Volume 10, Issue 1, January – 2025 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165 https://doi.org/10.5281/zenodo.14621424

Adaptive Traffic Signaling Control Using SUMO

Simulator
Taha Abdulwahid MAHMOOD*1; Muzamil Eltejani Mohammed ALI2;
1
Department of Electrical Engineering, Faculty of Engineering, Kirkuk University, Kirkuk, Iraq
2
R&D Center, MOSAS Smart Transportation Technologies Inc.,42160 Konya, Turkiye

Abstract:- In today's world, transportation vehicles are developed for traffic signaling, traffic situations can be
essential for meeting mobility needs and moving goods simulated on the computer at reasonable cost and without
efficiently. To reduce average waiting times across varying danger, which is why the computer simulations are selected in
traffic flow rates, a two-stage, three-module fuzzy logic the growth of traffic signal technology [4]. Traffic signal
system has been developed for real-time management of systems divide into three major categories: Human-controlled
signalized junctions. The first stage includes two modules: traffic light systems, fixed signal systems, and smart. The
the "next phase selection module," which monitors traffic human-controlled method utilizes hand motions and voice
conditions of all red phases (except the current green) and commands. Static-time systems follow predetermined
selects the most urgent one based on 30 fuzzy rules. The schedules based on historical data and lack the flexibility to
"extension time module" assesses the green phase's traffic make real-time changes. An intelligent traffic control system
conditions to decide whether to stop or extend it, using 12 uses sensors to continually monitor traffic conditions. This
fuzzy rules. The second stage features a "decision module" allows for dynamic modification of signal timing and signal
with 10 fuzzy rules, which determines whether to replace sequencing and improved intersection management. These
or maintain the current green phase based on inputs from approaches enhance the safety of both vehicles and
the previous modules. This system was implemented using pedestrians while increasing the capacity of the intersection at
the SUMO traffic simulation tool, utilizing real-world the same time [5].
traffic data from a congested intersection in Kilis, Turkey.
The fuzzy logic traffic management system outperformed This study discovers the improvement of an adaptive
conventional fixed-time control, achieving substantial traffic signal management system for smart cities utilizing
reductions in average waiting times: 76.46%, 56%, 50%, fuzzy logic and the SUMO simulation. The paper concentrates
and 60% for the four analyzed areas. on an isolated intersection and compares the adaptive system
performance based on fuzzy logic with a traditional fixed-time
Keywords:- Fuzzy Logic, Simulation Of Urban Mobility, traffic signal controller. The adaptive system uses real-time
Intelligent Traffic Control, Isolated Junction, Vehicles. traffic data, including vehicle density and queue length, to
dynamically adjust signal timing strategically and enhance
I. INTRODUCTİON efficiency. The results indicate that the fuzzy logic system
significantly reduces average waiting times compared to the
Today, the number of vehicles is increasing rapidly. As a fixed time method.
result, the problem of traffic congestion grows and in large
cities, this problem is inextricably affecting people's quality of II. RELATED WORKS
life. In recent years, high-speed living traffic has become an
important part of the modern city [1]. As the number of road This unit discusses traffic signal systems and presents
users, vehicles and pedestrians increases, light cycles recommendations for smart solutions:
controlled by traffic demand increase. Traffic congestion
brings with it many problems; Besides environmental, social Lee and Lee-Kwang (1999), the proposed traffic signal
and economic losses, it causes time loss and directly affects control system employs a fuzzy logic approach to manage
production and fuel usage [2]. High rates of vehicle use; multiple intersections adaptively. It dynamically adjusts signal
coupled with the lack of available space to build new transport phase sequences and durations to enhance performance. The
infrastructure and public funds, it serves to further complicate system consists of three key modules: observation, next phase,
the problem. Within this framework, it is necessary to develop and decision, all working together for real-time decision-
smart and economical solutions to improve the service quality making. Each module features its own fuzzy rule base,
for road users [3]. In order to reduce all these problems in a allowing for context-sensitive responses to traffic conditions.
relatively cheaper way and to ensure the best use of the Results show that the fuzzy logic system significantly
existing road network, a traffic signaling system technique has outperforms conventional vehicle-actuated methods, reducing
been implemented. Many simulation models have been average vehicle delays [6].

