See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/228942879

Evaluation of Effect of Traffic Signal Coordination System on Congestion

Conference Paper · July 2009

CITATIONS READS
3 3,960

3 authors:

Mahmood M.Nesheli Othman Che Puan

University of Toronto Universiti Teknologi Malaysia
23 PUBLICATIONS 317 CITATIONS 75 PUBLICATIONS 453 CITATIONS

SEE PROFILE SEE PROFILE

Arash M. Roshandeh

37 PUBLICATIONS 439 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

WEIGHTAGE FACTOR FOR MALAYSIA GREEN HIGHWAY ASSESSMENT View project

PTSF on Two-lane Highways View project

All content following this page was uploaded by Othman Che Puan on 30 December 2014.

The user has requested enhancement of the downloaded file.

 Mahmood Mahmoodi Nesheli, Othman Che Puan,
WSEAS TRANSACTIONS on ADVANCES in ENGINEERING EDUCATION Arash Moradkhani Roshandeh

Optimization of Traffic Signal Coordination System on

Congestion: A Case Study
MAHMOOD MAHMOODI NESHELI, OTHMAN CHE PUAN AND ARASH
MORADKHANI ROSHANDEH
Department of Geotechnics and Transportation
Universiti Teknologi of Malaysia, UTM
81310, Skudai, Johor Bahru
MALAYSIA
[email protected], [email protected], [email protected]

Abstract: Signal coordination is perceived by many agencies as an advantageous improvement to the community or
corridor in consideration. In many cases, signal coordination techniques have proven to be successful in improving the
quality of life and mobility through the area. This study determines the coordination system pattern of traffic signal for
four consecutive intersections spaced at 780 m distance. Data for vehicles movement were collected using video
camera during morning and evening peak hour with congested conditions. For evaluation of the possible coordination
of signalized intersections a simulation model, TRANSYT7F, was used. The results show after coordinating, the
amount of delay, travel time, and queue reduce.

Key-words: Congestion, Coordination System, Delay, Intelligent Transportation Systems, Maximum Back of Queue,
Travel Time

1 Introduction authors [3]. The coordinated actuated corridor or

Intelligent Transportation Systems (ITS), which apply network optimization became more necessary than
advanced technologies to surface transportation systems, before because more detectors are being installed on
are widely viewed as the solution to the transportation intersections. But there were not many software that are
problems that our society faces. ITS applications, in able to deal with actuated coordinated signals. Recent
which technology is used to increase the operating version of some software has improvement in actuated
efficiency and capacity of transportation infrastructure, signal simulation and optimization; therefore, it is
can supplement or even replace infrastructure possible to get more realistic simulation results on
development, providing more effective mobility actuated signal intersections.
solutions at less of a cost to society. Urban traffic At a higher level, however, traffic signals can part of
control is a major area in which ITS can be applied [1]. a broader control strategy. In this case, signal controllers
Traffic signal coordination is a method of timing are used as tools for managing traffic flow, either along
groups of traffic signals along a major roadway to a corridor or throughout a network, to provide a more
provide for a smooth flow of traffic with minimal stops. efficient use of the urban street network. ITS
The goal of coordination is to get the greatest number of applications for transit, or Advanced Public
vehicles through a system a group of coordinated traffic Transportation Systems (APTS), have the same goals,
signals with the fewest number of stops. While it would namely improvements of efficiency without the need for
be ideal if every vehicle entering the system could major infrastructure enhancements [4]. One such
proceed through without stopping, this is not possible application is Bus Rapid Transit (BRT), a transit
even in a well-spaced, well-designed system. concept that uses buses to provide a high level of service
Coordinated traffic signals also result in less stop-and- usually associated with rail transit. The reason that rail
go traffic. This can reduce driver frustration and stress transit can provide such a high level of service,
levels, and may reduce a driver’s potential to take risks however, is that it operates on a right-of-way that is
on the road [2]. fixed and exclusive. This is typically not the case for
Improvement of traffic signal timing and using city buses, which instead operate on a shared right-of-
traffic signal coordination is one of the most important way in an open and more chaotic system. In such an
strategies for reducing delay, travel time and queue environment, buses face delays caused by interactions
length in urban area. The comparison of corridor or with other vehicles and by the presence of traffic signals
network optimization for pre-timed signal system using at intersections. These two factors can have a significant
different software was done many times by many negative impact on operations [5].

