International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 11 Issue: 11 | Nov 2024 [Link] p-ISSN: 2395-0072
Smart Adaptive Traffic Signal Management System
Zarina Begum1, Anand Deshmukh2, Aneesh Deshmukh3, Ayush Gude4, Gautam Warvatkar5,
Shreyas Dhomane6
1Professor, Dept. of Artificial Intelligence and Data Science, BRACT’s Vishwakarma Institute of Technology, Pune,
India
2Student, Dept. of Artificial Intelligence and Data Science, BRACT’s Vishwakarma Institute of Technology, Pune,
India
3Student, Dept. of Artificial Intelligence and Data Science, BRACT’s Vishwakarma Institute of Technology, Pune,
India
4Student, Dept. of Artificial Intelligence and Data Science, BRACT’s Vishwakarma Institute of Technology, Pune,
India
5Student, Dept. of Artificial Intelligence and Data Science, BRACT’s Vishwakarma Institute of Technology, Pune,
India
6Student, Dept. of Artificial Intelligence and Data Science, BRACT’s Vishwakarma Institute of Technology, Pune,
India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - The conventional traffic management system, dynamic nature of modern urban traffic. These conventional
which relies heavily on fixed signal timers and manual control systems are inherently rigid, unable to adapt to fluctuating
by traffic police, often results in significant inefficiencies, traffic volumes throughout the day. Consequently, this
prolonged waiting times, and increased fuel consumption. As rigidity leads to several inefficiencies: prolonged waiting
urban areas continue to grow and vehicle numbers rise, these times for drivers, increased fuel consumption, heightened
issues become more pronounced, leading to heightened levels of air pollution due to idling vehicles, and a general
congestion and reduced transportation efficiency. This decrease in the quality of urban life.
research paper proposes a novel approach to traffic
management that dynamically adjusts signal timings based on Traffic congestion is not merely an inconvenience; it has far-
real-time vehicle counts at intersections. By accurately reaching economic, environmental, and social impacts.
measuring the number of vehicles and using this data to Economically, it results in substantial losses due to wasted
determine the optimal duration of green and red lights, the time and reduced productivity. Environmentally, the
system aims to streamline traffic flow, minimize delays, and continuous idling of vehicles at congested intersections
decrease overall congestion. This method not only enhances contributes significantly to greenhouse gas emissions,
the efficiency of traffic movement but also contributes to exacerbating climate change and deteriorating urban air
environmental sustainability by reducing idle times and quality. Socially, the frustration and stress associated with
subsequent emissions. Our study includes a comprehensive daily commuting in congested areas can lead to adverse
analysis of current traffic problems, the design of the proposed health effects and decreased overall well-being
system, and a discussion on the potential benefits and
Recognizing these issues, there is a pressing need to
challenges of implementation. The proposed solution offers a
innovate beyond traditional traffic management approaches.
practical and immediate improvement over traditional traffic
This paper explores an intelligent traffic control system
control methods, presenting a viable path towards smarter
designed to address the limitations of current methods by
and more efficient urban mobility.
dynamically adjusting signal timings based on real-time
Key Words: Intelligent Traffic Control, Real-time Vehicle vehicle counts. The proposed system aims to optimize traffic
Counting, Signal Timer Optimization, Traffic Congestion, flow through intersections by accurately counting vehicles
Urban Mobility, Traffic Management System, Transportation and adjusting signal timings accordingly. This real-time
Efficiency, Environmental Sustainability. adjustment allows for more efficient traffic movement,
reducing wait times and minimizing congestion.
[Link]
The core of this intelligent system is its ability to collect and
Urbanization and the exponential growth of vehicle process real-time data on vehicle numbers at intersections.
ownership have led to significant challenges in managing By employing straightforward counting mechanisms, the
traffic flow efficiently. Traditional traffic management system can determine the optimal duration for green and red
systems, which rely on fixed signal timings and manual lights, ensuring that traffic is managed more effectively. This
oversight by traffic police, often fall short in addressing the method stands in contrast to traditional fixed-timer systems,
which often operate on outdated assumptions about traffic
© 2024, IRJET | Impact Factor value: 8.315 | ISO 9001:2008 Certified Journal | Page 342
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 11 Issue: 11 | Nov 2024 [Link] p-ISSN: 2395-0072
volumes and patterns. The dynamic nature of the proposed and specific intersection issues, providing insights into the
system allows it to respond swiftly to changes in traffic intricate dynamics of urban traffic management.