IJISRT25JAN158 www.ijisrt.com 128

Volume 10, Issue 1, January – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.5281/zenodo.14621424

Askerzade and Mahmood (2010), fuzzy logic technology network (ANN) and Adaptive Neuro-Fuzzy Inference System
is utilized to control traffic lights, managing the duration of (ANFIS) models. The results indicate that the fuzzy logic
the green light based on traffic conditions. The fuzzy approach effectively reduces total delay time and outperforms
controller has three input variables: the number of vehicles the other systems in terms of intersection efficiency [11].
arriving, the number of vehicles waiting, and those making
right or left turns. Its output variable adjusts the green signal Gündoğan et al. (2014), this study presents the ATAK
duration based on these inputs, demonstrating their influence system, a real-time adaptive traffic control method that
on extension periods. The Tagaki-Sugeno method is employed manages signalized intersections through a combination of
to calculate these results, and its outcomes are compared with genetic algorithms and fuzzy logic. The ATAK system
those of the Mamdani method under various traffic scenarios. dynamically adjusts traffic signal timings based on variables
The Tagaki-Sugeno method yields superior results in reducing like traffic volume and occupancy rates. Its performance was
total waiting and transportation times. Decreased waiting assessed against a conventional multi-plan traffic control
times lead to lower fuel consumption and reduced system. The results show that the ATAK system can improve
environmental pollution [7]. average travel times by up to 15%. Furthermore, it reduced
carbon emissions by approximately 10% compared to the
Azimi et al. (2010), present a new fuzzy logic controller multi-plan control approach. [12].
for an isolated signalized intersection. The system consists of
the following variables: the status of the signal, the number of III. TRAFFIC SIGNALING
vehicles leaving traffic and the number of changes in queue
length. The exact value of the output was calculated by the The traffic light signal shown in Fig. 1, is a widely
centroid deviation technique. Simulation was carried out using recognized signaling device typically installed at road
fuzzy rules (81 in number), MATLAB 7.4 and fuzzy logic intersections or pedestrian crossing points to regulate the flow
toolbox. The results of this method and the improvement of traffic through the use of colored lights. This type of traffic
percentage have shown us that the fuzzy control system control system is widely implemented in cities across the
reduces the average waiting time at any intersection compared globe to manage the movement of vehicles and
to fixed time control [8,24]. pedestrians. All light signals are illuminated with two primary
colors, red light means “stop” and green light means allowing
Zaied and Al Othman (2011), This study introduces a passage. The first light signal was established in London in
fuzzy logic-based traffic signal control system designed for 1868, in 1913, tricolor (green, yellow, red) lights were
two-way intersections. The system dynamically adjusts signal invented in the United States. In 1922, the signal was found to
timing intervals in response to current traffic conditions and be working mechanically and then states began to widely
vehicle movement levels. Implemented in MATLAB, the establish light signals to control traffic [16-17].
fuzzy logic control algorithm features a robust set of 78 fuzzy
rules. These rules focus on optimizing signal cycle times and
minimizing unused green light periods across different phases.
When compared to traditional traffic control systems, this
fuzzy logic approach shows enhanced performance by
significantly reducing instances of unused green time [9].

