E3S Web of Conferences 445, 01004 (2023) https://doi.org/10.

1051/e3sconf/202344501004
GCEE 2023

Traffic Signal Coordination Based on Vissim

Software (Case Study of Sudirman Road in
Denpasar City, Indonesia)
Putu Alit Suthanaya , and Risma Putra
Department of Civil Engineering, Faculty of Engineering, Udayana University

Abstract. Transportation problems in the form of congestion, delays, as

well as air and noise pollution are very common in several big cities in
Indonesia such as Denpasar City (the capital of Bali Province). Traffic
congestion has been experienced on almost all major roads in Denpasar City
and one of them is Sudirman road. There are several signalized intersections
located at close distances. Traffic jam is usually experienced during peak
hour which causes an increase in travel time, energy consumption, and air
pollution. There is a potential to coordinate these intersections in order to
reduce congestion. The objective of this study is to optimize intersection
performances by coordinating traffic signals. This study applied a micro-
simulation approach by using Vissim software. The t-test method was used
to test the validity of the simulation. The results indicated that the
intersections were in the level of service F. Comparison of the intersection
performances showed that there was an increase in the intersection
performances as indicated by the decrease in the queue length, delay, stop
delay, and fuel usage. There was a decrease in the queue length of about -
18.52% (from 544.63 m to 443.74 m), a decrease in the delay of about -
20.41% (from 344.19 sec/pcu to 273.94 sec/pcu), a decrease in the stop
delays of about -21.61% (from 32.84 sec/pcu to 25.74 sec/pcu) and a
decrease in the fuel use of about -5.15% (from 17.19 liters/hour to 16.30
liters/hour).

1 Introduction
Cities in developing countries like Indonesia have experienced continuous growth in private
vehicle ownership which becomes one of the main concerns of the government. Based on the
study in Badung Regency, Bali, Indonesia, it was identified that there are a lot of elements
(safety, ability to accommodate more family members, and social standing) that affect the
choice between a motorcycle and a car, including monthly family income, monthly
transportation costs, the number of family members and students living in the home [1]. As
the traffic volume continues to increase, many cities in Indonesia (including Denpasar City)
have experienced increase in traffic congestion, especially at the signalized intersection. This
condition will be followed by the increase in the environmental impact. One of the ways to

Corresponding author: suthanaya@unud.ac.id

© The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons
Attribution License 4.0 (https://creativecommons.org/licenses/by/4.0/).
E3S Web of Conferences 445, 01004 (2023) https://doi.org/10.1051/e3sconf/202344501004
GCEE 2023

