  followsA stage DS 80
• B stage DS 70
• C stage 60

38
DSS

• Remember - other factors
• remaining constant, then DS is inversely
• proportional to the split of a stage.
• So, the projected DSs for each stage if split
  plan 1 is selected , will be

A stg DS 80 B stg DS 70 C stg DS 60 with
SP 2 running
39
DSS
Completing the projected DSs for the other split
plans - Which is the best plan to
select?
40
DSS

• SCATS selects the split plan with the lowest
  maximum DS.
• In the previous example, SCATS places a vote
  for split plan 3.
• (The maximum projected DS in SP3 was 73)
• Two votes for the same plan in the last 3 cycles
  will cause a plan change to that plan ( unless
  otherwise set)