Now-a-days, controlling traffic congestion relies on having an efficient and well-managed traffic signal control policy. Traffic signals operate in either pre-timed or actuated mode or some combination of the two. Pre-timed control consists of a series of intervals that are fixed in duration. They repeat a preset constant cycle. In contrast to pre-timed signals, actuated signals have the capability to respond to the presence of vehicles or pedestrians at the intersection. Actuated control consists of intervals that are called and extended in response to vehicle detectors. The controllers are capable of not only varying the cycle length & green times in response to detector actuation, but of altering the order and sequence of phases. Adaptive or area traffic control systems (ATCS) belong to the latest generation of signalized intersection control. ATCS continuously detect vehicular traffic volume, compute optimal signal timings based on this detected volume and simultaneously implement them. Reacting to these volume variations generally results in reduced delays, shorter queues and decreased travel times. Coordinating traffic signals along a single route so that vehicles get progressive green signal at each junction is another important aspect of ATCS. In the subsequent pages, the operating principles and features of Vehicle-Actuated Signals & Area Traffic Control Systems will be briefly discussed.
As stated earlier, Vehicle-Actuated Signals require actuation by a vehicle on one or more approaches in order for certain phases or traffic movements to be serviced. They are equipped with detectors and the necessary control logic to respond to the demands placed on them. Vehicle-actuated control uses information on current demands and operations, obtained from detectors within the intersection, to alter one or more aspects of the signal timing on a cycle-by-cycle basis. Timing of the signals is controlled by traffic demand. Actuated controllers may be programmed to accommodate:
Such variability allows the signal to allocate green time based on current demands and operations. A proper clearance interval between the green & the red phases is also ensured.
The various advantages of actuated signals are stated below:
The main disadvantages are as following :
There are three basic types of actuated control, each using signal controllers that are somewhat different in their design:
This type of controller is used at intersections where a major street having relatively uniform flow is crossed by a minor street with low volumes. Detectors are placed only on the minor street. The green is on the major street at all times unless a call on the side street is noted. The number and duration of side-street green is limited by the signal timing and can be restricted to times that do not interfere with progressive signal-timing patterns along the major street.
This type of controller is used at the intersections of streets or roads with relatively equal volumes, but where the traffic distribution is varying. In full actuated operation, all lanes of all approaches are monitored by detectors. The phase sequence, green allocations, and cycle length are all subjected to variation. This form of control is effective for both two-phase and multi-phase operations and can accommodate optional phases.
Volume-density control is basically the same as full actuated control with additional demand-responsive features. It is designed for intersections of major traffic flows having considerable unpredictable fluctuations.
The various types of detectors used for detection of vehicles are as following:
The vast majority of actuated signal installations use inductive loops for detection purpose. Now, the type of detection is of greater importance than the specific detection device(s) used. There are two types of detection that influence the design and timing of actuated controllers:
Regardless of the controller type, virtually all actuated controllers offer the same basic functions, although the methodology for implementing them may vary by type and manufacturer. For each actuated phase, the following basic features must be set on the controller:
Each actuated phase has a minimum green time, which serves as the smallest amount of green time that may be allocated to a phase when it is initiated. Minimum green times must be set for each phase in an actuated signalization, including the non-actuated phase of a semi-actuated controller. The minimum green timing on an actuated phase is based on the type and location of detectors.
![]() | (1) |
where, Gmin = minimum green time in second, tL = assumed start-up lost time = 4 sec, h = assumed saturation headway = 2 sec, d = distance between detector & stop line in m and x = assumed distance between stored vehicles = 6 m.
![]() | (2) |
where, tL = start-up lost time (sec) and n = number of vehicles stored in the detection area.
This time actually serves three different purposes:
In terms of signal operation, it serves as both the minimum allowable gap to retain a green signal and as the amount of green time added when an additional actuation is detected within the minimum allowable gap. The unit extension is selected with two criteria in mind:
The Traffic Detector Handbook recommends that a unit extension of 3.0 s be used where approach speeds are equal to or less than 30 mile per hour, and that 3.5 s be used at higher approach speeds. For all types of controllers, however, the unit extension must be equal to or more than the passage time.
It allows a vehicle to travel from the detector to the stop line. It is analogous with ’Unit Extension’.