density, leading to smoother traffic flow and reduced overall
delays. The paper in [3] discusses Intelligent Transportation
Systems (ITS) designed to alleviate urban congestion from
Furthermore, the implementation of an intelligent traffic rising travel demand and limited roadway capacity. It
control system can significantly contribute to environmental highlights adaptive traffic control systems like SCATS
sustainability. By minimizing idle times at intersections, the (Sydney Coordinated Adaptive Traffic System), which
system reduces fuel consumption and lowers emissions of optimize signal timings in real time to enhance traffic flow
pollutants, contributing to cleaner air and a reduction in the and minimize delays. Although certain studies report
urban carbon footprint. This environmental benefit is a benefits in travel time, fuel consumption, and emissions
critical consideration in the context of global efforts to reduction, evaluations like the one conducted in Troy,
combat climate change and promote sustainable urban Michigan, find no significant differences compared to pre-
development. timed control systems. These mixed findings suggest the
necessity for further research to optimize adaptive traffic
In this paper, we will provide a detailed analysis of the control systems and ensure their efficacy across diverse
current problems associated with traditional traffic urban settings.
management systems, outline the design and operational
principles of the proposed intelligent traffic control system, The study in [4] utilizes data from Los Angeles' Automated
and discuss the potential benefits and challenges associated Traffic Surveillance and Control (ATSAC) system, covering
with its implementation. Our research aims to demonstrate 75 detectors across nine intersections on Olympic
that with the integration of real-time vehicle counting and Boulevard, for simulating the Santa Monica Freeway
adaptive signal timing, cities can achieve a more efficient, Corridor using the INTEGRATION model. The research
sustainable, and driver-friendly traffic management system. focuses on efficient data extraction and integrity verification.
Through this innovation, we envision a future where urban Three data processing programs were developed:
mobility is significantly enhanced, making daily commutes [Link] for data formatting, [Link] for half-hourly
faster, safer, and more environmentally friendly. aggregation, and [Link] for detector data merging.
The analysis confirmed high data reliability, with only one
2. LITERATURE SURVEY detector malfunctioning among 45 assessed, validating the
data for use in traffic simulation models and future
The evaluation presented in [1] examines the SCATS Master performance evaluations.
Isolated (SMI) control for non-coordinated signalized
intersections, comparing its performance to traditional In [5], adaptive traffic control systems (ATCS), such as
vehicle-actuated (VA) control methods through both SCOOT (Split Cycle Offset Optimization Technique) and
simulations and field surveys. To assess effectiveness, SCATS (Sydney Coordinated Adaptive Traffic System), are
various SCATS-like adaptive control algorithms were examined for their ability to reduce delays, stops, and
developed and tested. Results indicate that SMI control journey times. Comparative studies indicate that both
typically achieved lower delays and reduced queue lengths systems generally outperform traditional methods, with
compared to VA control. While field surveys revealed only SCATS often achieving fewer stops. SCOOT employs online
slight improvements due to well-optimized VA settings, SMI optimization, while SCATS relies on heuristic adjustments
control demonstrated a marked advantage in scenarios using feedback. Recent evaluations via VISSIM simulation
involving detector malfunctions. The findings underscore highlight SCATS' efficient implementation of offsets and
SMI control's potential for more efficient traffic management splits under heavy traffic loads. The research identifies the
at isolated intersections. need for improved calibration and performance assessment
in oversaturated conditions to enhance the systems’ overall
The study in [2] analyzes the SCOOT traffic network efficiency.
optimization system, which has been implemented globally
and was specifically evaluated in Nijmegen, Netherlands, to The paper in [6] introduces SURTRAC (Scalable Urban
compare its effectiveness with the existing system. Initiated Traffic Control), a decentralized, real-time adaptive traffic
in July 1991, the project faced delays, completing calibration signal control system aimed at reducing urban congestion.
by September 1993, with data analysis finalized in January SURTRAC integrates traffic control theory with multi-agent
1995. Key metrics, including journey times, stops, and queue planning, allowing intersections to allocate green time
lengths across various transport modes, were assessed. independently based on real-time traffic, communicate with
SCOOT delivered mixed results, showing improvements neighboring signals, and adjust dynamically every few
during morning periods but poorer performance in the seconds. A pilot deployment in Pittsburgh, Pennsylvania,
evenings compared to the previous system, with no demonstrated significant reductions in travel times and
significant overall difference. The study also delves into emissions compared to traditional control systems. The
implementation challenges, such as calibration complexities system’s architecture, featuring a decentralized model,
© 2024, IRJET | Impact Factor value: 8.315 | ISO 9001:2008 Certified Journal | Page 343
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 11 Issue: 11 | Nov 2024 [Link] p-ISSN: 2395-0072
schedule-driven adjustments, and robust communication, patterns. The equations have been shown in the following
shows promise for scalable urban traffic management. figure [Figure 1].
