See discussions, stats, and author profiles for this publication at: [Link]

net/publication/381315472

Optimizing Traffic Flow with Fuzzy Logic and Machine Learning

Article · June 2024

DOI: 10.5281/zenodo.11582010

CITATIONS READS
0 313

2 authors:

Abdurazzag Ali A Aburas Abdullahi Bashi

Beykoz Vocational School of Logistics Beykoz Vocational School of Logistics
33 PUBLICATIONS 145 CITATIONS 1 PUBLICATION 0 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Abdullahi Bashi on 11 June 2024.

The user has requested enhancement of the downloaded file.

 Optimizing Traffic Flow with Fuzzy Logic and
Machine Learning

Abdullahi Ahmed Bashi., Abdurazzag Ali A Aburas,

Emails: Abdullahiahmedbashi@[Link], Abdurazzagaliaaburas@[Link]

Abstract— Fuzzy logic and machine learning present a Significantly enhance traffic flow and minimize
promising method for improving traffic signal systems.
delays.
Traditional traffic management systems often struggle
to adapt to dynamic urban traffic conditions, leading to Machine learning and fuzzy logic controllers (FLCs)
inefficiencies and congestion. This paper introduces a
present promising solutions for the development of
novel adaptive traffic signal control system that
employs machine learning and fuzzy logic. The intelligent traffic signal systems. FLCs are capable of
proposed system optimizes traffic flow and reduces processing intricate and ambiguous traffic data in a
congestion by dynamically adjusting signal timing in
response to real-time traffic conditions. Simulation manner that is comparable to human reasoning,
results using the Simulation of Urban Mobility (SUMO) thereby enabling more responsive signal timing
platform demonstrate significant improvements in
traffic efficiency and reduction in vehicle waiting times. adjustments that are based on real-time conditions.
This adaptability guarantees that traffic signals can
Keyword: adjust to changing traffic volumes and patterns,
Adaptive Traffic control, Fuzzy Logic, Sumo, Machine
Learning. thereby improving the safety and capacity of
intersections. Adaptive traffic control strategies can
be further informed by the identification of patterns
and trends in large datasets by machine learning
1. INTRODUCTION
algorithms. The installment of an intelligent traffic
signal system that integrates machine learning and