reduce the environmental impact of the road traffic congestion at the signalized intersection
is by implementing traffic signal coordination.
Traffic signal coordination is an important aspect of transportation engineering aimed at
improving the efficiency of urban traffic flow. There are many studies applying a micro-
simulation method to optimize signalized intersection performances. A study in Alexandria,
Egypt focused on applying optimization to simulate traffic signal timings under oversaturated
conditions, in order to find various solutions to the problem at hand [2]. The ExtendSim
simulation environment has been used to create a number of computer simulation models to
depict the actual system and suggested solutions. While the average wait time for a vehicle
has decreased by 32%, the average line length has decreased overall by 36%. Vissim is one
of the powerful simulation software [3]. VISSIM software can be used to simulate traffic
flow and evaluate the performance of traffic signal coordination systems. [4] investigated
''self-organizing signals,'' a novel traffic signal control paradigm based on local actuated
control but with certain extra principles that generate cooperation mechanisms. When
compared to an optimum coordinated-actuated scheme without transit priority, simulation
experiments in VISSIM on arterial corridors in Massachusetts and Arizona demonstrate
overall delay savings of up to 14%, and more than 30% in scenarios with transient
oversaturation.
An adaptive signal control plan along the Huangshan Road in Hefei, China has been
designed [5]. They used Transyt software to find the optimal fixed-time signal plan and
evaluating the plan using Vissim software. It was found that delay in the adaptive signal
control was reduced noticeably than that in the fixed time control. The VISSIM software is
used to simulate the traffic system before and after the improvement to overcome traffic delay
in the main riad in Beijing, China [6]. They discovered that while the average number of cars
per vehicle, the per capita delay, and the parking time have all dramatically decreased, the
number of vehicles and the number of passengers have clearly increased. The Vissim
software was applied to analyze the installation of a traffic signal at the skewed T-Intersection
in the Emakulam district of Kerala, India [7]. It was found that the vehicle delay has
decreased after the installation of the signal. The Vissim software was applied to design and
evaluate an isolated traffic light controlled intersection in Spisska Nova Ves, Slovakia [8].
From the two models developed have led to a 75% reduction of the queue length. The Vissim
software was used to analyze installation of traffic signal at unsignalized intersection in
Bekasi, Indonesia [9]. The study found that replenishing the intersection's two-phased signals
with a 106-second cycle time, as an alternative counter condition, was successful in reducing
24% of traffic conflict. Vissim has also been used to simulate traffic signal coordination. An
investigation on the benefits of signal coordination by means of traffic microsimulation
software, PTV Vissim in Bangalore (the capital city of Karnataka), India has been conducted
[10]. From Mother Theresa intersection to Sarvanton intersection, travel time and delay were
reduced after signal synchronization by 102.4 seconds and 110.32 seconds, respectively. The
traffic signal controllers have been optimized for a congested corridor in Malabe town, Sri
Lanka using Vissim software [11]. They found that emissions and fuel consumption were
reduced by 14.89%.
The applications of microsimulation modelling for assessing traffic safety that have been
subjected to critical analysis in terms of the application of various simulation tools [3]. They
stated that there is still a significant void in the development and application of simulation
model to evaluate traffic safety of non-lane based heterogeneous traffic environments that
predominate in many developing countries. Road traffic compositions in developed and
developing countries such as Indonesia are different. The road traffic in Indonesia is
dominated by motor cycle. Traffic signal coordination in Denpasar City is usually evaluated
manually. The application of the microsimulation model will be required to evaluate several
alternatives to obtain suitable traffic management solutions. The main objective of this study

2
E3S Web of Conferences 445, 01004 (2023) https://doi.org/10.1051/e3sconf/202344501004
GCEE 2023

was to minimize traffic congestion and fuel consumption by coordinating two signalized
intersections in Denpasar City as a case study, based on vissim simulation.

2 Traffic Signal Coordination

By allocating the right-of-way, one of the key goals of traffic signal timing (TST) settings is
to move people through an intersection safely and effectively. Some of the TST setting
parameters ought to be flexible enough to accommodate changes in traffic demand, while
others ought to be under the authority of the traffic management agency. Green time, cycle
length, phase sequence, change interval, and offset make up these control parameters. [12].
Some of the often used microsimulation programs for investigating the wide range of
dynamic problems in urban traffic are AIMSUN, CORSIM, MATSim, Paramics, SUMO,
and VISSIM. The features and traits of many widely used traffic simulation systems has been
assessed and also presented a comparative analysis by highlighting a few unique features
[13]. VISSIM, AIMSUN, and CORSIM were shown to be acceptable for modeling aspects
such arterial, freeway congestion, and integrated networks of freeways and streets, while
AIMSUN, CORSIM, and PARAMICS were found to be useful for modeling intelligent
transportation systems (ITS) [14].
According to [11]. the VISSIM microsimulation software is the most well-known and
often used of those. A multi-modal microscopic traffic flow simulation program called PTV
VISSIM was developed by PTV in Karlsruhe, Germany. Among the uses are traffic
engineering, public transportation, urban planning, fire safety, and 3D visualization for
illustrative and public communication. VISSIM can alter driver, vehicle, and other behaviors
from the point of origin to the point of destination. Vehicles such as vehicles, buses, and
trucks, as well as public transportation, bicycles, pedestrians, and rickshaws, can all be
simulated using VISSIM and interacted with. Priority rules, conflict zones, and signal heads
can all be used in VISSIM to depict vehicle conflict points.

3 Research Methods
Figure 1 shows the study location at two intersections along Sudirman Street in Denpasar
City, Bali, Indonesia. Denpasar City has an area of 127.78 km2 with a total population of
726,599 people [15].The majority of the road network's traffic volume occasionally
increases. Particularly at peak times in the morning and afternoon, there has been traffic
congestion. This condition has led to an increase in energy consumption and air pollution.