Prasetiyo et al. (2015), this work outlines the design and

implementation of an adaptive traffic light control system
using the Sugeno-style fuzzy logic approach. The system
leverages fuzzy logic to dynamically calculate the optimal
duration of the green signal phase at intersections. Although
the fuzzy logic-based adaptive control shows a slightly higher
average waiting time than fixed-time traffic signals, it Fig. 1. Traffic Light Signals
outperforms in reducing vehicle queues and increasing the
overall number of vehicles departing the intersection. [10].  Some Terms of Traffic Signaling
Here, some basic concepts related to traffic signaling are
Dereli et al. (2018), This study introduces the design of a briefly explained by making use of Fig. 2, and Fig. 3.
fuzzy logic-based traffic signal controller for a single
intersection, aiming to dynamically adjust green light duration, The most important concepts used in signaling:
reduce overall vehicle delays, and minimize waiting vehicles.  Junction: These are the areas where traffic flows coming
The fuzzy logic system was implemented using the Fuzzy from several roads are separated and intersected.
Inference System (FIS) in MATLAB. To assess its
 Road: is a part that generally used by vehicles.
performance, the researchers compared it with artificial neural

IJISRT25JAN158 www.ijisrt.com 129

Volume 10, Issue 1, January – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.5281/zenodo.14621424

 Lane: Each section of the road, separated by signs, for membership. This is represented by a membership function
vehicles to travel safely. that ranges from 0 to 1. This flexibility is a key distinguishing
 Traffic flow: The arrival and departure of vehicles on the characteristic of fuzzy logic compared to classical binary logic
road. [19]. Its origins date back to Zadeh's groundbreaking work in
 Red time: The time at which a red light is given to a phase. the 1960s, where he questioned the limitations of traditional
 Yellow time: Yellow light to signal phase is the known Boolean logic and paved the way for a new paradigm in
time. mathematics and computer science. Zadeh's key insight was
 Green time: When the green light is given to the phase. that many real-world problems involve imprecise or
ambiguous information, which classical logical frameworks
cannot effectively address [20]. Researchers such as Mamdani,
Sugeno, and Takagi-Sugeno-Kang played pivotal roles in
developing fuzzy control systems. These systems have been
widely applied in fields like industrial automation, robotics,
and consumer electronics. The ongoing advancements in fuzzy
logic, coupled with its seamless integration with emerging
technologies, suggest an exciting future where the power of
imprecision will play an increasingly vital role in shaping the
way we interact with and understand the world around us [21].

V. SIMULATION OF URBAN MOBILITY (SUMO)

Sumo was developed in 2001 by the German aerospace

center (DLR). With the program, which was published in
2002, users gained a basic microscopic program that can be
Fig. 2. Intersection, Road, Lane, Traffic Flow and Traffic studied on traffic networks, it is open-source software [22].
Lights With the TraCI module (Traffic Control Interface), this
interface allows for interactive monitoring and control of
 Phase: The time allowed to pass more than one flow at the various aspects of the traffic environment. All components can
junction. The time from the start of a current receiving the be manipulated in the simulation using the Python program. In
first green light to the next receiving a green signal is the this way, it is much easier to diversify than other simulation
phase. programs. SUMO can produce output data that can be used in
 Cycle time: The total time during which each phase of the various fields such as vehicle track data, vehicle type and
junction is allowed to pass once. number data passing through point, emission values of
vehicles, lane change information at desired frequency as
output with all specified features [23].

VI. APPLICATION AND RESULTS

Fuzzy logic-based traffic signal control provides a

flexible alternative to conventional fixed-time traffic lights,
adapting to various traffic patterns at intersections. This
system employs detectors to sense vehicle presence and count
the number of vehicles, effectively managing traffic density on
Fig. 3. Phase and Cycle Time individual lanes and improving the assessment of changing
conditions. The fuzzy logic controller dynamically adjusts
IV. FUZZY LOGIC green light durations, and the sequence of signal phases based
on current traffic conditions. In this two-stage traffic light
In a world dominated by the binary nature of traditional system, the controller not only decides whether to extend or
logic, where statements are either true or false, the concept of terminate the current green phase but also determines which
fuzzy logic emerges as a revolutionary approach to reasoning. red phase will become the next green phase and the
Fuzzy logic, pioneered by the visionary mathematician Lotfi appropriate extension time. As a result, the phase sequence
Zadeh in the 1960s, recognizes that many real-world remains uncertain. The performance of this fuzzy logic-based
phenomena defy the rigid constraints of classical logic [18]. system will be evaluated using average vehicle waiting time as
Fuzzy logic is based on the concept of fuzzy sets unlike the primary metric. Fig. 4, shows the schematic diagram of the
classical sets, where an element either fully belongs or does proposed controller.
not belong, fuzzy sets allow for varying degrees of