T RAFFIC congestion is a substantial impediment

to economic productivity in urban areas
fuzzy logic is the distinctive characteristic of this
project. This system ensures that traffic flows
worldwide, affecting the value of time spent in through intersections are seamless by dynamically
traffic, commuting expenses, and living costs. This changing traffic signals in accordance with the
issue is exacerbated by factors such as an increasing current traffic conditions. The machine learning
number of vehicles, outdated traffic signals, and component improves the efficacy of traffic
inadequate infrastructure. One of the primary causes management over time by optimizing the system's
of congestion is the inadequacy of traditional traffic capacity to adjust its control strategies through
lights, which operate on fixed schedules without continuous learning from traffic patterns.
adapting to fluctuating traffic conditions. Addressing The foundation for addressing the challenges
this issue through adaptive traffic signal systems can associated with ambiguous and uncertain traffic data
is provided by the concept of fuzzy logic, which was
introduced by Lotfi Zadeh in 1965. The optimistic traffic data[1]. The study employed fuzzy logic to
results of early applications of fuzzy logic to traffic model the relationship between traffic flow
management, such as the work of Pappis and parameters and queue lengths, allowing for more
Mamdani in 1977, encouraged further research. As accurate and efficient estimation. By integrating
evidenced by Kelsey and Bisset in 1993 and fuzzy logic with real-time data from traffic sensors,
Nakatsuyama et al. in 1984, subsequent research has the proposed method demonstrated improved
extended the application of fuzzy logic to more performance compared to traditional queue-length
intricate traffic scenarios. The automatic regulation of estimation methods.
signal timings through the use of fuzzy logic was
further advanced by Chiu's 1992 work. The 3.2 Rahman (2009): Review of the Fuzzy Logic
combination of FLCs and machine learning is a Based Approach in Traffic Signal Control: Prospects
highly effective method for circumventing the in Saudi Arabia [2].
constraints of conventional traffic management Rahman conducted a comprehensive review of
systems. Machine learning algorithms analyze existing studies on fuzzy logic-based traffic signal
extensive traffic data to provide insightful predictions control systems, with a focus on their potential
and modifications, while FLCs enhance the detail and application in Saudi Arabia. The study highlighted
adaptability of signal timing. This investigation the advantages of fuzzy logic in addressing the
suggests the implementation of an intelligent traffic uncertainties and complexities of traffic flow in urban
signal system that utilizes these technologies to environments. By analyzing various fuzzy logic-
modify signals in real-time, thereby optimizing based approaches, the study provided insights into
intersection efficiency and mitigating congestion. the prospects of adopting such systems in Saudi
Arabia.
This research endeavors to create an intelligent traffic
management system that significantly enhances the 3.3 Yıldız Budakone, Zülal Hilal, [2023]"Fuzzy
flow and safety of urban traffic by addressing the Logic Based Adaptive Traffic Signal Control for
limitations of conventional traffic signal systems. It Different Scenarios.
does so by utilizing the capabilities of fuzzy logic and Yıldız Budakone proposed an adaptive traffic signal
machine learning to develop a more adaptive and control system based on fuzzy logic to optimize
efficient solution. signal timings for different traffic scenarios [3]. The
study utilized fuzzy logic to optimize signal timings
in real-time according to current traffic
2. LITERATURE REVIEW circumstances, enhancing traffic flow and
minimizing congestion. By considering various
traffic scenarios, the proposed system demonstrated
3.1 Amini. (2016): Queue-Length Estimation Using robust performance across different road conditions
Real-Time Traffic Data and traffic patterns.
Amini proposed a method for estimating queue
lengths at signalized intersections using real-time
3.4 Abdelkader Chabchoub. (2021): Intelligent leading to improved traffic flow and reduced
Traffic Light Controller Using Fuzzy Logic and congestion.
Image Processing.
Abdelkader Chabchoub. Designed a sophisticated 3.7 Suhail M. Odeh (2015): Hybrid Algorithm: Fuzzy
traffic light controller that used fuzzy logic and Logic-Genetic Algorithm on Traffic Light Intelligent
image processing methods to enhance signal timings.. System Suhail M. Odeh developed a hybrid
The study utilized fuzzy logic to analyze traffic flow algorithm combining fuzzy logic with genetic
parameters extracted from video feeds and adjust algorithms for intelligent traffic light control [7]. The
signal timings accordingly [4]. The integration of study aimed to optimize signal timings using a
image processing with fuzzy logic in the suggested combination of fuzzy logic-based decision-making
controller resulted in enhanced precision and and genetic algorithm-based optimization. By
effectiveness in traffic signal regulation. iteratively adjusting signal timings based on feedback
3.5 Victor Manuel Madrigal Arteaga, José Roberto from traffic sensors, the proposed hybrid algorithm
Pérez Cruz, Antonio Hurtado-Beltrán and Jan achieved improved performance compared to
Trumpold. (2022). A Study on Pedestrian Safety at traditional traffic light control methods.
Signalized Intersections Using Fuzzy Logic.
The study investigated pedestrian safety at signalized 3.8 Ria Yuliani Kartikasari. (2020): Optimization of
intersections using fuzzy logic-based control systems Traffic Light Control Using Fuzzy Logic Sugeno
[5]. By incorporating fuzzy logic into pedestrian Method
detection and signal timing algorithms, the proposed Ria Yuliani Kartikasari proposed an optimization
system aimed to reduce pedestrian-vehicle conflicts method for traffic light control using the Sugeno
and enhance safety. The study highlighted the fuzzy logic approach. The study utilized fuzzy logic
potential of fuzzy logic in addressing pedestrian to model the relationships between traffic flow
safety concerns and improving the overall efficiency parameters and signal timings, allowing for adaptive
of signalized intersections. and efficient control [8]. By optimizing fuzzy logic
rules and parameters, the proposed method achieved
3.6 Mario Collotta. (2015): A Novel Approach for better performance in traffic signal control, leading to
Dynamic Traffic Lights Management based on reduced delays and improved traffic flow.
Wireless Sensor Networks and multiple fuzzy logic
controllers Mario Collotta. Proposed a novel 3.9 Roxanne Hawi. (2017): Smart Traffic Light
approach for dynamic traffic light management using Control Using Fuzzy Logic and Wireless Sensor
wireless sensor networks and multiple fuzzy logic Network Roxanne Hawi. Developed a smart traffic
controllers. The study utilized fuzzy logic to integrate light control system using fuzzy logic and wireless
data from wireless sensors and optimize signal sensor networks [9]. The study integrated data from
timings in real-time. By employing multiple fuzzy wireless sensors with fuzzy logic-based control
logic controllers, the proposed approach achieved algorithms to optimize signal timings in real-time. By
more robust and adaptive traffic signal control, leveraging wireless sensor networks and fuzzy logic,
the proposed system achieved more adaptive and signal timings, allowing for adaptive and efficient
efficient traffic signal control, leading to improved control [12]. By simulating the proposed controller in
traffic flow and reduced congestion. various traffic scenarios, the study demonstrated the
effectiveness of fuzzy logic-based control systems in
3.10 I. Adam. (2014): Adaptive Fuzzy Logic Traffic improving traffic management and safety.
Light Management System
I. Adam proposed an adaptive traffic light
management system based on fuzzy logic to optimize 2. PROBLEM STATEMENT
signal timings for varying traffic conditions [10]. The
study utilized fuzzy logic to adjust signal timings
dynamically based on real-time traffic data, Urban traffic congestion is caused by a combination