The study in [7] presents an adaptive traffic management
system designed for rapidly urbanizing cities in India. It
proposes an IoT-based approach employing camera sensors
and machine learning algorithms, such as TensorFlow, to
optimize signal timings dynamically based on traffic
conditions. The system prioritizes lanes with higher vehicle
densities, aiming to minimize wait times and congestion.
Techniques like Min Max Fairness, AIMD (Additive Increase
Multiplicative Decrease), and PCA (Principal Component
Analysis) are integrated to enhance traffic efficiency and Figure 1. Equations in IDM
pollution control. The approach addresses urban challenges
by improving emergency response and leveraging advanced The roads in our adaptive traffic signal control system
data analysis for dynamic traffic management simulate the physical and behavioral characteristics of road
segments, managing vehicle dynamics and interactions with
Paper [8] outlines an Intelligent Traffic Management System traffic signals. It initializes with start and end coordinates to
(ITMS) using a Deep-Neuro-Fuzzy model to tackle urban define the road's structure, and utilizes a deque for efficient
traffic congestion. It incorporates the Dijkstra algorithm for vehicle management. Geometric properties such as road
optimal route planning based on road segment weights length and orientation are calculated to support vehicle
calculated by the model. The research uses SUMO movement simulations. Traffic signal integration allows the
(Simulation of Urban Mobility) for simulation and class to control vehicle behavior in response to signal states,
demonstrates the system's effectiveness in real-world managing transitions between moving, slowing down, and
scenarios. The framework integrates IoT sensors and offers a stopping. By continuously updating vehicle positions and
user-friendly GUI for real-time data analysis and decision- responses to traffic signals, the Road class provides a
making. Performance comparisons of routing algorithms in realistic simulation of traffic flow, enabling the evaluation
SUMO show the Deep-Neuro-Fuzzy model's adaptability and and optimization of adaptive signal control strategies. Figure
accuracy, suggesting potential enhancements for future ITMS 2 demonstrates the start of the simulation.
iterations.
The paper in [9] introduces an Intelligent Traffic
Management System leveraging the Internet of Vehicles (IoV)
and Vehicle Ad-hoc Networks (VANETs) to improve signal
efficiency. The system adjusts signal timings dynamically
based on real-time vehicle density, offering a cost-effective
and scalable solution by utilizing existing infrastructure. It
discusses the role of IoV and VANETs in future Intelligent
Transportation Systems (ITS), emphasizing GNSS-based
vehicle positioning. The proposed system enhances traffic
flow, safety, and efficiency, using IoV technology for real-time
urban traffic management.
3. METHODOLOGY
3.1 Creating the Simulation
To realistically simulate vehicle behavior and interactions,
we employ the Intelligent Driver Model (IDM). The IDM is a
car-following model that simulates the acceleration and
Figure 2. Simulation generation
deceleration of individual vehicles based on the distance to
the vehicle in front, the relative speed, and the desired speed 3.2 The Algorithm
of the driver. This model is essential for creating realistic
traffic flow in the simulation as it captures the dynamics of The intersection features four primary directions: East,
real-world driving behavior. The IDM adjusts each vehicle's West, North, and South. Each direction includes three lanes
speed to ensure safe following distances, smooth for incoming traffic and three lanes for outgoing traffic. The
acceleration and deceleration, and efficient lane changes, lanes are designated as direction_in1, direction_in2,
contributing to a more accurate representation of traffic
© 2024, IRJET | Impact Factor value: 8.315 | ISO 9001:2008 Certified Journal | Page 344
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 11 Issue: 11 | Nov 2024 [Link] p-ISSN: 2395-0072
direction_in3 for incoming traffic, and similarly, multiple 1s in a row suggest several potential green light
direction_out1, direction_out2, direction_out3 for outgoing combinations, the system prioritizes the pair with the most
traffic. This configuration allows for a structured flow of traffic. The next steps involve implementing the Pygame
vehicles through the intersection. Vehicles in the first lane setup, initializing Pygame, and creating the display window
(designated as _in1) are restricted to turning right. Vehicles to begin the simulation. A function is defined to accurately
in the second lane (_in2) proceed straight through the draw the lanes and intersection layout on the screen. A
intersection, while vehicles in the third lane (_in3) are vehicle class is created to manage the properties and
designated for left turns. This specific lane designation helps movement of vehicles, implementing movement logic that
to organize traffic flow and reduce potential conflicts within adheres to lane directions. The traffic light control is
the intersection. implemented using the matrix, calculating traffic densities
and determining which lanes receive green lights. Finally, a
The traffic lights at the intersection regulate vehicle flow. main loop is created that continuously updates vehicle
The control system uses a matrix to decide which lanes positions, adjusts traffic lights based on real-time traffic
should receive green lights, based on current traffic densities, and renders the updated intersection and vehicles
densities. This matrix is an 8x8 grid where each row and on the screen. This loop ensures the dynamic and adaptive
column correspond to specific lanes. The values within the nature of the traffic signal control system.