IJISRT25JAN158 www.ijisrt.com 130

Volume 10, Issue 1, January – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.5281/zenodo.14621424

Fig. 4. Control unit diagram

The first stage involves two distinct modules:

 Next Stage Module

This module identifies the next candidate for the green phase by evaluating traffic conditions across all red phases, excluding the
current green one. It uses waiting times and queue lengths from these phases as inputs. The output is the selected phase along with its
urgency level. By comparing urgency levels of all input phases, the module determines the most urgent one. Inputs have six
membership functions for waiting time and queue length, ranging from zero to very long. The Urgency Level output consists of five
membership functions, from zero to very high, as shown in Fig. 5.

Waiting time
(a)

Queue length
(b)

IJISRT25JAN158 www.ijisrt.com 131

Volume 10, Issue 1, January – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.5281/zenodo.14621424

Urgency degree
(c)

Fig. 5. The input parameters include waiting time and queue length, while the output parameter is the degree of urgency.

Table 1 outlines the 30 fuzzy rules governing the decision-making process within this module. As an example, if the waiting time
is categorized as "Zero" and the queue length is deemed "Long", the corresponding urgency level would be determined as "Medium".

Table 1. Rules of the Next Stage Module.

Rule Waiting time Queue length Urgency level
R1 Z VS Z
R2 VS VS L
R3 S VS M
R4 M VS M
R5 L VS M
R6 VL VS M
R7 Z S Z
R8 VS S M
R9 S S M
R10 M S H
R11 L S H
R12 VL S H
R13 Z M L
R14 VS M L
R15 S M M
R16 M M M
R17 L M H
R18 VL M VH
R19 Z L M
R20 VS L M
R21 S L H
R22 M L H
R23 L L VH
R24 VL L VH
R25 Z VL M
R26 VS VL H
R27 S VL H
R28 M VL VH
R29 L VL VH
R30 VL VL VH

IJISRT25JAN158 www.ijisrt.com 132

Volume 10, Issue 1, January – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.5281/zenodo.14621424

 Extension time module

This module evaluates the traffic conditions during the current green phase. Using the collected data, it generates a stop degree to
decide whether the controller should terminate or extend the green light duration. The inputs for this module include the number of
remaining vehicles and the number of passing vehicles. The output consists of a decision to either stop or prolong the active green
light phase. Each input parameter has four membership functions: remaining vehicles are categorized as (Zero, Short, Medium, Long),
while passing vehicles are categorized as (Zero, Low, Medium, High). The output, or stop degree, features three membership
functions: (No, Maybe, Yes), show Fig. 6.

Remaining vehicles number

(a)

Passing vehicles number

(b)

Stop or Extend
(c)

Fig. 6. The fuzzy sets for the number of remaining vehicles, passing vehicles, and the decision to stop or extend.

IJISRT25JAN158 www.ijisrt.com 133

Volume 10, Issue 1, January – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.5281/zenodo.14621424

Table 2 presents the 12 fuzzy rules that govern the decision-making process within this module. As an example, if the number of
remaining vehicles is assessed as "Short" and the number of passing vehicles is deemed "High", the resulting decision would be to
either stop or extend the green light duration, which is classified as "No".