![]() | (3) |
where, P = passage time, sec, d = distance from detector to stop line, meter and S = approach speed of vehicles, m/s.
Each phase has a maximum green time that limits the length of a green phase, even if there are continued actuation that would normally retain the green. The maximum green time begins when there is a call (or detector actuation) on a competing phase. The estimation can be done by any of the following methods:
![]() | (4) |
where, Ci = Initial cycle length, sec, L = Total lost time, sec and V C = Sum of critical lane volumes, veh/hr. Knowing the initial cycle length, green times are then determined as:
![]() | (5) |
where gi = effective green time for Phase i, sec and V Ci = critical lane volume for Phase i, veh/hr. The effective green times thus obtained are then multiplied by 1.25 or 1.50 to determine the maximum green time.
The basic principle underlying all signal timing analysis is the queue accumulation polygon (QAP), which plots the number of vehicles queued at the stop line over the duration of the cycle. The QAP for a simple protected movement is illustrated in the Fig. 1.
From Fig. 1, it’s clear that queue accumulation takes place on the left side of the triangle (i.e., effective red) and the discharge takes place on the right side of the triangle (i.e., effective green).
There are two methods of determining the required green time given the length of the previous red time. The first employs a target v/c approach. Under this approach, the green-time requirement is determined by the slope of the line representing the target v/c of 0.9. If the phase ends when the queue has dissipated under these conditions, the target v/c will be achieved. The second method recognizes the way a traffic-actuated controller really works. It does not deal explicitly with v/c ratios; in fact, it has no way of determining the v/c ratio. Instead it terminates each phase when a gap of a particular length is encountered at the detector. Good practice dictates that the gap threshold must be longer than the gap that would be encountered when the queue is being served. Assuming that gaps large enough to terminate the phase can only occur after the queue service interval (based on v/c = 1.0), the average green time may be estimated as the sum of the queue service time and the phase extension time. Therefore, average green time = Queue Service Time + Phase Extension Time. Now,
![]() | (6) |
where, qr = red arrival rate (veh/s), qg = green arrival rate (veh/s), r = effective red time (s), s = saturation flow rate (veh/s) and fq = calibration factor = 1.08 - 0.1(actual green time∕maximum green time)2
![]() | (7) |
where, q = vehicle arrival rate throughout cycle (veh/s), u = unit extension time setting (s), t = time during which detector is occupied by a passing vehicle(s) = [3.6(Ld + Lv)]∕SA, Lv = Vehicle length, assumed to be 5.5 m, Ld = Detector length (m), SA= Vehicle approach speed (kmph), Δ = minimum arrival (intra-bunch) headway (s), λ = a parameter (veh/s) = Φq∕(1 - Δq), Φ = proportion of free (unbunched) vehicles in traffic stream = exp(-bΔq) and b = bunching factor.
This green-time estimation model is not difficult to implement, but it does not lead directly to the determination of an average cycle length or green time because the green time required for each phase is dependent on the green time required by the other phases. Thus, a circular dependency is established that requires an iterative process to solve. With each iteration, the green time required by each phase, given the green times required by the other phases, can be determined. The logical starting point for the iterative process involves the minimum times specified for each phase. If these times turn out to be adequate for all phases, the cycle length will simply be the sum of the minimum phase times for the critical phases. If a particular phase demands more than its minimum time, more time should be given to that phase. Thus, a longer red time must be imposed on all of the other phases. This, in turn, will increase the green time required for the subject phase.
Case | Δ(s) | b |
Single Lane | 1.5 | 0.6 |
Multi-lane | ||
2 lanes | 0.5 | 0.5 |
3 lanes | 0.5 | 0.8 |
Each actuated phase has a recall switch. The recall switches determine what happens to the signal when there is no demand. Normally, one recall switch is placed in the on position, while all others are turned off. In this case, when there is no demand present, the green returns to the phase with its recall switch on. If no recall switch is in the on position, the green remains on the phase that had the last ”call.”demand exists, one phase continues to move to the next at the expiration of the minimum green.
Yellow and all-red intervals provide for safe transition from green to red. They are fixed times and are not subject to variation, even in an actuated controller. They are found in the same manner as for pre-timed signals.