ISSN: 1790-1979 203 Issue 7, Volume 6, July 2009

 Mahmood Mahmoodi Nesheli, Othman Che Puan,
WSEAS TRANSACTIONS on ADVANCES in ENGINEERING EDUCATION Arash Moradkhani Roshandeh

One method of addressing these operational representation of phase state on a two-axis plane of
challenges is by the use of infrastructure solutions such distance versus time clearly shows a "green band" that
as exclusive bus lanes. While often effective in reducing has been established based on signalized intersection
delays due to congestion, these solutions can be spacing and expected vehicle speeds. In some countries
prohibitively expensive or, in many urban areas, (e.g. Germany, France and The Netherlands), this "green
infeasible due to inadequate street space. Another band" system is used to limit speeds in certain areas.
method is the use of control strategies, which use the Lights are timed in such a way that motorists can drive
existing traffic signal control system to give priority to through without stopping if their speed is lower than a
transit vehicles. This convergence of APTS and urban given limit, mostly 50 km/h in urban areas [6].
traffic control is known as transit signal priority. Transit Such systems were commonly used in urban areas of
signal priority strategies can be categorized into two the United States since the 1940s, but are less common
basic types: passive and active. Passive priority today. In modern coordinated signal systems, it is
strategies are those that use static signal settings to favor possible for drivers to travel long distances without
streets with transit routes. These rely on signal timing encountering a red light. This coordination is done easily
plans that are prepared offline and are designed to only on one-way streets with fairly constant levels of
impede transit vehicles as little as possible. Active traffic. Two-way streets are often arranged to
priority measures are those which employ dynamic correspond with rush hours to speed the heavier volume
detection and response to transit vehicles, altering signal direction. On the other hand, some traffic signals are
settings in real-time in order to reduce delay. coordinated to prevent drivers from encountering a long
Implementing transit signal priority can offer many string of green lights. This practice discourages high
challenges. One major concern is how to implement volumes of traffic by inducing delay yet preventing
transit priority within the existing signal control system. congestion. Speed is self-regulated in coordinated signal
Another is determining what impacts the priority systems; drivers travelling too fast will arrive on a red
implementation will have on other traffic. Most indication and end up stopping, drivers travelling too
fundamental, however, is the question of what benefits slowly will not arrive at the next signal in time to utilize
the priority implementation offers and whether these the green indication [7].
benefits outweigh the costs. In these cases, simulation
can be used to evaluate a proposed strategy before it is
implemented determining whether field implementation 2.1 Principle of traffic signal coordination
will have beneficial results. Traffic simulation is an An offset is defined as the time difference in the
ideal tool for these evaluations, as it simulates vehicle beginning of green between adjacent traffic control
movements at a detailed level, modeling interactions signals and is expressed in seconds. Traffic signal
with other vehicles and response to traffic control coordination is a method of establishing relationships
devices. between adjacent traffic control signals using offsets.
Traffic signal coordination reduces delay and
unnecessary stops at traffic signals. The benefit of traffic
2 Literature Review signal coordination is based on the relationship between
In general, the inflow on any intersection in an arterial the prevailing speed of vehicles on the main street, the
road is limited by the capacities and green times of the spacing of/distance between traffic signals, the hourly
different links on the upstream intersection. In order to traffic volume on a major street, hourly traffic volumes
maximize the possibility that vehicles passing an on the side streets, and number of non-signalized
intersection upstream will arrive at each intersection intersections along the roadway system [8].
downstream during the green phase, the traffic signals Travel speed along a roadway system is dependent
within an arterial road can be coordinated. on the signal spacing and the cycle length at traffic
Attempts are often made to place traffic signals on a signals. Travel speeds are lower when traffic signals are
coordinated system so that drivers encounter long strings closely spaced and operate under a short cycle length.
of green lights. The distinction between coordinated Conversely, higher travel speeds are a result of long
signals and synchronized signals is very important. cycle lengths and large spacing between intersections.
Synchronized signals all change at the same time and are Traffic signal coordination can be achieved at short
only used in special instances or in older systems. signal spacing, such as at 0.25 mile, as long as the traffic
Coordinated systems are controlled from a master volumes are low and short cycles 70 second or less can
controller and are set up so lights "cascade" in sequence be used. As arterial and cross-street traffic volumes
so platoons of vehicles can proceed through a increase, longer cycle lengths must be used in order to
continuous series of green lights. A graphical increase capacity by minimizing lost time. As a result,