3
E3S Web of Conferences 445, 01004 (2023) https://doi.org/10.1051/e3sconf/202344501004
GCEE 2023

.
Fig. 1. Study Location at Two Intersection Along Sudirman Street in Denpasar city.
Data collected include the intersection geometry, traffic signal (timing, phasing, and cycle
time), traffic volume, distance between the two intersections, and average travel time. Traffic
volume data were collected for 12 hours from 07:00 a.m. to 07:00 p.m. during the weekdays.
Using a measuring wheel, field measurements were used to gather geometric data. Using a
test vehicle, data on typical journey times was gathered. By making observations at the
signalized junctions, the signal timings, phasing, and cycle times were gathered.
The model of the chosen corridor was created with the help of the microsimulation
program VISSIM. In the early stages of the investigation, the queue length was taken into
consideration as the performance measure for calibrating and testing the program at the
junction level. The line length, average stop delay, average vehicle delay, CO emissions,
NOx emissions, and fuel consumption were taken into account to assess the effectiveness of
the traffic signal modification. Using the calibration parameters, the base model was adjusted
to the local circumstances.
By comparing the observed and anticipated values, the validation process was carried out
based on the t-test. Data for the vehicle volume, vehicle composition, turning movement, and
signal timing were entered into the VISSIM program while the model was being created. The
gathered and examined data, along with the preset calibration parameter values, were used to
create the corridor's base model. The VISSIM software was used to optimize the traffic
signals. Priority rules and a new signal program were developed as a stage-based signal
controller. The best traffic signal controller timings were then discovered by doing the signal
timing optimization.

4 Results And Discussion

4.1 Model idation

The average value of the two variables (mean) for the field is 433.2564, while for Vissim it
is 401.9744. Thus, descriptively, it can be concluded that there is a difference in the average
traffic volume results between the field and Vissim. Furthermore, to prove whether the
difference is significant or not, it can be seen in Table 1. The result of the independent sample
t-test indicates that there is no significant difference between the observed and predicted
values.

4
E3S Web of Conferences 445, 01004 (2023) https://doi.org/10.1051/e3sconf/202344501004
GCEE 2023

Table 1. Output independent t-test.

Independent Samples Test

Levene's
Test for
Equality of
Variances t-test for Equality of Means
Std. 95% Confidence
Mean Error Interval of the
Sig. (2- Differen Differenc Difference
F Sig. t df tailed) ce e Lower Upper
Out Equal , ,8 ,3 76 ,75 31,28 99,3126 - 229,08030
put varian 0 6 1 4 205 3 166,516
ces 3 3 5 20
assum 0
ed
Equal ,3 75,95 ,75 31,28 99,3126 - 229,08240
varian 1 0 4 205 3 166,518
ces 5 30
not
assum
ed
Intersection performance analysis was carried out using two methods: based on
regulations in the Indonesian Highway Capacity Manual and based on the output of a model
simulation using the Vissim software. The two methods produced different performance
outputs. Furthermore, the two output results were compared based on similar intersection
performance parameters. Table 2 shows the results of a comparison of the intersection
performances between IHCM and Output Vissim for the afternoon peak.
Table 2. Comparison of the intersection performances based on IHCM and Output Vissim for the
afternoon peak.

Traffic volume (pcu/hr)

No Condition Int FE Unud Int Waturenggong
N E S W N E
Traffic counting 1957,90 114,20 500,90 586,10 1821,70 1003,70
VISSIM 1837,00 135,00 365,00 557,00 1990,00 836,00
1
Difference 120,90 20,80 135,90 29,10 168,30 167,70
Percentage (%) 6,17% 15,41% 27,13% 4,97% 8,46% 16,71%
Queue length (m)
Condition Int FE Unud Int Waturenggong
N E N W N E
IHCM 710,75 34,94 38,29 320,32 145,67 822,44
VISSIM 680,66 40,50 36,36 338,85 215,83 873,43
2
Difference 30,09 5,56 1,93 18,52 70,15 51,00
Percentage (%) 4,23% 13,73% 5,04% 5,47% 32,50% 5,84%
Avg. Delay (sec/pcu)
Simpang
Condition Int FE Unud
Waturenggong
N E N W N E

