Genel Yayın Yönetmeni / Editor in Chief • C.

Cansın Selin Temana

Kapak & İç Tasarım / Cover & Interior Design • Serüven Yayınevi

Birinci Basım / First Edition • © Aralık 2024

ISBN • 978-625-5955-85-2

© copyright
Bu kitabın yayın hakkı Serüven Yayınevi’ne aittir.
Kaynak gösterilmeden alıntı yapılamaz, izin almadan hiçbir yolla çoğaltılamaz. The
right to publish this book belongs to Serüven Publishing. Citation can not be shown
without the source, reproduced in any way without permission.

Serüven Yayınevi / Serüven Publishing

Türkiye Adres / Turkey Address: Kızılay Mah. Fevzi Çakmak 1. Sokak
Ümit Apt No: 22/A Çankaya/ANKARA
Telefon / Phone: 05437675765
web: www.seruvenyayinevi.com
e-mail: [email protected]

Baskı & Cilt / Printing & Volume

Sertifika / Certificate No: 47083
INTERNATIONAL STUDIES AND EVALUATIONS IN THE FIELD OF

CITY, REGION AND

PLANNING

EDITORS

PROF. DR. SERTAÇ GÜNGOR

 CONTENTS

Chapter 1
INTELLIGENT TRANSPORTATION SYSTEMS: A
SUSTAINABLE AND EFFECTIVE SOLUTION FOR FUTURE
CITIES
İlker ATMACA..................................................................................... 1

Chapter 2

SCENARIO WRITING AS A STRATEGIC TOOL IN URBAN

AND REGIONAL PLANNING
Mehmet Akif SAĞ ............................................................................... 29

Chapter 3
COMPARISON OF TEXT-TO-IMAGE GENERATIVE AI
TOOLS FOR URBAN PORTRAITS: A CASE STUDY OF
INVISIBLE CITIES BY USING STABLE DIFFUSION, DALL-E
AND MIDJOURNEY
Özlem Kevseroğlu............................................................................... 49

Rifat Kurban ...................................................................................... 49

 Chapter 1
INTELLIGENT TRANSPORTATION SYSTEMS: A
SUSTAINABLE AND EFFECTIVE SOLUTION FOR
FUTURE CITIES

İlker ATMACA1

1 Asst. Prof. Dr., Yozgat Bozok University, Faculty of Engineering and Architecture, Depart-
ment of City and Regional Planning, [email protected] ORCID: 0000-0001-9950-
2833
 2 İlker ATMACA

1. INTRODUCTION
Intelligent Transportation Systems (ITS) represent a significant
advancement in urban mobility, integrating various technologies to
enhance the efficiency, safety, and sustainability of transportation networks.
The core of ITS lies in the deployment of connected vehicles (CVs) and
autonomous vehicles (AVs), which utilize Vehicle-to-Everything (V2X)
communication technologies to facilitate real-time interactions between
vehicles and their surroundings. This connectivity is crucial for improving
traffic management, reducing congestion, and enhancing safety by enabling
vehicles to share vital information regarding road conditions, traffic signals,
and potential hazards (Chen et al., 2020; Sumalatha et al., 2019; Liu, 2023).
The implementation of such systems is not merely a technological upgrade;
it signifies a paradigm shift in how urban transportation is conceptualized
and managed.
The impetus for the rise of ITS is largely driven by the challenges posed
by increasing urban populations and the corresponding demand for more
sustainable and efficient mobility solutions. Traditional transportation
systems are often ill-equipped to handle the complexities of modern urban
environments, leading to issues such as traffic congestion, pollution, and
safety concerns (Ning, 2021; Namazi et al., 2019). Intelligent transportation
systems address these challenges by employing advanced technologies,
including the Internet of Things (IoT), artificial intelligence (AI), and edge
computing, to optimize traffic flow and minimize emissions (Liu, 2023;
Zheng et al., 2023). For instance, connected vehicles equipped with sensors
can gather and transmit data about traffic patterns, allowing traffic signals
to adjust dynamically to real-time conditions, which significantly reduces
delays and enhances overall traffic efficiency (Chen et al., 2020; Sumalatha
et al., 2019).
Moreover, the scope of ITS extends beyond vehicles to encompass
a comprehensive ecosystem that includes infrastructure, public
transportation, and pedestrian interactions. The integration of smart
technologies into public transit systems can significantly enhance user
experiences through real-time updates and improved scheduling, thereby
promoting shared mobility solutions (Sumalatha et al., 2019; Chen et al.,
2018). Additionally, the emergence of Urban Air Mobility (UAM) systems,
which aim to incorporate aerial vehicles into the transportation mix,
underscores the potential of ITS to redefine urban mobility paradigms
(Banjanovic-Mehmedovic et al., 2018; Jiao & Huang, 2020). This holistic
approach to transportation management not only improves the functionality
of existing networks but also lays the groundwork for innovative solutions
that can adapt to the evolving needs of urban populations.
 International Studies and Evaluations in the Field of City, Region and Planning 3