ISSN: 1790-1979 204 Issue 7, Volume 6, July 2009

 Mahmood Mahmoodi Nesheli, Othman Che Puan,
WSEAS TRANSACTIONS on ADVANCES in ENGINEERING EDUCATION Arash Moradkhani Roshandeh

cycle lengths of 90 to 120 seconds are commonly used to a central computers. The traffic flow data is gathered
in those areas. A spacing of 0.5 miles will enable traffic using a detector located in each lane at the signalized
flow at a wide range of speeds, with cycle lengths intersection.
ranging from 60 to 120 seconds [6].

2.4 Advantages and Disadvantages of Traffic

2.2 Need for Traffic Signal Coordination Signal Coordination
Traffic signal coordination is typically needed to process Signal coordination is perceived by many agencies as an
traffic efficiently through a group of intersections. This advantageous improvement to the community or
is an attempt to utilize the existing roadway corridor in consideration. In many cases, signal
infrastructure by insuring optimum travel speeds while coordination techniques have proven to be successful in
reducing delay. Traffic coordination may delay or even improving the quality of life and mobility through the
eliminate the need for roadway widening. Since traffic area. Project experience from around the United States
signal coordination attempts to reduce the number of has indicated that interconnecting previously un-
stops and slow down of traffic, there is a reduction in coordinated signals and providing newly optimized
accident potential. In addition to traffic and safety timing plans and a central master control system can
concerns, the need for signal coordination may be result in a reduction in travel time ranging from 10
justified by high levels of vehicle emissions and poor air percent to 20 percent [9].
quality.
An engineering study may be required to determine 2.4.1 Some of the Advantages of Traffic Signal
the need for traffic signal coordination. The need is Coordination
based on a detailed investigation of the existing Traffic signal coordination has some advantages that are
conditions which include travel speeds and delay, traffic dedicated as follow:
volumes and accident experience [9]. (i) Improves mobility and access through the area;
(ii) Reduces vehicle accidents in the area;
2.3 Types of Coordinated Signal Systems (iii) Reduces energy and fuel consumption;
A signal system can be defined as a group of traffic (iv) Reduces stops;
signals that are coordinated. The selection of a type of (v) May control travel speeds;
signal system is based upon the available budgetary (vi) Provides environmental benefits from reduced
resources and the applicability of that system in the vehicle emissions; and,
given area [10].
The most common signal systems are Urban Traffic (vii) Ability to monitor daily traffic operations
Control Systems (UTCS), Closed-Loop Systems, Time- (UTCS and Closed-Loop).
Based Coordination (TBC) Systems, and traffic adaptive
signal control systems. The TBC system operates on a 2.4.2 Some of the Disadvantages of Traffic Signal
time clock that is used to take actions automatically Coordination
based upon the time of day and day of week. In contrast, Beside of any advantage for using the one system or
both UTCS and the Closed-Loop systems react to real- method there are some disadvantages. Disadvantages of
world conditions as they are happening, based on actual coordination system are as follow.
traffic volume and signal timing data stored in the (i) Increase in travel speeds may have a negative
system. impact in the community;
In UTCS and Closed-Loop systems, traffic signals (ii) May attract additional traffic through the
are interconnected using different types of cables or corridor; Maintenance and equipment costs may
communication mechanisms. Electrical cables are the be high based on the type of hardware and
most commonly used method of signal system software used; and,
interconnection. Fiber-optic cables are slowly getting (iii) Requires qualified staff for maintenance and
recognition in signal systems. Connecting cables are not monitoring of daily operations.
needed in the TBC system, as adjacent intersections are
coordinated by the timing of their individual controlling
clocks. 3 Aim and Objectives
Traffic-adaptive signal control systems are designed The aim of this study is to implement a traffic control
to develop coordination patterns in real-time based on coordination system within a macroscopic simulation
traffic flow data gathered, processed, and communicated environment, thus there is a case study for the evaluation