Table 2. Rules of Extension Time Module

Rule Remaining vehicles Passing vehicles Stop or Extend
R1 Z L Y
R2 Z M MB
R3 Z H MB
R4 L L MB
R5 L M MB
R6 L H N
R7 M L N
R8 M M N
R9 M H N
R10 H L N
R11 H M N
R12 H H N

 Second Stage Decision module

The primary function of this module is to determine whether the current green phase should be changed or maintained based on
the outputs from the preceding stages. The inputs to this decision module are the urgency degree and the stop/extend decision
generated by the earlier modules. The output of this module is the final decision, which is represented using two membership
functions: "No" and "Yes", as shown in Fig. 7.

This decision-making process takes into account the evaluated urgency degree as well as the recommendation to either stop or
extend the green phase duration. By synthesizing these inputs, the Decision Module arrives at the appropriate course of action,
whether to keep the existing green phase or initiate a change to the next phase. The flexibility of this two-stage approach ensures that
the traffic signal control system can adaptively respond to the prevailing traffic conditions and optimize the signal timing accordingly.

Fig. 7. The fuzzy sets of decision

Table 3 outlines the 10 fuzzy rules that define the decision-making process within this module. To illustrate, if the urgency
degree is classified as "Very High" and the stop/extend decision is recognized as "Yes", the final output of the Decision Module would
be "Yes", signifying the need to transition to the next green phase.

Table 3. Rules of the Decision Module

Rules Urgency degree stop/extend decision Decision
R1 Z MB N
R2 L MB Y
R3 M MB Y
R4 H MB Y
R5 VH MB Y
R6 Z Y N

IJISRT25JAN158 www.ijisrt.com 134

Volume 10, Issue 1, January – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.5281/zenodo.14621424

R7 L Y Y
R8 M Y Y
R9 H Y Y
R10 VH Y Y

VII. WORKSPACE

The fuzzy logic and fixed-time methods we developed were implemented in simulations at four different intersections in Kilis
city. Additionally, a 3-second transition time was allocated for each workspace to prevent accidents during phase changes. Our
simulation studies are detailed as follows:

 First Workspace
As the first Workspace, in Fig. 8, the junction of Ali Metin Dirimtekin Street in Ekrem Çetin Quarter of Kilisili was selected and
studied on 24-hour data.

Fig. 8. First Workspace

At a nine-lane, four-phase intersection, the first and second phases each have three lanes, the third phase has two lanes, and the
fourth phase has a single lane. A detector is installed on the road for each lane, as illustrated in Fig. 9. Utilizing the SUMO program,
the cycle time is set to 102 seconds to minimize the average waiting time in the fixed-time traffic control system. Specifically, the
durations for phases 1, 2, 3, and 4 are 15, 15, 14, and 40 seconds, respectively.

Fig. 9. Modeling the first workspace on Simulation

The results obtained from the simulation as shown in Table 4, fuzzy logic system reduces total waiting time from 996204
seconds to 234751 seconds compared to fixed time system, The total trip time is reduced from 1775617 seconds to 1037921 seconds,

IJISRT25JAN158 www.ijisrt.com 135

Volume 10, Issue 1, January – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.5281/zenodo.14621424

the average vehicle travel time is reduced from 92 seconds to 53 seconds, and the average vehicle waiting time is reduced from 51
seconds to 12 seconds. Fuzzy logic system has been observed to be better.

Table 4. Comparison of Results

Control Methods Fixed Time Method Fuzzy Logic Method
Total number of vehicles 19230 19230
Total Waiting Time (Sec) 996204 234751
Total Travel Time (Sec) 1775617 1037921
Average Vehicle Travel Time (Sec) 92 53
Average Vehicle Waiting Time (Sec) 51 12

 Second Workspace
As our second Workspace, the junction of Kilis Yaşar Aktürk Neighborhood, Vatan Street, which is seen in Fig. 10, was selected
and studied on 24-hour data.