One of the primary advantages of ITS is its ability to enhance

transportation quality through improved traffic management. The
integration of information and communication technology (ICT) with
transportation infrastructure enables real-time monitoring and control
of traffic conditions, which is particularly vital for urban areas with high
traffic volumes (Liu, 2023; Su, 2023). This capability allows for the efficient
management of traffic flow, resulting in reduced travel times and lower
vehicle emissions, thus promoting environmental sustainability (Zheng et
al., 2023; Ahanin et al., 2023). For example, ITS can facilitate automatic
accident detection and provide timely information to drivers, significantly
improving road safety (Yang et al., 2015; Namazi et al., 2022).
Furthermore, ITS plays a crucial role in addressing the challenges
posed by urbanization, such as traffic congestion and difficulties in vehicle
tracking. By leveraging emerging technologies, ITS can process vast
amounts of data to identify patterns and predict traffic conditions, enabling
proactive measures to alleviate congestion (Jayaprakash et al., 2023; Zhang,
2014). This predictive capability is essential for modern cities, where the
demand for efficient transportation solutions is continuously increasing.
The application of advanced algorithms and machine learning techniques
further enhances the effectiveness of ITS in managing urban traffic
dynamics (He, 2023; Gulati & Srinivasan, 2019).
In addition to improving traffic management, ITS fosters economic
development by enhancing accessibility and mobility within urban areas.
Efficient transportation systems are critical for facilitating trade and
commerce, ensuring the timely movement of goods and services (Ning,
2021; Namazi et al., 2019). The deployment of ITS can lead to better
transport planning and organization, making transportation systems more
sustainable, accessible, and economically viable (Mane et al., 2023; Guillen-
Perez & Cano, 2021). Moreover, ITS contributes to the overall quality of life
in urban settings by reducing travel times and enhancing the reliability of
public transportation systems (Vallati et al., 2021; Pan et al., 2017).
The multifaceted importance of ITS in contemporary transportation
cannot be overstated. By leveraging advancements in ICT, ITS improves
traffic management, reduces congestion, and enhances safety, thereby
significantly contributing to urban mobility and sustainability (Hamadeh
et al., 2021; Ristama, 2023). As cities continue to grow and evolve, the
integration of intelligent transportation systems will be paramount in
addressing the myriad challenges associated with urbanization.
 4 İlker ATMACA

2. TECHNOLOGICAL FOUNDATIONS OF INTELLIGENT

TRANSPORTATION SYSTEMS
The foundation of ITS is intricately woven with several key
technological components, including the Internet of Things (IoT), data
analytics, artificial intelligence (AI), and communication technologies.
These elements work synergistically to enhance the efficiency, safety, and
sustainability of transportation systems, thereby addressing the myriad
challenges posed by urbanization and increasing mobility demands.
2.1. Internet of Things (IoT) and ITS
The IoT plays a pivotal role in the development of ITS by enabling the
interconnection of various devices and systems. This connectivity allows
for real-time data collection and sharing, which is essential for effective
traffic management and monitoring. IoT devices, such as sensors and
cameras, gather data on traffic flow, vehicle speeds, and environmental
conditions, which can be analyzed to optimize traffic signals and reduce
congestion (Wang et al., 2018). The integration of IoT in ITS not only
improves operational efficiency but also enhances the user experience by
providing real-time information to travelers about traffic conditions and
alternative routes (Sen, 2021). For instance, smart traffic lights can adjust
their timings based on real-time traffic data collected from IoT sensors,
thereby minimizing delays and improving traffic flow (Andrienko et al.,
2017).
Moreover, the IoT facilitates the development of smart infrastructure
that can communicate with vehicles, creating a more cohesive transportation
ecosystem. This interconnectedness allows for the implementation of
advanced traffic management strategies, such as adaptive traffic signal
control, which can dynamically respond to changing traffic conditions (Çelik
et al., 2023). The ability to collect and analyze data from various sources in
real-time enables transportation agencies to make informed decisions that
enhance safety and efficiency on the roads (Abbas et al., 2021). As urban
areas continue to expand, the role of IoT in ITS will become increasingly
critical in managing the complexities of modern transportation networks.
2.2. Data Analytics and Machine Learning
Data analytics serves as another cornerstone of ITS, enabling the
processing and interpretation of vast amounts of data collected from
various sources. Advanced analytics techniques, including machine
learning algorithms, are employed to predict traffic patterns, identify
potential incidents, and optimize routing for vehicles (Hao & Qin, 2020).
For instance, predictive analytics can forecast traffic congestion based on
historical data and current conditions, allowing transportation agencies
 International Studies and Evaluations in the Field of City, Region and Planning 5

to implement proactive measures to mitigate delays (Maciejewski et al.,

2016). The ability to analyze data in real-time enhances decision-making
processes and contributes to more effective traffic management strategies.
Machine learning models can also be utilized to improve the accuracy
of traffic predictions by continuously learning from new data inputs,
thereby refining their algorithms over time (Bhuvana et al., 2023). This
adaptability is crucial in urban environments where traffic conditions can
change rapidly due to various factors, including weather, accidents, and
road construction. By leveraging data analytics, ITS can provide valuable
insights that inform infrastructure investments, traffic policy decisions,
and public transportation planning (Raja et al., 2020). The integration of
data analytics into ITS not only enhances operational efficiency but also
contributes to a more sustainable urban mobility framework.
2.3. Communication Technologies
Effective communication technologies are essential for the seamless
operation of ITS. V2X communication is a critical aspect, allowing vehicles
to communicate with each other and with infrastructure elements such
as traffic signals and road signs (Çelik et al., 2023). This communication
facilitates the exchange of information regarding traffic conditions, hazards,
and navigation assistance, thereby improving overall safety and efficiency
on the roads (Zheng, 2019). Additionally, the use of dedicated short-
range communications (DSRC) and cellular networks ensures that data is
transmitted quickly and reliably, which is vital for real-time applications in
ITS.
The implementation of V2X communication enables vehicles to
receive timely updates about road conditions, traffic signals, and potential
hazards, allowing drivers to make informed decisions (Andrienko et
al., 2017). Furthermore, this communication technology supports the
development of cooperative driving strategies, where vehicles can work
together to optimize traffic flow and reduce congestion (Wang et al., 2018).
As communication technologies continue to evolve, their integration into
ITS will play a crucial role in enhancing the safety and efficiency of urban
transportation systems.
2.4. Advanced Traffic Management Systems (ATMS)
Advanced Traffic Management Systems (ATMS) are integral
components of ITS that utilize advanced technologies to monitor and
manage traffic flow. These systems employ a combination of sensors,
cameras, and data analytics to provide real-time information about traffic
conditions and incidents. By integrating various data sources, ATMS can
optimize traffic signal timings, manage lane usage, and provide timely
 6 İlker ATMACA