![]() | (8) |
![]() | (9) |
where, y = yellow time, sec, ar = all red interval, sec, S85 = 85th percentile speed, m/s, S15 = 15th percentile speed, m/s, t = reaction time of the driver = 1 sec (standard), a = deceleration rate = 3 m/s2 (standard), g = grade of approach in decimal, w = width of street being crossed, m and l = length of a vehicle, m.
The Fig. 2 illustrates the operation of an actuated phase based on the three critical settings: minimum green, maximum green, and unit or vehicle extension. When the green is initiated for a phase, it will be at least as long as the minimum green period. The controller divides the minimum green into an initial portion and a portion equal to one unit extension. If an additional call is received during the initial portion of the minimum green, no time is added to the phase, as there is sufficient time within the minimum green to cross the STOP line (yellow and all-red intervals take care of clearing the intersection). If a call is received during the last U seconds (Unit Extension) of the minimum green, U seconds of green are added to the phase. Thereafter, every time an additional call is received during a unit extension of U seconds, an additional period of U seconds is added to the green. Note that the additional periods of U seconds are added from the time of the actuation or call. They are not added to the end of the previous unit extension, as this would accumulate unused green times within each unit extension and include them in the total green period. The green is terminated in one of two ways:
The maximum green begins timing out when a call on a competing phase is noted. During the most congested periods of flow, however, it may be assumed that demand exists more or less continuously on all phases. The maximum green, therefore, begins timing out at the beginning of the green period in such a situation. Now-a-days, in India, detectors are placed mostly at stop lines. In that case, the green times for phases are primarily determined by arrival headway. The green time is extended until the gap between two vehicles becomes equal to or greater than the pre-determined threshold value. Generally threshold of 4 seconds is considered.
Approach | Detector Set-Back | Mimi-mum | Passage |
Speed | (To front of loop) | Green | Time |
(kmph) | (m) | (sec) | (sec) |
24 | 12 | 8.0 | 3.0 |
32 | 18 | 10.0 | 3.0 |
40 | 24 | 12.0 | 3.0 |
48 | 30 | 14.0 | 3.5 |
56 | 41 | 18.0 | 3.5 |
64 | 52 | 22.0 | 3.5 |
72+ | Volume density or multiple detectors recommended
| ||
Volume-Density Controllers are designed for intersections of major traffic flows having considerable unpredictable fluctuations. They are generally used at intersections with high approach speeds (≥ 45 mi/hr). Here, detectors are placed on all approaches. Generally this type of controller is used with Area Detectors. To operate efficiently, this type of control needs to receive traffic information early enough to react to existing conditions. So, it is essential that detectors be placed far in advance of the intersection.
An isolated suburban intersection of two major arterial is to be signalized using a full actuated controller. Area detection is to be used, and there are no driveways or other potential entry points for vehicles within 90 m of the STOP line on all approaches. The intersection is shown in the figure and all volumes have already been converted to tvus for convenience. Left-turn slots of 75 m in length are provided for each approach. The tvu conversions assume that a protected left-turn phase will be provided for all approaches.
Solution: Step 1: Phasing: The problem statement indicates that protected left-turn phasing will be implemented on all approaches. Note that Kennedy Avenue has double left-turn lanes in each direction and that Monroe Street has a single left-turn lane in each direction. At a heavily utilized intersection such as this, quad-eight phasing would be desirable. Each street would have an exclusive LT phase followed by a leading green in the direction of heavier LT flow and a TH/RT phase. Such phasing provides much flexibility in that LT phasing is always optional and can be skipped in any cycle in which no LT demand is noted. The resulting signalization has a maximum of four phases in any given cycle and a minimum of two. It is treated as a four-phase signal, as this option leads to the maximum lost times. Quad-eight phasing involves overlaps that would be taken into account if this were a pre-timed signal. As an actuated signal, the worst-case cycle, however, would occur when there are no overlap periods. This would occur when the LT flow in opposing directions are equal. Thus, the signal timing will be considered as if this were a simple four-phase operation without overlaps. The controller, however, will allow one protected LT to be terminated before the opposing protected LT, creating a leading green phase. The four phases are:
Step 2: Unit Extension: For approach speeds of 64 kmph, the recommended unit extension (from Table) is 3.5 s.