matrix indicate whether a particular pair of lanes
(designated by the respective row and column) can have 3.3 Flowchart
green lights simultaneously. The matrix values are binary,
where a value of 1 indicates that the corresponding lanes can
be green at the same time. For example, if matrix[0][2]
equals 1, this means that westin1 and eastin1 lanes can
simultaneously have green lights. Multiple 1s in a row
suggest several potential green light combinations, but the
system will select the pair of lanes with the highest traffic
density. The system continuously calculates the traffic
density for each lane. Based on these densities, it determines
which two lanes, having the highest traffic, should be
granted green signals to optimize flow and minimize
congestion.
The simulation flow begins by setting up the Pygame
environment and creating the display window, which forms
the foundation for the graphical simulation. A function is
then defined to draw the lanes and intersection layout on the
screen, using lines or rectangles. Each direction includes
three lanes for both incoming and outgoing traffic, accurately
depicting the traffic setup. Vehicle classes and their behavior
are defined according to lane directions (right turn, straight,
left turn), and vehicle positions are updated based on the
status of traffic lights, ensuring realistic vehicle movement
within the simulation. Traffic signal logic utilizes the
provided matrix to determine possible green light pairs,
calculates the number of vehicles in each lane to assess
traffic density, and then determines the two lanes with the
maximum traffic to grant green signals, effectively managing
congestion.
The 8x8 matrix represents allowable green light
combinations for the lanes, with rows and columns
corresponding to specific lanes: Row 0 = Col 0 = westin1
lane, Row 1 = Col 1 = southin1 lane, Row 2 = Col 2 = eastin1 Figure 3. Flowchart
lane, Row 3 = Col 3 = northin1 lane, Row 4 = Col 4 = westin2
lane, Row 5 = Col 5 = southin2 lane, Row 6 = Col 6 = eastin2 4. RESULTS
lane, and Row 7 = Col 7 = northin2 lane. The matrix indicates
which pairs of lanes can be green simultaneously. For In this section, we present the outcomes of our simulation
example, matrix[0][2] = 1 signifies that westin1 and eastin1 experiments. The results highlight congestion levels,
can both have green lights at the same time. Although response times, and the efficiency of our traffic management
© 2024, IRJET | Impact Factor value: 8.315 | ISO 9001:2008 Certified Journal | Page 345
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 11 Issue: 11 | Nov 2024 [Link] p-ISSN: 2395-0072
system. Each figure and table corresponds to specific
observations made during the simulations, providing
insights into the impact of various factors on traffic flow and
emergency vehicle prioritization.
Figure 4 illustrates the scenario at the beginning of the
simulation, with 74 seconds having elapsed. At this point, the
congestion is low, and an emergency vehicle has been
granted passage through the traffic.
Figure 6. At 441 seconds
The vehicle counts at Intersection 1 are presented in Table 1,
showing the number of vehicles waiting at each incoming
road at different time intervals. At 74 seconds, traffic
remains light with minimal congestion, as only one vehicle is
waiting at each of the North, West, and East roads, while the
South road has no waiting vehicles. By 167 seconds, the
vehicle count has slightly increased, with small queues
forming at the South and East roads. At 441 seconds,
congestion rises notably, especially on the North and South
roads, with 15 and 13 vehicles waiting, respectively, while
Figure 4. At 74 seconds
the West road sees no vehicles waiting. As the simulation
Figure 5 depicts the scenario at 167 seconds, where traffic progresses, the traffic volume continues to grow, with
congestion remains low, and no significant queues have significant queues forming at all roads by 600 and 750
formed. seconds, reflecting an increase in overall congestion at the
intersection.