Fig. 10. Second Workspace

As illustrated in Fig. 11, the intersection features eight lanes organized into four phases, with each phase containing two lanes. A
detector is positioned along the path for each lane. Using the SUMO program, the fixed-time traffic control system is configured with
a cycle time of 56 seconds to minimize the average waiting time. Accordingly, the durations for phases 1, 2, 3, and 4 are set to 11
seconds each.

Fig. 11. Modeling the second workspace on Simulation

IJISRT25JAN158 www.ijisrt.com 136

Volume 10, Issue 1, January – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.5281/zenodo.14621424

The results obtained from the simulation in Table 5, fuzzy logic system compared to fixed time system the total waiting time is
reduced from 421547 seconds to 185397 seconds, The total trip time is reduced from 1052305 seconds to 790478 seconds, the average
vehicle travel time is reduced from 63 seconds to 47 seconds, and the average vehicle waiting time is reduced from 25 seconds to 11
seconds. Fuzzy logic system has been observed to be better.

Table 5. Comparison of results

Control Methods Fixed Time Method Fuzzy Logic Method
Total number of vehicles 16597 16597
Total Waiting Time (Sec) 421547 185397
Total Travel Time (Sec) 1052305 790478
Average Vehicle Travel Time (Sec) 63 47
Average Vehicle Waiting Time (Sec) 25 11

 Third Workspace
As the third Workspace, in Fig. 12, Kilis Mehmet Rıfat Kazancıoğlu Neighborhood, Gen. Safter Necioğlu Boulevard was
selected and 18:30 hours data were studied.

Fig. 12. Third Workspace

As depicted in Fig. 13, the intersection consists of eight lanes divided into three phases: the first phase has three lanes, and the
second phase has two lanes. A detector is installed in the path for each lane. Using the SUMO program, the fixed-time traffic control
system is designed with a cycle time of 65 seconds to minimize the average waiting time. Specifically, the durations for phases 1, 2,
and 3 are set to 20, 20, and 16 seconds, respectively.

Fig. 13. Third workspace modeling on Simulation

IJISRT25JAN158 www.ijisrt.com 137

Volume 10, Issue 1, January – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.5281/zenodo.14621424

Table 6 results obtained from the simulation, fuzzy logic system compared to fixed time system, the total waiting time is reduced
from 269940 seconds to 131490 seconds, the total travel time is reduced from 977125 seconds to 826411 seconds, the average vehicle
travel time is reduced from 61 seconds to 51 seconds, and the average vehicle waiting time is reduced from 16 seconds to 8
seconds. Fuzzy logic system has been observed to be better.

Table 6. Comparison of results

Control Methods Fixed Time Method Fuzzy Logic Method
Total number of vehicles 15977 15977
Total Waiting Time (Sec) 269940 131490
Total Travel Time (Sec) 977125 826411
Average Vehicle Travel Time (Sec) 61 51
Average Vehicle Waiting Time (Sec) 16 8

 Fourth Workspace
As the fourth Workspace, in Fig. 14, Kilis Ekrem Çetin Neighborhood, Doğan Güreş Pasha Boulevard was selected and 24-hour
data were studied.

Fig. 14. Fourth Workspace

At an intersection with eleven lanes organized into five phases, the first phase has three lanes, while the second, third, fourth, and
fifth phases each have two lanes. A detector is installed for each lane, as shown in Fig. 15. Utilizing the SUMO program, the cycle
time is configured to 85 seconds to minimize the average waiting time in the fixed-time traffic control system. Specifically, the
durations for phases 1, 2, 3, 4, and 5 are set to 16, 11, 16, 16, and 11 seconds, respectively.

Fig. 15. Model workspace modeling on Simulation

IJISRT25JAN158 www.ijisrt.com 138

Volume 10, Issue 1, January – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.5281/zenodo.14621424

Table 7 results obtained from the simulation, the fuzzy logic system reduces the total waiting time from 413544 seconds to
165492 seconds, the total travel time decreases from 824562 seconds to 559979 seconds, the average vehicle travel time decreases
from 61 seconds to 41 seconds and the average vehicle standby time decreases from 30 seconds to 12 seconds. Fuzzy logic system has
been observed to be better.