alerts to drivers about accidents or road closures. The implementation of

ATMS has been shown to reduce travel times, improve safety, and decrease
emissions by minimizing stop-and-go traffic conditions (Sen, 2021).
ATMS can also facilitate the coordination of traffic management
efforts across different jurisdictions, ensuring a more cohesive approach
to urban mobility. By leveraging real-time data, these systems can adapt
to changing traffic patterns and implement strategies that enhance overall
traffic flow. For example, ATMS can reroute traffic in response to accidents
or road closures, thereby minimizing delays and improving the efficiency of
the transportation network. The continued development and deployment
of ATMS will be essential in addressing the growing challenges of urban
traffic management.
2.5. Autonomous Vehicles and ITS
The advent of autonomous vehicles represents a significant
advancement in the realm of ITS. These vehicles rely on a combination of
sensors, cameras, and AI algorithms to navigate and make decisions in real-
time. The integration of autonomous vehicles into existing transportation
systems has the potential to enhance safety by reducing human error, which
is a leading cause of traffic accidents. Furthermore, autonomous vehicles
can communicate with other vehicles and infrastructure, contributing to
more efficient traffic flow and reduced congestion (Raja et al., 2020).
The deployment of autonomous vehicles can also lead to changes in
urban planning and infrastructure design, as these vehicles may require
different road configurations and traffic management strategies. For
instance, dedicated lanes for autonomous vehicles could be established
to optimize their performance and safety. Additionally, the data collected
from autonomous vehicles can provide valuable insights into traffic patterns
and behaviors, further informing traffic management strategies. As the
technology continues to evolve, the integration of autonomous vehicles
into ITS will play a crucial role in shaping the future of urban mobility.
2.6. Blockchain Technology in ITS
Blockchain technology is emerging as a valuable tool in the context
of ITS, particularly in enhancing security and data integrity. By providing
a decentralized and tamper-proof ledger, blockchain can facilitate
secure transactions and data sharing among various stakeholders in the
transportation ecosystem (Abbas et al., 2021). For example, blockchain
can be used to manage vehicle ownership records, track the maintenance
history of vehicles, and ensure the authenticity of data collected from IoT
devices. This technology not only enhances trust among participants but
also streamlines processes within the transportation sector.
 International Studies and Evaluations in the Field of City, Region and Planning 7

The integration of blockchain with IoT can further enhance the security
of connected vehicles and infrastructure, protecting against potential
cyber threats. Additionally, smart contracts enabled by blockchain can
automate various processes within the transportation ecosystem, such as
toll payments and vehicle registration, thereby improving efficiency and
reducing administrative burdens. As the transportation industry continues
to embrace digitalization, the role of blockchain technology in ITS will
become increasingly significant in ensuring secure and efficient operations.
3. KEY COMPONENTS OF INTELLIGENT TRANSPORTATION
SYSTEMS
The key components of ITS encompass a variety of technologies and
methodologies that work together to create a cohesive and responsive
transportation environment (Figure 1.).

Figure 1. Key components of intelligent transportation systems

3.1. Traffic Management Systems

Traffic management systems are integral components of ITS that utilize
advanced technologies to monitor and manage traffic flow. These systems
employ a combination of sensors, cameras, and data analytics to provide
real-time information about traffic conditions and incidents Tzvetkova
(2020). By integrating various data sources, traffic management systems can
optimize traffic signal timings, manage lane usage, and provide timely alerts
to drivers about accidents or road closures. The implementation of these
systems has been shown to reduce travel times, improve safety, and decrease
emissions by minimizing stop-and-go traffic conditions (Rakhmatulloh et
 8 İlker ATMACA

al., 2021). For instance, real-time data from traffic management systems
can lead to dynamic adjustments in signal timings, which can significantly
alleviate congestion during peak hours. Furthermore, these systems can
facilitate better coordination among different transportation modes,
enhancing the overall efficiency of urban mobility (Geurs et al., 2016).
3.2. Connected Vehicles and V2X
Communication Connected vehicles (CVs) are a cornerstone of
ITS, enabling real-time communication between vehicles and their
environments through Vehicle-to-Everything (V2X) technologies. V2X
communication allows vehicles to interact with each other (V2V), with
infrastructure (V2I), and with pedestrians (V2P) (Buehler et al., 2018).
This connectivity enhances traffic management and safety by enabling
vehicles to share critical information about road conditions, traffic signals,
and potential hazards. For instance, when a vehicle detects an accident
ahead, it can communicate this information to other vehicles, allowing
them to reroute and avoid congestion (Cools et al., 2016). The integration
of V2X communication is essential for the development of autonomous
vehicles, as it enhances situational awareness and decision-making
capabilities (Komarova, 2021). Moreover, V2X communication can also
support emergency vehicle prioritization, ensuring that first responders
can navigate through traffic more efficiently (Corsar et al., 2017).
3.3. Data Analytics and Artificial Intelligence
Data analytics is another critical component of ITS, enabling the
processing and interpretation of vast amounts of data generated by
connected vehicles and infrastructure. Advanced analytics techniques,
including machine learning and AI, are employed to identify patterns,
predict traffic conditions, and optimize routing for vehicles (Hardiyansyah
et al., 2023). For example, AI algorithms can analyze historical traffic
data alongside real-time information to forecast congestion and suggest
alternative routes, thereby minimizing delays and improving overall
traffic efficiency (Cheng & Chen, 2015). The integration of AI in ITS
enhances decision-making processes, allowing transportation agencies to
implement proactive measures based on predictive analytics. Additionally,
AI can facilitate the development of intelligent algorithms that adapt to
changing traffic conditions, further improving the responsiveness of traffic
management systems (Gogolova, 2017).
3.4. Public Transportation Integration
The integration of smart technologies into public transportation
systems is another key component of ITS. This integration enhances user
experience through real-time updates and improved scheduling, while also
 International Studies and Evaluations in the Field of City, Region and Planning 9