ISSN: 1790-1979 205 Issue 7, Volume 6, July 2009

 Mahmood Mahmoodi Nesheli, Othman Che Puan,
WSEAS TRANSACTIONS on ADVANCES in ENGINEERING EDUCATION Arash Moradkhani Roshandeh

of traffic congestion. A simulation model, TRANSYT7F calculates the maximum back of queue.
TRANSYT7F, was used to evaluate the possible This calculation includes any vehicles which join to the
coordination of signalized intersections. A comparison back of the queue after the signal indication has turned
of the traffic parameters after modeling system is green and the front of the queue is moving. The queue
performed, and recommendations for improvement and lengths calculated are the queues that occur due to the
further study are offered. To achieve this aim, the study given flow rates and they do not build over time [12].
was carried out based on the following objectives: Therefore, it is assumed that the queues have cleared at
To evaluate travel time, speed, delay and maximum the end of each cycle and there are no residual vehicles.
back of queue in the case study, and Comparison of the TRANSYT-7F also reports a queue capacity value for
traffic parameters performance after modeling the each link based upon a user input value such as a left
system with software. turn storage length or it is calculated by TRANSYT-7F
based upon the link length. This value can then be
compared to the estimated queue lengths to determine
4 Methodology where spillover may occur. For measuring queue the
For implementing this study some data like traffic flow reference lane at each intersection was considered.
at each intersection during A.M. and P.M. peak hour, During the peak hour the number of vehicle in queue
maximum back of queue and control delay at each was measured.
approach and travel time between intersections were
collected. 4.1.2 Travel Time
The delay experienced by a motorist is made up of a
number of factors that relate to control, geometrics,
4.1 Field Data traffic, and congestion. Total delay is the difference
Field data were collected at four signalized intersections between the travel time actually experienced and the
that were located successively in the case study. At each reference travel time that would result during base
site the video cameras were set up. The total intersection conditions, in the absence of congestion, control, traffic,
traffic volumes were grouped into 15-minute time or geometric delay. To estimate the travel time, one
periods. technique, floating car method, was used. Some
In addition to the total intersection volumes and the checkpoints involved all intersections were selected to
maximum back of queue length data, the saturation flow measure the travel time. This measurement was done
rates for the critical approaches were measured. The during evening and morning peak hour.
Highway Capacity Manual method for measuring
saturation flow rates was used. Consistent with ideal 4.1.3 Delay
values in TRANSYT7F and Chapter 16 of the Highway The values derived from the delay calculations represent
Capacity Manual [11]. the average control delay experienced by all vehicles
These intersections are located in the vicinity of that arrive in the analysis period, including delays
Johor Bahru. Video cameras were set up at each of the incurred beyond the analysis period when the lane group
locations during the A.M. and P.M. peak hour period. is oversaturated. Delay that practically measured was
Data were collected for the heaviest traveled weekday according to HCM (2000) chapter 16, appendix A. The
peak hour conditions. Each of these locations were measuring procedure of delay in TRANSYT is
videotaped for a minimum one hour. On certain according to HCM too. The average control delay per
intersection approaches where the traffic volumes were vehicle is estimated for each lane group and aggregated
heavy and back of queues long, an additional dedicated for each approach and for the intersection as a whole.
camera was assigned to tape the approach. LOS is directly related to the control delay value. The
The traffic volume, delays, vehicle classifications, criteria are listed in HCM (2000).
and maximum back of queue were obtained for most
approach lane groups at each of the intersection
locations. These data were collected by cycle for 16 4.2 Case Study Description
cycles. Signal timing is not fixed over the study period Selection of the study area is based on road sections that
since some signals analyzed are actuated and some are are spaced less than 270 meters for two signalized
pretimed. Cycle lengths, signal phases and offsets were intersections. This study requires area where there is
measured before and after each analysis period and they straight path with series junctions. The Bukit Indah is
were unchanged. one of the most important areas in Skudai, Johor Bahru,
Malaysia. The network is extremely congested, and has
4.1.1 Queue heavy traffic, resulting from a varied mix of commuters
and travelers.