improving traffic flow and reducing delays. By of more vehicles on the road and insufficient

incorporating adaptive fuzzy logic control infrastructure. This puts a lot of pressure on

algorithms, the proposed system demonstrated robust traditional traffic signal systems, making flow

performance across different traffic scenarios. problems even worse. Fuzzy logic and machine

3.11 M. Fatih Adak; Musa Balta. (2021). Fuzzy learning offer solutions by adaptively modifying

Logic-based Adaptive Traffic Light Control of an signal timings in intelligent traffic lights. By

Intersection: A Case Study. integrating these technologies, traffic management

M. Fatih Adak. Presented a case study on the systems can adjust to intricate urban traffic situations,

implementation of adaptive traffic signal enhancing the efficiency and safety of traffic flow.

management using fuzzy logic at an intersection [11]. Fuzzy logic controllers facilitate intelligent

The study implemented fuzzy logic control engagement with road users by analyzing imprecise

algorithms to adjust signal timings based on real-time inputs, while machine learning detects traffic

traffic data collected from sensors. By optimizing congestion trends to enable timely actions.

fuzzy logic rules and parameters, the proposed Nevertheless, achieving broad implementation of

system achieved improved performance in traffic these sophisticated systems still appears to be a

signal control, leading to reduced congestion and distant objective. Current systems frequently fail to

delays at the intersection. recognize the potential collaboration between fuzzy

logic and machine learning, which restricts their

3.12 Erwan Eko Prasetiyo et al. (2015): Design and ability to effectively handle the intricate nature of

Simulation of Adaptive Traffic Light Controller urban traffic. This project endeavors to create an

Using Fuzzy Logic Control Sugeno Method intelligent traffic signal system that use fuzzy logic

Erwan Eko Prasetiyo et al. designed and simulated an and machine learning to dynamically optimize signal

adaptive traffic light controller using the Sugeno timing. The suggested system aims to mitigate

fuzzy logic control method. congestion, optimize traffic flow, and increase urban