promoting the use of shared mobility solutions (Sani, 2020). For instance,
real-time tracking of buses and trains allows passengers to receive accurate
arrival times, reducing wait times and improving overall satisfaction.
Furthermore, effective integration of public transport with other modes of
transportation, such as cycling and walking, can enhance accessibility and
encourage greater use of public transit (Smith et al., 2018). This multimodal
approach not only improves the efficiency of public transport systems but
also contributes to reducing the environmental impact of urban mobility
(Chen, 2023).
3.5. Intelligent Parking Management
Parking management is another key component of ITS. Intelligent
Parking Systems (IPS) are designed to address the challenges associated
with traditional parking management, which often leads to inefficiencies,
increased traffic congestion, and environmental pollution. By utilizing a
combination of sensors, cameras, and communication technologies, IPS
can monitor parking spaces in real-time, providing drivers with accurate
information about available spots (Ewhrudjakpor et al., 2019). This
capability not only improves the efficiency of parking operations but also
enhances the overall user experience by reducing the time spent searching
for parking. Moreover, intelligent parking solutions can integrate with
navigation systems to guide drivers to available spaces, further alleviating
congestion in urban areas (Rifai & Arifin, 2020). The implementation of
such systems can also facilitate dynamic pricing strategies that encourage
the efficient use of parking resources (Yun et al., 2023).
3.6. Infrastructure Development
The physical infrastructure that supports ITS is crucial for its successful
implementation. This includes the development of smart roads equipped
with sensors and communication technologies that facilitate data collection
and sharing (Sochor, 2015). Upgrading and expanding urban infrastructure
to accommodate connected and autonomous vehicles is essential for
maximizing the benefits of ITS. This infrastructure must be designed to
support various modes of transportation, ensuring seamless connectivity
between different systems and enhancing overall mobility (DeRobertis et
al., 2020). Additionally, the integration of renewable energy sources and
sustainable materials in infrastructure development can contribute to the
environmental sustainability of transportation systems (Nag et al., 2019).
As cities evolve, the continuous investment in infrastructure will be vital
to support the growing demands of urban mobility and the integration of
advanced transportation technologies.
10 İlker ATMACA

3.7. User-Centric Design and Public Acceptance

The success of ITS also hinges on user acceptance and the design
of systems that prioritize user experience. Engaging with stakeholders,
including the public, during the design and implementation phases is
essential for addressing concerns and ensuring that the systems meet the
needs of users (Merkert et al., 2020). Public education campaigns can help
raise awareness about the benefits of ITS, fostering a positive perception
and encouraging adoption (Errampalli et al., 2020).
Moreover, incorporating user feedback into the design process can
lead to more intuitive and accessible systems, ultimately enhancing user
satisfaction and promoting greater use of public transportation and other
ITS features (Hu et al., 2020). The emphasis on user-centric design is crucial
in ensuring that ITS solutions are not only technologically advanced but
also socially acceptable and beneficial for all users.
4. APPLICATIONS OF INTELLIGENT TRANSPORTATION
SYSTEMS
The applications of ITS address various challenges faced by urban
environments, including traffic congestion, safety concerns, and
environmental impacts.
4.1. Traffic Management and Control
One of the primary applications of ITS is traffic management and
control. Intelligent Traffic Management Systems utilize real-time data from
various sources, such as sensors and cameras, to monitor traffic conditions
and optimize traffic flow (Al-Abassi & Al-Jameel, 2021). By analyzing this
data, traffic management agencies can implement adaptive signal control
systems that adjust traffic light timings based on current traffic volumes,
thereby reducing congestion and improving overall traffic efficiency (Bıyık
et al., 2021). Additionally, incident detection systems can quickly identify
accidents or breakdowns, allowing for rapid response and minimizing
disruptions to traffic flow (Lile, 2023). The effectiveness of these systems
is evidenced by studies showing significant reductions in travel times and
improvements in safety metrics in cities where such technologies have been
implemented (Roman et al., 2018).
4.2. Public Transportation Optimization
ITS plays a crucial role in optimizing public transportation systems. By
integrating real-time tracking and scheduling technologies, public transit
agencies can provide passengers with accurate information about arrival
times and service availability (Al Harbi et al., 2017). This enhances the user
experience and encourages greater use of public transportation, which can
 International Studies and Evaluations in the Field of City, Region and Planning 11

help alleviate traffic congestion and reduce emissions. Furthermore, ITS

can facilitate the coordination of different modes of transportation, such as
buses, trains, and rideshare services, creating a seamless travel experience
for users (Ge, 2023). The integration of public transport with other mobility
solutions not only enhances accessibility but also promotes a shift from
private vehicle use to more sustainable transport options (Alarbi, 2023).
4.3. Smart Parking Solutions
IPS are another significant application of ITS. These systems utilize
sensors and mobile applications to provide real-time information about
parking availability, allowing drivers to locate and reserve parking spaces
before arriving at their destination (Buldakov, 2020). By reducing the time
spent searching for parking, smart parking solutions can decrease traffic
congestion and improve air quality in urban areas. Additionally, dynamic
pricing models can be implemented to optimize the use of parking
resources, encouraging turnover and maximizing revenue for parking
facilities (Iacobescu et al., 2021). Research indicates that smart parking
systems can lead to a more efficient allocation of parking spaces, ultimately
enhancing the overall urban mobility experience (Joshi et al., 2021).
4.4. Vehicle-to-Everything (V2X) Communication
V2X communication is a transformative application of ITS that enables
vehicles to communicate with each other, infrastructure, and pedestrians.
This connectivity enhances safety by allowing vehicles to share information
about road conditions, traffic signals, and potential hazards (Rizwan et al.,
2019). For instance, if a vehicle detects an impending collision, it can alert
nearby vehicles to take evasive action, thereby preventing accidents. V2X
communication also supports the development of autonomous vehicles, as
it provides critical information for navigation and decision-making (Hirai
& Murase, 2020). The integration of V2X technologies is essential for
creating safer and more efficient transportation networks, particularly in
urban settings where traffic density is high (El-Jamassi et al., 2022).
4.5. Environmental Monitoring and Management
ITS applications extend beyond traffic management to include
environmental monitoring and management. By integrating data from
various sources, such as air quality sensors and traffic flow data, ITS
can provide insights into the environmental impacts of transportation
systems This information can be used to develop strategies for reducing
emissions, promoting sustainable transportation options, and improving
urban air quality. For example, cities can implement congestion pricing or
promote the use of electric vehicles based on data-driven insights from ITS
(Petrov et al., 2018). The ability to monitor environmental conditions in
12 İlker ATMACA