ISSN: 1790-1979 206 Issue 7, Volume 6, July 2009

 Mahmood Mahmoodi Nesheli, Othman Che Puan,
WSEAS TRANSACTIONS on ADVANCES in ENGINEERING EDUCATION Arash Moradkhani Roshandeh

4.3 Simulating Model local and global optimum solutions, refer to the
Using TRANSYT7F for evaluating traffic signal documentation on Hill- Climb Optimization.
network is one of the ways to achieve delay and queue While running G.A. optimization, the T7F10
in the system. In this study current traffic was evaluated interface continuously displays the optimal function
with simulating the system in the software, it means the value and the average function value. The optimal
base case was simulated according to the uncoordinated function value represents the best performance index
traffic signal. Next step of simulation was applied for value when considering all candidate timing plans in
coordination system. Also optimization and estimation that particular generation. The average function value is
the system to accomplish minimum delay were done. the average performance index value from all candidate
Modeling a network in software requires detailed input solutions in that particular generation. (TRANSYT 7F,
data. These data are included traffic flow in each lane 2005).
and movement, traffic signal data such as cycle time,
kind of control (actuated or pretimed) peak hourly factor
(PHF) for each lane, geometric data, saturation flow and 4.4 Optimization
etc. The primary qualities of the TRANSYT-7F optimization
process include the availability of multiple search
4.3.1 Disutility Index techniques (hill-climb and genetic algorithm), numerous
This is a measure of disadvantageous operation; that is, optimization objective functions (e.g., involving
stops, delay, fuel consumption, etc. Unless the disutility combinations of progression opportunities, delay, stops,
index (DI) has specifically been defined as excess fuel fuel consumption, throughput, and queuing), extensive
consumption, its value has no intrinsic meeting, since it ability to customize the optimization process, and the
is simply a linear combination of delay and stops, whose ability to optimize all signal settings (cycle length,
units differ. An excess maximum back of queue penalty phasing sequence, splits, and offsets). When using
can optionally be included within the disutility index. genetic algorithm optimization, it is now possible to
Note that "Disutility Index (DI)" was the optimization allow offset optimization at certain intersections but not
objective function for this run, and average delay is a others. In addition, genetic algorithm optimization now
significant component of the DI. holds the offset constant at the master controller, if a
In general, TRANSYT-7F always attempts to master controller has been defined [13].
maximize the PI (Performance Index), unless the Phase sequences may consist of numerous
disutility index (DI) has been selected as the PI. Since combinations of protected and permitted movements.
the DI can be a combination of delay, stops, queuing, While any phase sequence may be specified in
and fuel consumption, this value must be minimized. TRANSYT, the traffic engineer must exercise
Other objective functions that involve progression or professional judgment as to which sequences are most
throughput must be maximized. practical (or safe) for the intersections under
The DI is a combination of vehicle delay, stops, and consideration. Phasing sequence optimization is
fuel consumption. Weighting factors are available in primarily effective at improving progression for
order to place more emphasis on any of these three DI coordinated intersections [14].
components if desired. Delay-only optimization results TRANSYT-7F explicitly optimizes phasing
in excessive stops and fuel consumption. Excess fuel sequences, phase lengths (splits), and offsets for a given
consumption (minimization) is considered to be a good cycle length. Optimization consists of a series of trial
compromise between bandwidth and delay-based simulation runs, using the TRANSYT-7F simulation
optimization. Although the DI has been shown over the engine. Each simulation run is assigned a unique signal
years to be a practical and effective combination of timing plan by the optimization "umbrella", or
minimizing delay and stops, it does not necessarily processor. The optimizer typically applies the hill-climb
produce a wide bandwidth (TRANSYT7F, 2005). and/or genetic algorithm searching strategies. The trial
simulation run resulting in the best performance is
4.3.2 Genetic Algorithm Optimization reported as optimal. To determine the best cycle length,
Genetic algorithm (G.A.) optimization is a theoretical an evaluation of a user-specified range of cycle lengths
improvement over the traditional hill-climb optimization may also be made. Prior the optimization project, field
technique that has been employed by TRANSYT-7F for studies related to travel times as well as advance
many years. G.A. has the ability to avoid becoming considerations for a signal timing review process is
trapped in a "local optimum" solution, and is recommended[15].
mathematically best qualified to locate the "global
optimum" solution. For a discussion on the meaning of