The study employed fuzzy logic to model the safety by integrating various strategies, thereby

relationships between traffic flow parameters and addressing the changing complexities of
contemporary urban transportation.
 The system proceeds to the next simulation after
assessing and modifying the traffic conditions. The
3. METHODOLOGY
simulation continues until it reaches 5000 steps, at
which point it adequately tests and validates the
effectiveness of all traffic control strategies.
3.1 Proposed System

The proposed traffic light control system is designed

to enhance the control of traffic, especially within
urban areas, and it’s developed using SUMO
simulation software and Python. This system also
regulates traffic, ensuring less congestion while at the
same time providing access to emergency vehicles.
The process begins with the start of the SUMO
simulation, which generates traffic data. Some of this
data includes the number of vehicles in each lane, the
time the vehicle waited in the lane, and the
identification of an emergency vehicle on the road.
The system checks for emergency vehicles on the
road first. The system initiates the 'Emergency
Vehicle Handling' procedure if it detects an
emergency vehicle. This protocol helps to change the
operation of the traffic lights to give way to
emergency vehicles crossing the intersection within
the shortest time possible.
If there are no emergency vehicles, the total number
of vehicles in the red and green zones is checked. The
system extends the duration of the green light display
to alleviate congestion in lanes with high traffic
turnover. On the other hand, if the green lane has less
traffic, the system reduces the time for which the
green light remains on and switches to the red lanes
with a heavy inflow to reduce congestion.
The system also evaluates different waiting times for
Figure 1. Proposed System Flowchart
vehicles. The system recalculates the green light's
The flowchart shows the process of simulating a
waiting time if it exceeds the defined limit, ensuring
traffic light. The simulation begins by acquiring
a more efficient flow of traffic. Otherwise, the
traffic data. The system checks for emergency
current light timings remain intact to ensure the free
vehicles, and if any are present, handles them
flow of traffic.
accordingly. If no emergency vehicles are detected, it
evaluates the vehicle count on red lanes and, if
needed, extends the green light for these lanes. It also
assesses the vehicle count on green lanes and, if
necessary, shortens the green light duration. If neither (1)
condition is met, it assesses the waiting times and Where:
determines whether to maintain the existing length of Y is the crisp value.
the traffic signal. Upon implementing the decisions, μY change(y) and μY maintain(y) are the degrees of
the program verifies whether the simulation has membership of Y in the fuzzy sets "change" and
executed for a total of 5000 steps. If it has, the "maintain," respectively.
program terminates. If not, the process continues.
The centroid value obtained indicates a single
representative value for the traffic light phase,
3.2 Mathematical Approach in Fuzzy Logic indicating the optimal timing for switching between
In the field of traffic management, the application of "change" and "maintain" actions. This value is then
fuzzy logic-grounded traffic control sets has gained used to determine the duration of each phase in the
traction due to their ability to create complex traffic light cycle, ensuring efficient traffic flow and
decision rules that effectively address arising minimizing congestion.
uncertainties and inaccurate data. In that regard,
defuzzification is an integral part of these systems
because it transforms fuzzy output sets into precise 3.3 Existing System
values that are traffic signals' control directives. Fuzzy logic merged with Webster and modified
This study seeks to provide a mathematical Webster formulae are added to an existing traffic
representation of defuzzification in the context of signal regulation approach in a study done in 2021 by
traffic control, employing the centroid technique. To Muzamil El Tejani Mohammed Ali, Akif Durdu,
get a more accurate mathematical representation of Seyit Alperen Celtek, and Alper Yilmaz. The study
the given inputs and outputs, it is important to create seeks to find the best cycle times and green phase
precise fuzzy sets, membership functions, fuzzy timings based on actual traffic circumstances using
rules, and a defuzzification technique. Triangular the SUMO traffic simulator. Adaptive control
membership functions will be implemented. solutions are tested using both simulated traffic
The fuzzy output set will be denoted as Y, with fuzzy scenarios and real-world field data, compared to
sets "change" and "maintain" represented by μY fixed-time and fuzzy logic-based systems.
change(y) and μY maintain(y), respectively.
The centroid C is calculated using equation (1) as Comparing the adaptive procedures to standard
follows: methods, the results show a considerable
improvement in journey times, speed, and average
vehicle delay reduction. Inductive loop detectors are
used in this method to analyze traffic flow and
calculate effective cycle lengths and green times. The
comparison analysis highlights the adaptive 3. EXPERIMENT & RESULTS
techniques' advantages over the current fixed-time
and fuzzy logic control technologies, particularly in
situations with high traffic congestion. The proposed traffic control system utilizes a
combination of SUMO simulation and fuzzy logic