real-time allows urban planners to make informed decisions that enhance

sustainability and public health (Pham et al., 2015).
4.6. Emergency Response and Management
ITS also play a vital role in emergency response and management. By
utilizing real-time data and communication technologies, ITS can facilitate
the rapid deployment of emergency services to incidents, improving
response times and enhancing public safety (Banode, 2023). For instance,
traffic signal preemption systems can prioritize the movement of emergency
vehicles at intersections, allowing them to reach their destinations more
quickly. Additionally, ITS can provide real-time information to the public
during emergencies, helping to guide evacuation efforts and inform citizens
about potential hazards (Otchere & Opare, 2019). The integration of ITS
in emergency management not only improves the efficiency of response
efforts but also enhances community resilience in the face of disasters
(Shawly et al., 2022).
4.7.Data-Driven Urban Planning
The integration of ITS into urban planning processes is becoming
increasingly important. By analyzing traffic patterns, travel behavior, and
environmental data, urban planners can make informed decisions about
transportation infrastructure and land use (Barriga et al., 2020). This
data-driven approach allows for the development of more sustainable and
efficient transportation systems that meet the needs of growing urban
populations. Additionally, ITS can support the implementation of smart
city initiatives, enhancing the overall livability and functionality of urban
environments (Rizvi et al., 2019). The insights gained from ITS data can
inform policies that promote equitable access to transportation and support
economic development.
5. CASE STUDIES AND GLOBAL PERSPECTIVES OF
INTELLIGENT TRANSPORTATION SYSTEMS
ITS have been implemented in various forms across the globe,
showcasing their versatility and effectiveness in addressing transportation
challenges.
5.1. Case Study: Singapore’s Smart Traffic Management
Singapore is often cited as a leading example of ITS implementation.
The city-state has developed a comprehensive smart traffic management
system that integrates various technologies to optimize traffic flow and
reduce congestion (Figure 2.). The system utilizes a network of sensors,
cameras, and data analytics to monitor real-time traffic conditions across
the city (Kim & Hong, 2016).
 International Studies and Evaluations in the Field of City, Region and Planning 13

Figure 2. ITS Vision for Singapore [URL 1]

One of the key features of Singapore’s ITS is its Electronic Road Pricing
(ERP) system, which employs dynamic tolling based on real-time traffic
conditions. This system adjusts toll rates according to traffic demand,
encouraging drivers to use alternative routes during peak hours (Abood et
al., 2023). The integration of Vehicle-to-Everything (V2X) communication
allows vehicles to receive real-time updates about traffic conditions, further
enhancing the efficiency of the transportation network. The success of
Singapore’s ITS can be attributed to its focus on data-driven decision-
making and public engagement. The government actively involves citizens
in the planning process, ensuring that the system meets the needs of the
community while promoting sustainable urban mobility (Ahn et al., 2018).
5.2. Case Study: Barcelona’s Smart Parking Solutions
Barcelona has implemented an Intelligent Parking System (IPS) that
utilizes IoT technology to optimize parking resource utilization in the city.
The system employs sensors installed in parking spaces to detect vehicle
occupancy and transmit real-time information to a central management
system (Figure 3.) (Naik et al., 2019).

Figure 3. IPS for Barcelona [URL 2]

14 İlker ATMACA

Drivers can access this information through a mobile application,

which provides guidance to available parking spots, reducing the time spent
searching for parking. The IPS also incorporates dynamic pricing, adjusting
parking fees based on demand to encourage turnover and maximize revenue
for parking facilities (Khalid, 2024). The implementation of Barcelona’s IPS
has led to a significant reduction in traffic congestion and emissions, as
drivers no longer need to circle the streets in search of parking. This case
study highlights the importance of integrating smart technologies into
urban infrastructure to create more efficient and sustainable transportation
systems (Blinova, 2024).
5.3. Global Perspective: ITS in Developing Countries
While many ITS implementations are concentrated in developed
countries, there is a growing interest in adopting intelligent transportation
solutions in developing nations. For instance, countries like India and
Brazil are exploring the potential of ITS to address urban mobility
challenges, such as traffic congestion and inadequate public transportation
systems (Naderpour et al., 2017). In India, cities like Delhi have begun
implementing ITS technologies to improve traffic management and
enhance public transportation services. The Delhi Traffic Police has
introduced a centralized traffic management system that utilizes real-time
data to monitor traffic conditions and respond to incidents more effectively
(Jameel et al., 2021). Similarly, Brazil’s Smart Mobility program aims to
integrate ITS technologies into urban transportation planning, focusing on
improving public transit efficiency and reducing emissions. The program
emphasizes the importance of collaboration between government agencies,
private sector stakeholders, and citizens to create sustainable transportation
solutions (Guizar et al., 2020).
5.4. Emerging Trends and Future Directions
As ITS continues to evolve, several emerging trends are shaping the
future of intelligent transportation systems. The integration of AI and
machine learning is expected to enhance the capabilities of ITS, enabling
more sophisticated data analysis and predictive modeling (Tal et al., 2013).
Additionally, the rise of connected and autonomous vehicles presents
new opportunities for ITS implementation. These vehicles rely on V2X
communication to interact with their environment, improving safety and
efficiency on the roads (Sangeetha et al. 2024). Moreover, the adoption
of blockchain technology in ITS is gaining traction, particularly in
enhancing data security and privacy. Blockchain can facilitate secure data
sharing among various stakeholders, ensuring the integrity of information
exchanged within the transportation network (Kwon et al., 2021). The
convergence of these technologies promises to create more resilient and
 International Studies and Evaluations in the Field of City, Region and Planning 15