Time North_in East_in South_in West_in
Elapsed
74 1 1 0 1
167 1 1 1 3
441 15 0 13 13
600 20 2 18 22
750 25 4 22 28
Table 1. Readings at Intersection 1
The vehicle counts at Intersection 2 are presented in Table 2,
Figure 5. At 167 seconds
showing the number of vehicles waiting at each incoming
Figure 6 shows the scenario at 441 seconds into the road at different time intervals. At 74 seconds, traffic is light
simulation, where traffic congestion has risen. Despite the with one vehicle waiting at each of the North, West, South,
increased congestion, vehicles continue to flow at regular and East roads. By 167 seconds, the vehicle count has
intervals, and the emergency vehicle is given priority. increased slightly, with two vehicles waiting at the North
road while the other roads remain relatively unaffected. At
441 seconds, congestion rises, particularly at the North and
East roads, where 10 and 14 vehicles are waiting,
respectively, while the South and West roads see moderate
queues. As the simulation continues, the number of waiting
vehicles increases further, with more significant congestion
© 2024, IRJET | Impact Factor value: 8.315 | ISO 9001:2008 Certified Journal | Page 346
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 11 Issue: 11 | Nov 2024 [Link] p-ISSN: 2395-0072
at all roads by 600 and 750 seconds, indicating higher [5] Renjith Soman "Traffic Light Control and Violation
overall traffic volume at the intersection. Detection Using Image Processing”.” IOSR Journal of
Engineering (IOSRJEN), vol. 08, no. 4, 2018, pp. 23-27
Time North_in East_in South_in West_in
Elapsed [6] A. Kanungo, A. Sharma and C. Singla, "Smart traffic lights
switching and traffic density calculation using video
74 1 1 1 1 processing," 2014 Recent Advances in Engineering and
167 2 1 1 1 Computational Sciences (RAECS), Chandigarh, 2014, pp. 1-6,
doi: 10.1109/RAECS.2014.6799542.
441 10 5 7 14
[7] Sabeen Javaid, Ali Sufian, Saima Pervaiz and Mehak
600 15 8 12 18 Tanveer,"Smart Traffic Management System Using Internet
750 20 10 18 25 of Things",IEEE 2018.
[8] F. Tang, Y. Kawamoto, N. Kato, and J. Liu, “Future
Table 2. Readings at Intersection 2 intelligent and secure vehicular network toward 6G:
machine-learning approaches,” Proceedings of the IEEE, vol.
4. CONCLUSION 108, no. 2, pp. 292–307, 2020.
In conclusion, the results of the simulation demonstrate the [9] Y. Ni, L. Cai, J. He et al., “Toward reliable and scalable
impact of varying traffic conditions on congestion levels and internet of vehicles: performance analysis and resource
the efficiency of traffic flow at the two intersections. The data management,” Proceedings of the IEEE, vol. 108, no. 2, pp.
presented reveals that while initial traffic volumes are 324–340, 2020.
manageable, congestion increases significantly over time,
particularly during peak periods. The ability to prioritize [10] H. Zhou, W. Xu, J. Chen, and W. Wang, “Evolutionary
emergency vehicles, even under high congestion, was V2X technologies toward the internet of vehicles: challenges
effectively maintained throughout the simulation. The and opportunities,” Proceedings of the IEEE, vol. 108, no. 2,
findings highlight the importance of dynamic traffic pp. 308–323, 202
management systems that can adapt to changing conditions
and ensure both the smooth flow of vehicles and the timely
passage of emergency vehicles. Further studies and
simulations could refine these models and incorporate
additional factors to enhance the performance and
scalability of traffic management systems in real-world
scenarios.
REFERENCES
[1] [Link], 'Tom Tom World Traffic Index', 2019.
[Online]. Available:
[Link]
[2] Khushi, "Smart Control of Traffic Light System using
Image Processing," 2017 International Conference on
Current Trends in Computer, Electrical, Electronics and
Communication (CTCEEC), Mysore, 2017, pp. 99-103, doi:
10.1109/CTCEEC.2017.8454966.
[3] A. Vogel, I. Oremović, R. Šimić and E. Ivanjko, "Improving
Traffic Light Control by Means of Fuzzy Logic," 2018
International Symposium ELMAR, Zadar, 2018, pp. 51-56,
doi: 10.23919/ELMAR.2018.8534692.
[4] A. A. Zaid, Y. Suhweil and M. A. Yaman, "Smart controlling
for traffic light time," 2017 IEEE Jordan Conference on
Applied Electrical Engineering and Computing Technologies
(AEECT), Aqaba, 2017, pp. 1-5, doi:
10.1109/AEECT.2017.8257768.
© 2024, IRJET | Impact Factor value: 8.315 | ISO 9001:2008 Certified Journal | Page 347