5
E3S Web of Conferences 445, 01004 (2023) https://doi.org/10.1051/e3sconf/202344501004
GCEE 2023

IHCM 555,67 41,54 37,08 346,19 36,32 722,04

VISSIM 512,72 40,22 37,43 336,11 44,82 643,56
3
Difference 42,95 1,32 0,35 10,09 8,50 78,48
Percentage (%) 7,73% 3,17% 0,93% 2,91% 18,96% 10,87%
LoS F F

Based on Table 2, it is known that during peak hour conditions, the largest percentage
difference in volume occurs in the southern approach of the FE Unud intersection, namely
27.13%; the largest percentage difference in queue length occurs in the southern approach of
the FE Unud intersection, at 12.44%; and the largest percentage difference in the average
delay occurs in the northern approach of the Waturenggong intersection, at 18.96%.

4.2 Analysis Traffic Signal Coordination for Existing Condition (Before

Coordination)
Based on the survey results at the study site, the average travel speed (v) from the FE Unud
Intersection to the Waturenggong Intersection was 30 km/hour, converted to 8.333 m/s.
Travel speed (v) also applies in the reverse direction. The distance (l) from the FE Unud
Intersection (FE) to the Waturenggong Intersection (WR) is 400 m. Offset values for existing
conditions were 48 seconds. The bandwidth used was the fastest green time, which was 45
seconds from the direction of Sudirman Street (N) to Waturenggong Street (W), and 25
seconds from the direction of Waturenggong Street (E) to Sudirman Street (N). Based on this
analysis, it can be seen that the signal arrangement in the afternoon is not coordinated
between the two intersections.

4.3 Analysis Traffic Signal Coordination (After Coordination)

From the results of the coordination analysis of the existing condition intersections, it can be
concluded that the two intersections do not yet have signal coordination during the morning
peak hours, afternoon peak hours, and evening peak hours. During morning peak hours, the
results of the intersection performance from the IHCM analysis and the Vissim software
output show that the longest queues and delays are at the northern approach to the FE Unud
intersection and the northern approach to the Waturenggong Intersection. Because the queue
length and the biggest delay are on the northern approach, the green time distribution must
be greater on the northern approach than the east approach at the Waturenggong intersection.
At noon peak hours, the intersection performance from the IHCM analysis and Vissim
software output shows similar results to the intersection performance in the morning. The
difference occurs in the afternoon peak hours. The results of the intersection performance
from the analysis of IHCM and the output of the Vissim software show that the longest
queues and delays are at the north approach of the FE Unud intersection and the east approach
of the Waturenggong Intersection. Because the queue length and the biggest delay are on the
east approach, the green time distribution must be greater on the east approach than the north
approach at the Waturenggong intersection.
The bandwidth used is the fastest green time, which is 40 seconds from the direction of
Sudirman Street (N) to Waturenggong Street (W) and 30 seconds from Waturenggong Street
(E) to Sudirman Street (N). After setting the signal phase timing, the two intersections have
also been coordinated in the afternoon peak. Based on the results obtained from the analysis,
several alternative signal timings were inputted into the Vissim software, as can be seen in
Figure 2. The two intersections were coordinated based on the same cycle time of 110
seconds.

6
E3S Web of Conferences 445, 01004 (2023) https://doi.org/10.1051/e3sconf/202344501004
GCEE 2023

Fig. 2. Vissim Software Program Signals at the Waturenggong Intersection Before and After the
Intersection Coordination Settings at Afternoon Peak Hours.

4.4 Intersection Performance Results After Setting Signal Coordination in the

Vissim Software
Based on the signal coordination analysis that has been carried out, the new intersection
performance results are obtained from the output of the Vissim software. The performance
results will be compared with the intersection performance resulting from the output of the
Vissim software when the existing conditions were obtained previously to better understand
the differences between the two. The performance parameters being compared are queue
length, delay, stop delay, and fuel usage. Table 3 shows a comparison of before and after
coordination.
Table 3. Comparison of the Intersection Performance Before and After Coordination (Afternoon
Peak).
fuel
Intersectio qlenmax vehdelayall stopdelayall
Movement consumption
n
(m) sec/pcu sec/pcu liter/hr
Condition Existing Coordinated Existing Coordinated Existing Coordinated Existing Coordinated
N 680,66669,86 512,72 487,39 46,73 43,60 11,72 11,52
40,2
E 40,50 36,50 40,27 33,24 31,24 1,80 1,80
2
37,4
FE Unud S 36,36 33,36 35,93 28,31 28,17 3,52 3,52
3
W 338,85 332,85 336,11 326,11 57,62 55,50 3,88 3,88
Avg 274,09 268,14 231,62 222,42 41,47 39,63 5,23 5,18
Percentage % -2,17% -3,97% -4,45% -0,95%
44,8
N 215,83 273,83 54,82 31,06 32,03 27,75 26,36
2
Waturenggong E 873,43 613,65 643,56 493,06 34,62 19,46 6,62 6,24
Avg 544,63 443,74 344,19 273,94 32,84 25,74 17,19 16,30
Percentage % -18,52% -20,41% -21,61% -5,15%