adaptive transportation systems capable of meeting the challenges of urban

mobility in the future.
6. FUTURE TRENDS IN INTELLIGENT TRANSPORT
SYSTEMS
ITS are rapidly evolving, driven by advancements in technology,
increasing urbanization, and the need for sustainable transportation
solutions. As cities around the world grapple with challenges such as traffic
congestion, pollution, and safety concerns, the future of ITS holds significant
promise for enhancing urban mobility. This section explores key future
trends in ITS, highlighting emerging technologies, innovative applications,
and the potential for transformative change in transportation networks.
6.1. Integration of Artificial Intelligence and Machine Learning
AI and machine learning are set to play a pivotal role in the future
of ITS. These technologies enable advanced data analysis and predictive
modeling, allowing transportation agencies to make informed decisions
based on real-time data Hoang (2020). For instance, AI algorithms can
analyze traffic patterns, weather conditions, and historical data to predict
congestion and optimize traffic flow. This predictive capability will enhance
traffic management systems, allowing for dynamic adjustments to traffic
signals and routing strategies based on current conditions (Savin, 2021).
Moreover, AI can facilitate the development of autonomous vehicles (AVs)
that rely on real-time data from their environment to navigate safely
and efficiently. As AV technology matures, its integration into existing
transportation systems will require sophisticated AI-driven solutions to
ensure seamless operation and safety.
6.2. Expansion of V2X Communication
The future of ITS will see a significant expansion of V2X communication
technologies. V2X enables vehicles to communicate with each other,
infrastructure, and pedestrians, creating a connected ecosystem that
enhances safety and efficiency (Ibanez et al., 2018). As the deployment of
5G networks continues to grow, V2X communication will become more
reliable and widespread, allowing for real-time data exchange and improved
situational awareness for drivers and automated systems (Mustakim, 2020).
This connectivity will facilitate advanced applications such as cooperative
adaptive cruise control, where vehicles can communicate to maintain
optimal distances and speeds, reducing the likelihood of collisions and
improving traffic flow (Azizalrahman & Hasyimi, 2020). Furthermore, V2X
communication will support the integration of smart traffic management
systems that can adapt to changing conditions based on real-time data
from connected vehicles.
16 İlker ATMACA

6.3. Emphasis on Sustainable Transportation Solutions

Sustainability will be a key focus in the future development of ITS.
As urban populations continue to grow, the demand for efficient and
environmentally friendly transportation solutions will increase. IPS and
smart public transportation solutions will play a crucial role in promoting
sustainable urban mobility (Maksimchuk et al., 2022). For example, IPS
can reduce the time spent searching for parking, thereby decreasing vehicle
emissions and congestion (Razizadeh, 2023). Additionally, the integration
of electric and hybrid vehicles into ITS will contribute to reducing the
carbon footprint of urban transportation. As cities implement policies to
promote electric vehicle (EV) adoption, ITS will facilitate the development
of charging infrastructure and optimize routing for EVs to ensure efficient
energy use (Javed et al., 2020).
6.4. Enhanced Cybersecurity Measures
As ITS becomes increasingly reliant on data exchange and connectivity,
cybersecurity will be a critical concern. The potential vulnerabilities
associated with connected vehicles and infrastructure necessitate robust
security measures to protect sensitive data and ensure the integrity of
transportation systems (Khurana et al., 2019). Future ITS will need to
incorporate advanced cybersecurity protocols, including encryption,
authentication, and intrusion detection systems, to safeguard against
cyber threats (Chang, 2023). Moreover, the use of blockchain technology
is being explored as a means to enhance data security and privacy within
ITS. Blockchain can provide a decentralized and tamper-proof ledger for
transactions and data sharing, ensuring the authenticity and integrity of
information exchanged among various stakeholders (Riki et al., 2020).
6.5. Smart Infrastructure Development
The future of ITS will also involve the development of smart
infrastructure that integrates advanced technologies to support connected
and autonomous vehicles. This includes the installation of sensors, cameras,
and communication devices in roadways, traffic signals, and parking
facilities to facilitate real-time data collection and sharing (Maaroufi et al.,
2021). Smart infrastructure will enable more efficient traffic management,
enhance safety, and improve the overall user experience. Additionally,
the concept of “smart cities” will drive the integration of ITS with urban
planning and development. By incorporating ITS into the design of
urban spaces, cities can create more efficient transportation networks that
prioritize accessibility, sustainability, and safety (Scarinci et al., 2017).
 International Studies and Evaluations in the Field of City, Region and Planning 17