Step 3: Minimum Green Times and Detector Placement: The problem specifies that area detection shall be employed. For area detection, the far end of the detection zone is placed such that the passage time is equal to unit extension. Since all approaches (including LT approaches) have a 64 kmph approach speed, the far end of detectors should be located as follows:
Step 4: Critical-Lane Volumes: As the volumes given have already been converted to tvus, critical-lane volumes for each phase are easily identified:
Therefore, VC = (200+400+110+700) = 1,060 tvu/h.
Step 5: Yellow & All-Red times With a 64 kmph average approach speed for all movements, the S85 may be estimated as (64 + 8) = 72 kmph, and the S15 may be estimated as (64 - 8) = 56 kmph. Then:
Step 6: Maximum Green Times and the Critical Cycle: The initial cycle length for determining maximum green time is: Ci = 25.6/[1-1060/(1615*0.96*0.98)] = 84.8 sec. Green times are found as:
Consider an intersection of two streets with a single lane in each direction. Each approach has identical characteristics and carries 675 veh/h with no left or right turns. The average headway is 2.0 s per vehicle and the lost time per phase is 3.0 s. Detectors are 9.1 m long with no setback from the stop line. The actuated controller settings are as follows:
Setting | Time (s) |
Initial interval | 10 |
Unit extension | 3 |
Maximum green | 46 |
Intergreen | 4 |
Determine the phase time for this intersection with actuated controller for approach speed 50 kmph.
Solution: The maximum phase time for each phase will be (46 + 4) = 50 s. The minimum phase time will be 10 + 3 + 4 = 17 s. The first iteration will be used with a 34-s cycle with 17 s of green time on each approach. The effective green time will be 14 s, and the effective red time will be 20 s for each phase. For purposes of traffic-actuated timing estimation It is recommended (HCM 2000) that, for a specified lost time of n seconds, 1 s be assigned to the end of the phase and (n - 1) s be assigned to the beginning. Here, start-up lost time = 2.0 secs. The following are the steps to calculate the phase time required:
Step 1. Compute the arrival rate throughout the cycle, q: q = 675/3600 = 0.188 veh/s
Step 2. Compute the net departure rate (saturation flow rate - arrival rate): (s - q) =1800/3600- 0.188 = 0.312 veh/s
Step 3.Compute the queue at the end of 20 s of effective red time: qrr = 20 × (0.188) = 3.760veh
Step 4. Compute the queue calibration factor,fq: fq = 1.08 - 0.1(13∕46)2 = 1.072
Step 5. Compute the time required to serve the queue, gs: gs = 1.072(3.760∕0.312) = 12.919s
Step 6. Determine λ: Δ = 1.5 and b = 0.6 (for single lane from table in HCM)
Step 7. Determine the occupancy time of the detector: t0 = 3.6(9.1+ 5.5)/50, vehicle length=5.5m, detector = 1.051 s length=9.1 m, approach speed=50 kmph
Step 8. The expected green extension time, ge:
Step 9. Compute the total phase time:
Step 10. Compute the phase time deficiency as the difference between the trial phase time and the computed phase time: or 25.469 - 17.0 = 8.469 s. For next iteration: Trial green time = 25.469 s. Cycle length = 50.968 s. This process is continued through successive iterations until the solutions converge or the phase deficiency i.e. the error is negligible practically. The following figure shows the results of successive iterations for this problem and the final convergence.
The final phase time is 37.710 s giving a cycle length of 75.420 s. The convergence was considered for threshold of 0.1 difference in successive cycle times.
Modern actuated controllers give the traffic engineers a great deal of flexibility in dealing with variations in demand. Area traffic control system along with Vehicle actuated signals can reduce traffic delays substantially. These are highly complex subject. Timing of VA signals is almost as much an art as a science, and more then one solution is possible. Regarding ATC systems, SCOOT and SCAT are popular in advance countries but such systems cannot cope up with Indian situations without adaptation to Indian traffic scenario. Presently, an advance ATC system known as CoSiCoSt has been developed considering the Indian Traffic scenario.
I wish to thank several of my students and staff of NPTEL for their contribution in this lecture. I wish to thank specially my student Mr. Kaniska Ghosh for his assistance in developing the lecture note, and my staff Mr. Rayan in typesetting the materials. I also appreciate your constructive feedback which may be sent to tvm@civil.iitb.ac.in
Prof. Tom V. Mathew
Department of Civil Engineering
Indian Institute of Technology Bombay, India
_________________________________________________________________________
Thursday 28 September 2023 10:47:46 AM IST