ISSN: 1790-1979 207 Issue 7, Volume 6, July 2009

 Mahmood Mahmoodi Nesheli, Othman Che Puan,
WSEAS TRANSACTIONS on ADVANCES in ENGINEERING EDUCATION Arash Moradkhani Roshandeh

4.5 Estimation 5.1 Delay

The estimation module is capable of computing traffic- Delay that measured in the case study is control delay.
actuated phase times with precision included are an Control delay is the portion of the total delay attributed
improved green time reallocation algorithm, to traffic signal operation for signalized intersections.
comprehensive critical link search, and improved According to appendix A, HCM (2000), field
treatment of stochastic effects. The end result is that measurement of intersection control delay was obtained
performance estimates generated by TRANSYT-7F, at each approach of all intersections during morning and
after computing actuated phase times under the new evening peak hours. Therefore 32 samples from 16
methodology, are more realistic. The actuated estimation approaches control delay during peak period were
module can be used to determine maximum green times measured.
for implementation in the field [13].

5.2 Queue and Travel Time

4.6 Calibration of TRANSYT7F Traffic stream for the site in the morning and evening
The process of comparing model parameters with real- peak hour is different. Traffic flow in the morning from
world data is to ensure that the model realistically Johor Bahro to Jusco has maximum amount and in the
represents the traffic environment. The objective is to evening this flow is inverse. The value of queue in the
minimize the discrepancy between model results and morning from the Johor Bahru to Jusco and Jusco to
measurements or observations [11]. Before using the Johor Bahru are 19 vehs per lane, and 21 vehs per lane
model to simulate the traffic flow, it has to be calibrated, respectively at major routes. It depict high amount of
so that it gives a good estimate of the results. For this maximum back of queue in the system when there is
study, amount of delay is the most important measure of congestion. The results of travel time show there is
effectiveness, because it was found to be directly significant difference between running speed and
proportional to the traffic performance in the system. In journey speed (more than 50%). It shows there is long
order to estimate the traffic performance correctly, the delay in the case study.
model should be able to estimate the value of delay
correctly. In order to calibrate the value of delay, travel
time and maximum back of queue obtained from the 5.3 Calibration of TRANSYT7F
TRANSYT7F model is compared to the delay, travel Scatter plot to determine R- squared value for field data
time and maximum back of queue obtained from the and TRANSYT7F results were used. Correlation of
field studies. model with TRANSYT, based on all three experiments,
are illustrated in Fig. 1, Fig. 2 and Fig. 3. For travel time
just major approaches were considered, for control delay
5 Results and Analysis and maximum back of queue data of all approaches at
each intersection were considered. It seems that the
goodness of fit (R2) of the calibration was obtained.

Figure 1. Travel time calibration

ISSN: 1790-1979 208 Issue 7, Volume 6, July 2009

 Mahmood Mahmoodi Nesheli, Othman Che Puan,
WSEAS TRANSACTIONS on ADVANCES in ENGINEERING EDUCATION Arash Moradkhani Roshandeh