6.6. Focus on User-Centric Solutions

User-centric design will be increasingly important in the development
of ITS. Engaging with stakeholders, including the public, during the design
and implementation phases is essential for ensuring that systems meet the
needs of users (Premkumar & Saminadan, 2015). Future ITS will prioritize
user experience by providing intuitive interfaces, real-time information,
and personalized services that enhance the overall travel experience. Public
education campaigns will also play a crucial role in fostering acceptance
and encouraging the adoption of ITS technologies. By raising awareness
about the benefits of intelligent transportation solutions, cities can promote
greater participation in smart mobility initiatives (Guido et al., 2022).
7. CONCLUSION
In conclusion, ITS represent a paradigm shift in how we approach
the management and operation of transportation networks. By leveraging
advanced technologies such as connectivity, data analytics, AI, and
automation, ITS enhances the efficiency, safety, and sustainability of urban
mobility. The integration of these technologies addresses the pressing
challenges posed by increasing urbanization, traffic congestion, and
environmental concerns, ultimately leading to smarter and more responsive
transportation systems.
The key components of ITS, including connected vehicles, traffic
management systems, smart parking solutions, and public transportation
integration, work synergistically to optimize transportation networks.
For instance, connected vehicles equipped with V2X communication
capabilities allow for real-time data exchange, enhancing situational
awareness and enabling proactive traffic management. Traffic management
systems utilize data from various sources to monitor and control traffic flow,
reducing delays and improving safety at intersections. Similarly, intelligent
parking systems streamline the parking experience, minimizing the time
drivers spend searching for available spaces and reducing congestion in
urban areas.
Case studies from cities like Singapore and Barcelona illustrate the
successful implementation of ITS and its potential to transform urban
mobility. Singapore’s comprehensive traffic management system, which
includes dynamic tolling through its Electronic Road Pricing (ERP) system,
effectively manages traffic demand and encourages the efficient use of road
space. Barcelona’s Intelligent Parking System, which utilizes IoT technology
to provide real-time parking availability information, has significantly
improved the user experience and reduced traffic congestion. These
examples highlight the effectiveness of ITS in enhancing transportation
efficiency and promoting sustainable urban development.
18 İlker ATMACA

Moreover, the global perspective on ITS reveals a growing interest in

adopting intelligent transportation solutions in developing countries. As
urban populations continue to rise, cities in regions such as India and Brazil
are exploring the potential of ITS to address their unique transportation
challenges. By implementing ITS technologies, these cities can improve
traffic management, enhance public transportation services, and promote
sustainable mobility solutions, ultimately contributing to economic
development and improved quality of life for residents.
Looking ahead, several future trends are poised to shape the
evolution of ITS. The integration of AI and machine learning will enable
more sophisticated data analysis and predictive modeling, allowing
transportation agencies to make informed decisions based on real-time
data. The expansion of V2X communication technologies, supported
by the rollout of 5G networks, will enhance connectivity and facilitate
the development of autonomous vehicles, further improving safety and
efficiency on the roads. Additionally, the emphasis on sustainability will
drive the adoption of electric and hybrid vehicles within ITS, contributing
to reduced emissions and a smaller carbon footprint.
However, the successful implementation of ITS is not without
challenges. Issues related to infrastructure costs, data privacy, and
cybersecurity must be addressed to ensure the effective deployment of
intelligent transportation solutions. As ITS continues to evolve, it will be
essential for stakeholders including government agencies, private sector
partners, and the public to collaborate and engage in the planning and
implementation processes. Public education campaigns will also play a
crucial role in fostering acceptance and encouraging the adoption of ITS
technologies.
In summary, Intelligent Transportation Systems hold immense
potential to revolutionize urban mobility and improve the quality of life
for citizens. By harnessing the power of connectivity, data analytics, and
emerging technologies, ITS can create more efficient, safe, and sustainable
transportation networks. As research and development in this field continue
to advance, the future of ITS promises innovative solutions that can adapt
to the evolving needs of urban populations, ultimately paving the way for
smarter, more resilient cities.
 International Studies and Evaluations in the Field of City, Region and Planning 19

REFERENCES

Abbas, K., Tawalbeh, L., Rafiq, A., Muthanna, A., Elgendy, I., & El‐Latif, A. (2021).
Convergence of blockchain and iot for secure transportation systems in
smart cities. Security and Communication Networks, 2021, 1-13.

Abood, M., Wang, H., He, D., Kang, Z., & Kawoya, A. (2023). Intelligent network
slicing in v2x networks – a comprehensive review. Journal of Artificial
Intelligence and Technology.

Ahanin, F., Mustapha, N., Zolkepli, M., & Husin, N. (2023). A review of traffic
state prediction (tsp) methods in intelligent transportation systems (its).
International Journal of Academic Research in Business and Social
Sciences, 13(3).

Ahn, J., Kim, Y., & Kim, R. (2018). A novel wlan vehicle-to-anything (v2x)
channel access scheme for ieee 802.11p-based next-generation connected
car networks. Applied Sciences, 8(11), 2112.

Al-Abassi, S. & Al-Jameel, H. (2021). Design and study of a smart park system:
university of kufa as a case study. Kufa Journal of Engineering, 9(4), 128-
145.

Alarbi, M. (2023). Scope: smart cooperative parking environment. Ieee Access,

11, 120346-120361.

Al Harbi, A., Alotaibi, B., Baatya, M., Jastania, Z., & Meccawy, M. (2017). A
smart parking solution for jeddah city. International Journal of Computer
Applications, 171(7), 4-9.

Andrienko, G., Chen, W., Maciejewski, R., & Zhao, Y. (2017). Visual analytics
of mobility and transportation: state of the art and further research
directions. Ieee Transactions on Intelligent Transportation Systems,
18(8), 2232-2249.

Azizalrahman, H. & Hasyimi, V. (2020). Towards a generic framework for smart

cities. IntechOpen.

Banjanovic-Mehmedovic, L., Butigan, I., Mehmedovic, F., & Kantardzic,

M. (2018). Hybrid automaton based vehicle platoon modelling and
cooperation behaviour profile prediction. Tehnicki Vjesnik - Technical
Gazette, 25(3).

Banode, M. (2023). Smart parking system using mern stack. International

Journal for Research in Applied Science and Engineering Technology,
11(5), 2811-2814.
20 İlker ATMACA

Barriga, J., Sulca, J., León, J., Ulloa, A., Portero, D., García, J., & Yoo, S. (2020).
A smart parking solution architecture based on lorawan and kubernetes.
Applied Sciences, 10(13), 4674.