The study does, however, point out certain control to manage traffic flow efficiently. During the

limitations, such as the effect of transitioning at five- simulation, the system continuously monitors vehicle

phase crossings on average speed, and it mentions movements and traffic light states, calculating

that the suggested approaches could not work as well feedback based on waiting times and queue lengths.

in terms of average velocity under specific traffic The fuzzy traffic controller receives this feedback

demand scenarios. The study recommends employing and uses predefined rules to determine the timing of

neural networks or genetic algorithms to refine fuzzy traffic lights. These rules consider factors such as the

logic memberships in order to lessen these number of vehicles in red and green lanes to

restrictions. Future research and development could determine the appropriate traffic command (low,

take use of these improvements to better maximize medium, or high). Through iterative simulation steps,

the efficacy and performance of adaptive traffic the system dynamically adjusts traffic light signals to

signal control systems [4]. All things considered, the optimize traffic flow and minimize waiting times,

study provides insightful information about particularly prioritizing emergency vehicles. For

improving traffic flow optimization and reducing analysis, we gather data on vehicle movements and

delays at intersections. control decisions. Testing involves verifying the

correctness of control logic, validating the system's
effectiveness in improving traffic flow, and assessing
Case No G1 G2 G3 G4 Co Σ(y_i)
Case 1 60 41 53 28 194 0.881667 its performance and robustness across various traffic
Case 2 23 17 27 11 90 0.745
scenarios. Ultimately,
Case 3 6 8 8 8 42 0.453333
Combined 60 41 53 28 194 0.881667
case The system aims to enhance overall traffic
Real data 9 3 27 13 64 0.636111
Table 1. The optimal cycle and the distribution of management by intelligently adapting traffic light
effective green times for Webster [4]. control in real time based on observed traffic
conditions.
Case No G1 G2 G3 G4 Co Σ(y_i)
Case 1 42 28 38 20 140 0.881667
Case 2 22 17 26 11 88 0.745
Case 3 7 10 10 10 49 0.453333
Combined 42 28 38 20 140 0.881667
case
Real data 10 3 29 14 68 0.636111
Table 2. The optimal cycle and the distribution of
effective green times for modified Webster [4].
 In figure 3 the graph illustrates the total waiting time
for vehicles in each test case using fuzzy logic and
random forest methods. Across all scenarios, fuzzy
logic consistently demonstrates lower waiting times
compared to random forest, indicating its efficiency
in minimizing overall vehicle wait times.

Figure 2. Simulation of 4 way intersection

Figure 4. Comparison of Emergency Vehicle

Waiting time
In figure 2 the simulation of a 4-way intersection
involves replicating the behavior and interactions of Figure 4 shows the waiting time for emergency
vehicles at a junction where four roads intersect. vehicles in each test case, comparing the fuzzy logic
and random forest techniques. Fuzzy logic regularly
Utilizing mathematical models and algorithms, outperforms random forest by constantly maintaining
vehicles are simulated approaching, navigating, and lower emergency vehicle waiting times in all cases,
departing the intersection. Factors such as traffic indicating its reliability in prioritizing urgent vehicle
movements.
flow, vehicle speeds, signaling, and right-of-way
rules are incorporated to mimic real-world conditions
accurately. 5.1 Results

The evaluation of the traffic control system's

performance is conducted by considering key
metrics, including the overall waiting time for
vehicles, the waiting time for emergency vehicles, the
total feedback received from vehicles, and the
number of emergency vehicles handled. The system's
efficacy in mitigating congestion and enhancing
traffic flow is evaluated using these parameters.