From Table 3 it can be seen that at peak hours in the afternoon there was a decrease in
performance parameters after an alternative application of signal coordination was carried
out, namely at the FE Unud intersection there was a decrease in queue length of -2.17%,
delay of -3.97%, stop delay of -4.45% and fuel use of -0.95%, while at the Waturenggong

7
E3S Web of Conferences 445, 01004 (2023) https://doi.org/10.1051/e3sconf/202344501004
GCEE 2023

intersection, there was a decrease in queue length of -18.52%, -20.41% delay, -21.61% stop
delay, and -5.15% fuel usage.

5 Conclusion
Based on the results of the validation using statistical analysis of the t-test, it can be concluded
that there is no significant difference between observed and predicted traffic volume. Based
on the intersection coordination analysis, alternative results of signal coordination between
the two intersections are obtained, which are inputted and re-analyzed with the Vissim
software using the node results to obtain a new intersection performance. After carrying out
the application of signal coordination, it is known that there is a decrease in the performance
parameters of queue length, delays, stop delays, and fuel usage, respectively, during morning
peak hours, afternoon peak hours, and evening peak hours. The most significant decrease
occurred at the Waturenggong intersection arm during the afternoon peak hours, namely a
decrease in the queue length parameter of -18.52%, a delay of -20.41%, stop delay of -
21.61%, and a fuel consumption savings of -5.15%.

References
1. P. A. Suthanaya and P. P. Winaya, Civil Engineering and Architecture 11, 1498
(2023)
2. J. Erdmann, R. Oertel, P. Wagner, A. a Ezzat, H. a Farouk, K. S. El-kilany, and A.
F. A. Moneim, 2014 International Conference on Industrial Engineering and
Operatios Management Bali, Indonesia. 3007 (2014)
3. S. M. S. Mahmud, L. Ferreira, M. S. Hoque, and A. Tavassoli, IATSS Research 43,
27 (2019)
4. B. Cesme and P. G. Furth, Transp Res Part C Emerg Technol 48, 1 (2014)
5. R. Tian and X. Zhang, Transportation Research Procedia 21, 141 (2017)
6. X. Guo, H. Sun, and X. Wang, AIP Conf Proc 1839, (2017)
7. M. I. P. M. and H. H. R. Salim, L. Samuel, L.Francis, International Journal of
Engineering Science and Computing (IJESC) 9, 23296 (2019)
8. J. Fabianova, P. Michalik, J. Janekova, and M. Fabian, Open Engineering 10, 48
(2020)
9. I. M. Septyaningrum, R. Y. Anindita, and Y. Ktopianto, RSF Conference Series:
Engineering and Technology 2, 294 (2022)
10. N. M. and P. S. G. A. Amaranatha, International Journal of Future Generation
Communication and Networking 13, 3164 (2020)
11. D. Gunarathne, N. Amarasingha, and V. Wickramasighe, Science, Engineering and
Technology 3, 9 (2023)
12. S. S. S. M. Qadri, M. A. Gökçe, and E. Öner, European Transport Research Review
12, 1 (2020)
13. N. T. Ratrout and S. M. Rahman, A Comparative Analysis of Currently Used
Microscopic and Macroscopic Traffic Simulation Software (n.d.)
14. D. Salgado, D. Jolovic, P. T. Martin, and R. M. Aldrete, Procedia Comput Sci 83,
441 (2016)
15. D. Bureau of Statistics, Denpasar City in Figure Bureau of Statistics, (2022)