Table 7. Comparison of results

Control Methods Fixed Time Method Fuzzy Logic Method
Total number of vehicles 13500 13500
Total Waiting Time (Sec) 413544 165492
Total Travel Time (Sec) 824562 559979
Average Vehicle Travel Time (Sec) 61 41
Average Vehicle Waiting Time (Sec) 30 12

VIII. CONCLUSION

We developed a two-stage traffic signal control system for an isolated intersection, utilizing fuzzy logic principles. The
performance of this proposed system was compared to that of the traditional traffic control method. The traffic simulation was
conducted using the SUMO platform. In this system, the traffic signal controllers manage the intersection based on the traffic
information collected by their respective detectors. Using this real-time data, the fuzzy rule-based system determines the optimal
phase sequence and timing for the traffic signals. The results obtained from the simulation showed significant improvements when
compared to the conventional logic controller. Specifically, the proposed two-stage fuzzy logic-based system demonstrated 76.46%,
56%, 50%, and 60% enhancements in the respective working areas, as detailed in Table VIII. These findings suggest that the
recommended fuzzy logic-based approach outperforms the traditional fixed-time control method in effectively managing the traffic
flow at the isolated intersection.

Table 8. Average Vehicle Waiting Time İmprovement percentage

Fixed Time Method Fuzzy Logic Method
Variable Workspace İmprovement
First 51 12 %76.46
Average Vehicle Waiting Second 25 11 %56
Time (Sec) Third 16 8 %50
Fourth 30 12 %60

ACKNOWLEDGMENT [5]. Srinivasan, D., Choy, M. C., & Cheu, R. L. (2006).

Neural networks for real-time traffic signal control. IEEE
The authors express their gratitude to the MOSAŞ Transactions on Intelligent Transportation Systems, 7(3),
GROUP (Akıllı Ulaşım ve Tartım Sistemleri Şti.) for kindly 261-272.
providing the database used in this study. [6]. Lee, J. H., & Lee-Kwang, H. (1999). Distributed and
cooperative fuzzy controllers for traffic Intersections
REFERENCES group. IEEE Transactions on Systems, Man, and
Cybernetics, Part C (Applications and Reviews), 29(2),
[1]. Mahmood, T., Ali, M. E. M., & Durdu, A. (2019). A two 263-271.
stage fuzzy logic adaptive traffic signal control for an [7]. Askerzade, I. N., & Mahmood, M. (2010). Control the
isolated intersection based on real data using SUMO extension time of traffic light in single junction by using
simulator. Electron. Commun. Eng, 3, 656-659. fuzzy logic. International Journal of Electrical &
[2]. Papageorgiou, M., Diakaki, C., Dinopoulou, V., Computer Sciences IJECS–IJENS, 10(2), 48-55.
Kotsialos, A., & Wang, Y. (2003). Review of road traffic [8]. Azimirad, E., Pariz, N., & Sistani, M. B. N. (2010). A
control strategies. Proceedings of the IEEE, 91(12), novel fuzzy model and control of single intersection at
2043-2067. urban traffic network. IEEE Systems Journal, 4(1), 107-
[3]. Litman, T. (2017). Evaluating transportation equity. 111.
Victoria, BC, Canada: Victoria Transport Policy [9]. Zaied, A. N. H., & Al Othman, W. (2011). Development
Institute. o f a fuzzy logic traffic system for isolated signalized
[4]. Sommer, C., Tonguz, O. K., & Dressler, F. (2010, intersections in the State of Kuwait. Expert Systems
December). Adaptive beaconing for delay-sensitive and with Applications, 38(8), 9434-9441.
congestion-aware traffic information systems. In 2010 [10]. Prasetiyo, E. E., Wahyunggoro, O., & Sulistyo, S.
IEEE Vehicular Networking Conference (pp. 1-8). IEEE. (2015). Design and simulation of adaptive traffic light