Figure 3. Comparison of Total Vehicle Waiting time

Table 3 Fuzzy Logic Test Case

Test Total EMV Total Total Total the total vehicle waiting time was 10,212,957.0,
Case Vehicle Waiting Feedback EMV Vehicles
which is considerably higher than that of the Fuzzy
Waiting Time
Logic approach. This suggests that while Random
Time
1 561293.0 12338.0 19197 1415 5000 Forest excels in handling complex datasets and large
2 1122690.0 30355.0 38518 2425 10000 traffic volumes, it may not always be the optimal
3 1667623.0 47003.0 57595 3736 15000 choice for minimizing waiting times in real-time
4 2754181.0 146599.0 92273 6439 50000
scenarios.
In real-time applications, the ability to make prompt,
The test cases (1-4) in table 3 of Fuzzy Logic transparent, and understandable decisions is critical.
illustrate different degrees of performance in the While Random Forest offers adaptability and
management of traffic issues. Although the proposed robustness in handling high-dimensional data, Fuzzy
approach suggests that Fuzzy Logic excels at Logic provides clear advantages in terms of
minimizing waiting times, a crucial factor in real- interpretability, targeted prioritization of emergency
time systems where promptness is essential. Test vehicles, and transparency. These attributes make
Case 4 demonstrates lower waiting periods and the Fuzzy Logic particularly well-suited for real-time
accumulation of feedback, indicating that Fuzzy traffic management systems, where the need for
Logic's interpretability and rule-based nature make it quick, reliable, and comprehensible decision-making
suitable for scenarios where human-like reasoning processes is paramount.
and transparency are crucial. This aspect is
particularly beneficial in environments where 5.2 Analysis
decisions need to be easily understood and justified.
Real Data / Fuzzy Logic Program
Webster Real Data: G1 = 9, G2 = 3, G3 = 27, G4 =
13; Σ (y_i) = 0.636111
Table 4 Random Forest Test case Modified Webster Real Data: G1 = 10, G2 = 3, G3 =
Test Total EMV Total Total Total
29, G4 = 14; Σ(y_i) = 0.636111
Feedback Fuzzy Logic Program Test Case 4:
Case Vehicle Waiting EMV Vehicles
Total Vehicle Waiting Time: 2754181.0
Waiting Time
EMV Waiting Time: 146599.0
Time
Observation: Even with large-scale data, fuzzy logic
1 2043376.0 76225 1415 5000
30536.0 demonstrates better handling of traffic with lower
2 4115188.0 52886.0 151399 2425 10000
waiting times.
3 6166601.0 80224.0 228268 3736 15000
4 10212957.0 146599.0 399124 6439 50000 Based on the comparison table 1, 2, 3, it seems that
fuzzy logic manages traffic flow more effectively
than the Webster and modified Webster approaches.
Significantly reduced EMV and overall vehicle
The test cases (1-4) in Table 4 of the Random Forest waiting times across all test instances serve as
approach demonstrate significant improvements in indicators of this. The green time allocations in the
fuzzy logic.
managing feedback accumulation compared to Fuzzy
Logic. However, it often results in higher total
vehicle waiting times. For instance, in Test Case 4,
 6. CONCLUSION Prospects in Saudi Arabia," IEEE Intell.
Transp. Syst. Conf., Available online, Dec.
1, 2009. DOI: 10.1109/ITSC.2016.7795752.
[3] Z. H. Yıldız Budakone, "Fuzzy logic based
In this study, we conducted a comprehensive
investigation of two traffic management approaches: adaptive traffic signal control for different
fuzzy logic and random forest. We examined the scenarios," 2023. [Online]. Available:
performance of these strategies in optimizing traffic
[Link]
flow and minimizing wait times for both total
vehicles and emergency vehicles using simulations 4153407_Fuzzy_Logic_Based_Adaptive_Tr
with the SUMO traffic simulator and Python affic_Signal_Control_for_Different_Scenari
programming language.
os. [Accessed: Jun. 11, 2024].
Our findings show that that the fuzzy logic-driven
system routinely surpasses methods, such Webster [4] A. Chabchoub, A. Hamouda, S. Al-Ahmadi,
and modified Webster approaches, especially when it
and A. Cherif, "Intelligent traffic light
comes to reducing wait times for both emergency and
general vehicles. Fuzzy logic system's adaptiveness controller using fuzzy logic and image
guarantees responsiveness and effective traffic processing," 2021. DOI:
management, emphasizing its fit for real-time
10.14569/IJACSA.2021.0120450.
applications where interpretability and transparency
are essential.