Figure 2. Delay calibration

5.4 Evaluation of Performance of System without cycle time to reduce delay and minimize it, was applied
Coordination in TRANSYT7F. This coordination provides
Objective of this part is modeling the base case information about the links that are affected by the
congestion. The results for the whole segment of system
condition to obtaine curent traffic performance. The first
show that the amount of delay, travel time and disutility
study was done without coordination in the system.
index (DI) reduced. The benefits of coordination are
Results show for the whole system, the high amount of
clearly visible. For the whole segments, the maximum
travel times, delay and maximum back of queue are
experienced during the peak period by most of the average travel time comes down to 314veh-veh/hr, from
396veh-veh/hr, an improvement of about 21% and
travelers. However, disutility Index (DI) shows the
travel time and delay are larger indicating that these 325veh-veh/hr from 419veh-veh/hr, an improvement
about 22% during the morning and evening peak hours
routes are highly congested. Delay obtained 65.1 sec/veh
respectively. From the results, we can see that control
(A.M.) and 68.3 sec/veh (P.M.), with refer to (HCM,
delay faced by the vehicles has come down significantly,
2000) dictate level of service (LOS) E. Also travel time
achieved 396 veh-hr/hr (A.M.) and 419 veh-hr/hr (P.M.). where 65.1sec/veh has been changed to 35.6sec/veh and
68.3 comes down to 37.2sec/veh during A.M. and P.M.
peak hours respectively that indicates LOS D. Average
travel times for the peak hour period decreases during
5.5 Evaluation of Performance of System with
the whole duration. With the help of coordination
Coordination system, maximum back of queue per link in major lane
Simulation of the system with coordination mode and also comes down by a significant value.
with the help of optimization (genetic algorithm) and
estimation of maximum green time for finding the best

Figure 3. Queue calibration

ISSN: 1790-1979 209 Issue 7, Volume 6, July 2009

 Mahmood Mahmoodi Nesheli, Othman Che Puan,
WSEAS TRANSACTIONS on ADVANCES in ENGINEERING EDUCATION Arash Moradkhani Roshandeh

Figure 4. Travel time vs number of approach

It leads to increase the system speed. Without
5.6 Comparison coordination system occurrence increases the delay as
This section compares the results obtained from different well. Coordination system usage can bring this down by
scenarios, travel time, control delay and the maximum about 45% during peak hours that means improve the
back of queue for each approach. level of service to D. Fig. 4 and Fig. 5 show the amount
of travel time and delay at both systems during the peak
5.6.1 Comparison Base on Travel Time and Delay hours at each approach. All obtained result will be
Table 1 summarizes the travel times, average delay, represented by performance index.
system speed and performance index during A.M. and
P.M. peak hours for whole segment of the route. As can Table 1. Summary of the Result of Uncoordinated and
be observed from these results, using a coordination Coordinated During the A.M and P.M Peak Hour.
system can significantly improve all parameters. It may
be observed that using coordination system can save
about 21% and 22% vehicle hours for travel time during
the peak hours in the morning and evening respectively.

Figure 5. Delay vs number of approach

ISSN: 1790-1979 210 Issue 7, Volume 6, July 2009

 Mahmood Mahmoodi Nesheli, Othman Che Puan,
WSEAS TRANSACTIONS on ADVANCES in ENGINEERING EDUCATION Arash Moradkhani Roshandeh

Figure 6. Queue vs number of approach

For the whole segment, the performance index (DI) signal timing, coordinating system leads to reduce
comes down to 186.4, from 233.7, an improvement of delay and reduction in the overall system number of
about 20% and 194.3 from 251.2, an improvement about stops.
23% during the morning and evening peak hours 4. Simulating system with TRANSYT7F shows when
respectively. intersections are uncoordinated, the amount of delay,
queue, travel time and system speed are almost near
to practical measuring.
5.6.2 Comparison Base on maximum Back of Queue 5. The value of delay, maximum back of queue, travel
(Veh/Link) time and speed that obtained from practical
The comparison based on the maximum back of queue measurement in the site show the level of service in
within various ranges of travel times and delay for whole the case study is E and drivers have experienced long
segment of route during peak hours at all approaches is delay, long travel time and low speed for passing the
shown in Fig. 6. The impact of the coordination system system.
is obvious from the fact that there is a significant 6. During the evening and morning peak hour after
reduction in the value of queue with major approaches. applying coordinating system obvious effect has been
seen, where average delay from 68.3 sec/veh
decreases to 37.2 sec/veh and from 65.1sec/veh
6 Conclusion comes down to 35.6sec/veh respectively. It means
1. Because of consecutive short distance signalized LOS has been improved from E in uncoordinated
intersections in the site, the performance of traffic system to D in coordinated one. This leads to increase
parameters are not appropriate, so one of the methods speed from 10.9 km/hr to 14.7 km/hr and 11.2 km/hr
for improving the network performance is to 15.1 km/hr during P.M. and A.M. peak hour
coordinating traffic signal at intersections. respectively.
2. Maximum traffic flow in the morning is from Also for the whole system, performance index (DI)
southbound and in the evening is from northbound. improved about 23% for evening and about 20% for
During the A.M. and P.M. peak hours drivers have morning peak hour.
experienced long queue in mentioned approaches. In 7. In a coordinated signal system the cycle length is
addition the amount of traffic flow in the evening peak constant for all controllers during any given control
hour is more than morning peak hour. period. In this study the capability of TRANSYT-7F
3. One of the most important measurements of (optimization and estimation model) has been used
effectiveness in traffic studies is the delay to vehicles for evaluating a range of cycle lengths and advising
in the system. Delay represents indirect costs to the of a "best" cycle length for investigation along with
motorist in terms of lost time, discomfort and the system.
frustration, and a direct cost in terms of fuel A general conclusion from the evaluation is that
consumption during idling. Excessive delay at the effectiveness of coordination traffic signal is
signalized intersections reflects inefficiency in the obvious for the site that was investigated, but it is