IJISRT25JAN158 www.ijisrt.com 139

Volume 10, Issue 1, January – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.5281/zenodo.14621424

controller using fuzzy logic control sugeno method. [24]. Murat, Y.S., & Gedizlioglu, E. (2005). A fuzzy logic
International Journal of Scientific and Research multi-phased signal control model for isolated junctions.
Publications, 5(4), 1-6. Transportation Research Part C: Emerging Technologies,
[11]. Dereli, T., Cetinkaya, C., & Celik, N. (2018). Desıgnıng 13(1), 19-36, 0968-090X.
a Fuzzy Logıc Controller for a Single Intersectıon: a
Case Study ın Gazıantep. Sigma: Journal of Engineering
& Natural sciences/mühendislik ve Fen Bilimleri
Dergisi, 36(3).
[12]. Gündoğan, F., Karagoz, Z., Kocyigit, N., Karadag, A.,
Ceylan, H., & Murat, Y. Ş. (2014). An evaluation of
adaptive traffic control system in istanbul, turkey.
Journal of Traffic and Logistics Engineering Vol, 2(3).
[13]. Dameri, R. P. (2013). Searching for smart city definition:
a comprehensive proposal. International Journal of
computers & technology, 11(5), 2544-2551.
[14]. Neirotti, P., De Marco, A., Cagliano, A. C., Mangano,
G., & Scorrano, F. (2014). Current trends in Smart City
initiatives: Some stylised facts. Cities, 38, 25-36.
[15]. Meijer, A., & Bolívar, M. P. R. (2016). Governing the
smart city: a review of the literature on smart urban
governance. International review of administrative
sciences, 82(2), 392-408.
[16]. Tainter, J. A., & Taylor, T. G. (2014). Complexity,
problem-solving, sustainability and resilience. Building
Research & Information, 42(2), 168-181.
[17]. Chattaraj, A., Bansal, S., & Prakash, A. (2017). Traffic
Signals: A Review. International Journal of Engineering
Technology Science and Research, 4(7), 204-208.
[18]. Aldarraji, I., Kakei, A., Ismaeel, A. G., Tsaramirsis, G.,
Khan, F. Q., Randhawa, P., ... & Jan, S. (2021). Takagi-
sugeno fuzzy modeling and control for effective robotic
manipulator motion. arXiv preprint arXiv:2112.03006..
[19]. IBRAHEM, S. A., SAHINER, A., & IBRAHIM, A. A.
(2018). Fuzzy logic modeling for prediction of the
nuclear tracks. Journal of Multidisciplinary Modeling
and Optimization, 1(1), 33-40.
[20]. Faris, M. R., Ibrahim, H. M., Abdulrahman, K. Z.,
Othman, L. S., & Marc, K. D. (2021). Fuzzy Logic
Model for Optimal Operation of Darbandikhan
Reservoir, Iraq. Journal homepage: http://iieta.
org/journals/ijdne, 16(4), 335-343.
[21]. Muhammad, A. H., & Akbar, H. S. (2015). Algorithms
for edge detection by using fuzzy logic technique.
Kirkuk University Journal-Scientific Studies, 10(1), 173-
190.
[22]. Krajzewicz, D., Hertkorn, G., Rössel, C., & Wagner, P.
(2002). SUMO (Simulation of Urban MObility)-an open-
source traffic simulation. In Proceedings of the 4th
middle East Symposium on Simulation and Modelling
(MESM20002) (pp. 183-187).
[23]. Behrisch, M., Bieker, L., Erdmann, J., & Krajzewicz, D.
(2011). SUMO–simulation of urban mobility: an
overview. In Proceedings of SIMUL 2011, The Third
International Conference on Advances in System
Simulation. ThinkMind.

IJISRT25JAN158 www.ijisrt.com 140