In summary, fuzzy logic offers a strong answer to [5] V. M. Madrigal Arteaga, J. R. Pérez Cruz,
today's traffic management problems. The capacity of
the suggested system to dynamically adapt to A. Hurtado-Beltrán, and J. Trumpold, "A
changing traffic conditions improves urban safety study on pedestrian safety at signalized
and efficiency while optimizing traffic flow. This intersections using fuzzy logic," 2022. DOI:
suggests a promising direction for intelligent
10.3390/app12126024.
transportation systems in the future.

[6] S. M. Odeh, "Hybrid algorithm: Fuzzy

logic-genetic algorithm on traffic light
REFERENCES
intelligent system," 2015. DOI:
10.1109/FUZZ-IEEE.2015.7338117.

[1] Z. Amini, R. Pedarsani, A. Skabardonis, and

P. Varaiya, "Queue-length estimation using [7] R. Y. Mario Collotta, L. L. Bello, and G.
real-time traffic data," in Proc. IEEE Intell. Pau, "A novel approach for dynamic traffic
Transp. Syst. Conf., Rio de Janeiro, Brazil, lights management based on Wireless
Nov. 2016, pp. 517-522. DOI: Sensor Networks and multiple fuzzy logic
10.1109/ITSC.2016.7795752. controllers," 2015. DOI:
10.1016/[Link].2015.02.011.
[2] S. M. Rahman, "Review of the fuzzy logic
based approach in traffic signal control:
 [8] G. P. Kartikasari, D. Pradeka, "Optimization [14] I. Moumen, J. Abouchabaka, and N.
of traffic light control using fuzzy logic Rafalia, "Adaptive traffic lights based on
Sugeno method," 2020. DOI: traffic flow prediction using machine
10.47194/ijgor.v1i2.37. learning models," International Journal of
Electrical and Computer Engineering, vol.
[9] R. Hawi, G. Okeyo, and M. Kimwele, 13, no. 5, pp. 5813-5823, 2023. DOI:
"Smart traffic light control using fuzzy logic 10.11591/ijece.v13i5.pp5813-5823.
and wireless sensor network," 2014. DOI:
10.1109/SAI.2017.8252137.

[10] I. Adam, A. Wahab, M. Yaakop, A. Abdul

Salam, and Z. Zaharudin, "Adaptive fuzzy
logic traffic light management system,"
2014. DOI: 10.1109/ICE2T.2014.7006274.

[11] M. F. Adak and M. Balta, "Fuzzy logic-

based adaptive traffic light control of an
intersection: A case study," 2021. DOI:
10.1109/ISMSIT52890.2021.9604556.

[12] E. E. Prasetiyo, O. Wahyunggoro, and S.

Sulistyo, "Design and simulation of adaptive
traffic light controller using fuzzy logic
control Sugeno method," 2015. [Online].
Available:
[Link]
8835093_Design_and_Simulation_of_Adapt
ive_Traffic_Light_Controller_Using_Fuzzy
_Logic_Control_Sugeno_Method.

[13] A. Vogel, I. Oremović, R. Šimić, and E.

Ivanjko, "Improving traffic light control by
means of fuzzy logic," 2018. DOI:
10.23919/ELMAR.2018.8534692.

View publication stats