ISSN: 1790-1979 211 Issue 7, Volume 6, July 2009

 Mahmood Mahmoodi Nesheli, Othman Che Puan,
WSEAS TRANSACTIONS on ADVANCES in ENGINEERING EDUCATION Arash Moradkhani Roshandeh

specific in every site and condition. Therefore, results [13] TRANSYT-7F User’s Guide, U.S. Department of
from one application may not necessarily apply in Transportation, University of Florida, 2005.
different situations. For this reason, in one case study [14]McTrans, Highway Capacity Software.
simulation is a valuable tool for the design and Transportation Research Center. University of
evaluation of different strategies, as it can be used to Florida, Newsletter, Volume 21 Fall 2000.
determine if and how such strategies can be most [15] Avery, D.J., A Design Review Process for the
effectively implemented. Development of Traffic Signal Timing Plans,
Masters Report, University of Florida
Transportation Research Center, 1991.
Acknowledgment:
The authors are thankful to MOSTI through RMC of
Universiti Teknologi Malaysia (UTM) and UTM for
providing financial aid to carry out this research.

References:
[1] Angus P. Davol, Modeling of Traffic Signal
Control and Transit Signal Priority Strategies in
a Microscopic Simulation Laboratory, Master
Thesis, Massachusetts Institute of Technology,
September 2001.
[2] McShane, W.R., Roess, R.P., Prassas, E.S.,
Traffic Engineering. 2rd. ed. Prentice Hall,
New Jersey, 1998.
[3] Dong Nyong KIM, Evaluation on Simulation
Models For Urbancorridor Signal Optimizations,
Proceedings of the Eastern Asia Society for
Transportation Studies, Vol.4, October, 2003.
[4] Booz Allen Hamilton., Assessment of benefits
relative to costs of ITS facilities, VicRoads,
Melbourne., 2003.
[5] Roger R. Stough, George., Intelligent
Transportation Systems, Cases and Policies.USA:
Mason University., 2001.
[6] McShane, W.R., Roess, R.P., Prassas, E.S.,
Traffic Engineering. 3rd. ed. Prentice Hall, New
Jersey., 2004.
[7] Nicholas J. Garber, Lester A. Hoel., “Traffic and
Highway Engineering”. West Publishing Co.,
1999.
[8] Richard Whelan., (1995). Smart Highways, Smart
Cars. Boston: Artech House
[9] Traffic Engineering Handbook, Institute of
Transportation Engineers, 5th Edition, 1999.
[10] Fred L. Orcutt Jr., The Traffic Signal Book,
Prentice Hall, Englewood Cliffs, N.J., 1993.
[11] TRB, Highway Capacity Manual.
Transportation Research Board. National
Research Council, Washington, D.C., 2000.
[12] Wallace, Charles, Chang, Edmond, Messer,
Carroll, and Courage, Kenneth, Methodology for
Optimizing Signal Timing: PASSER II-90 Users
Guide, Office of Traffic Operations and Intelligent
Vehicle/Highway Systems – Federal Highway
Administration, Volume 3, December 1991.

ISSN: 1790-1979 212 Issue 7, Volume 6, July 2009

View publication stats