Bhuvana, J., Hashmi, H., Adhvaryu, R., Kashyap, S., Kumari, S., & Wadhwa, D.
(2023). Intelligent analytics algorithms in breach detection systems for
securing vanets and data for smart transportation management.

Bıyık, C., Allam, Z., Pieri, G., Moroni, D., Fraifer, M., O’Connell, E., & Khalid,
M. (2021). Smart parking systems: reviewing the literature, architecture
and ways forward. Smart Cities, 4(2), 623-642.

Blinova, T. (2024). Data-intensive traffic management: real-time insights from

the traffic management simulation test. Bio Web of Conferences, 86,
01089.

Buehler, R., Pucher, J., & Dümmler, O. (2018). Verkehrsverbund: the evolution and
spread of fully integrated regional public transport in germany, austria,
and switzerland. International Journal of Sustainable Transportation,
13(1), 36-50.

Buldakov, N. (2020). An open source solution for smart contract-based parking.

16th International Conference on Open Source Systems, Innopolis,
Russia. 55-69.

Chang, C. (2023). An infrastructure management humanistic approach for

smart cities development, evolution, and sustainability. Infrastructures,
8(9), 127.

Chen, L. (2023). Research on smart mobility in public transportation and

solutions for mobility. Highlights in Science Engineering and Technology,
56, 498-505.

Chen, X., Chen, C., Cai, Y., Wang, H., & Ye, Q. (2018). Kernel sparse representation
with hybrid regularization for on-road traffic sensor data imputation.
Sensors, 18(9), 2884.

Chen, X., Lu, J., Zhao, J., Qu, Z., Yang, Y., & Xian, J. (2020). Traffic flow prediction
at varied time scales via ensemble empirical mode decomposition and
artificial neural network. Sustainability, 12(9), 3678.

Cheng, Y. & Chen, S. (2015). Perceived accessibility, mobility, and connectivity

of public transportation systems. Transportation Research Part a Policy
and Practice, 77, 386-403.

Cools, M., Fabbro, Y., & Bellemans, T. (2016). Free public transport: a socio-
cognitive analysis. Transportation Research Part a Policy and Practice,
86, 96-107.
 International Studies and Evaluations in the Field of City, Region and Planning 21

Corsar, D., Edwards, P., Nelson, J., Baillie, C., Papangelis, K., & Velaga, N. (2017).
Linking open data and the crowd for real-time passenger information.
SSRN Electronic Journal.

Çelik, Y., Aydın, M., Petri, I., & Potoglou, D. (2023). Intelligent transportation
systems applications: safety and transfer of big transport data.,
Transbaltica XIII: Transportation Science and Technology, 59-73.

DeRobertis, M., Ferrell, C., Lee, R., & Eells, J. (2020). Characteristics of effective
metropolitan areawide public transit: a comparison of european,
canadian, and australian case studies.

Di, H. (2021). Risk management in the construction of intelligent transportation

platform based on artificial intelligence. Converter, 543-549.

El-Jamassi, A., Al-Sultan, H., & El-Saidy, A. (2022). A smart parking system‫‏‬,
case study. Iraqi Journal of Civil Engineering, 15(2), 97-103.

Errampalli, M., Patil, K., & Prasad, C. (2020). Evaluation of integration between
public transportation modes by developing sustainability index for
indian cities. Case Studies on Transport Policy, 8(1), 180-187.

Ewhrudjakpor, O., Poliakova, A., & Poliak, M. (2019). The integration of

passenger transport and integration barriers. The Archives of Automotive
Engineering - Archiwum Motoryzacji, 85(3), 103-113.

Ge, L. (2023). Landscape architecture design and implementation based on

intelligent monitoring sensing network. Journal of Sensors, 2023(1).

Geurs, K., Puello, L., & Weperen, S. (2016). A multi-modal network approach to
model public transport accessibility impacts of bicycle-train integration
policies. European Transport Research Review, 8(4).

Gogolova, M. (2017). Methodological proposal of profit distribution among

the participants of integrated transport system. Universal Journal of
Management, 5(3), 133-138.

Guido, G., Haghshenas, S., Haghshenas, S., Vitale, A., Gallelli, V., & Astarita,
V. (2022). Prioritizing the potential smartification measures by using an
integrated decision support system with sustainable development goals (a
case study in southern italy). Safety, 8(2), 35.

Guillen-Perez, A. and Cano, M. (2021). Intelligent iot systems for traffic

management: a practical application. Iet Intelligent Transport Systems,
15(2), 273-285.

Guizar, A., Mannoni, V., Poli, F., Denis, B., & Berg, V. (2020). Lte-v2x performance
evaluation for cooperative collision avoidance (coca) systems. IEEE
92nd Vehicular Technology Conference (VTC2020-Fall), Victoria, BC,
Canada, 2020, 1-5.
22 İlker ATMACA

Gulati, I. & Srinivasan, R. (2019). Image processing in intelligent traffic

management. International Journal of Recent Technology and
Engineering, 8(2S4), 213-218.

Hamadeh, N., Karouni, A., Farhat, Z., Ghor, H., Ghor, M., & Katea, I. (2021).
Intelligent transportation systems to mitigate road traffic congestion.
Intelligenza Artificiale. 15(2):91-104.

Hao, Q. & Qin, L. (2020). The design of intelligent transportation video processing
system in big data environment. Ieee Access, 8, 13769-13780.

Hardiyansyah, B., Sutikno, F., Zeibots, M., & Wijayaratna, K. (2023). Evaluation
of accessibility commuter-line station with jitney transportation
modes by using gis application-a case study of depok station. Civil and
Environmental Science, 006(01), 43-55.

He, C. (2023). Optimization of urban intelligent transportation travel service

system based on feature extraction and traffic prediction. Applied
Mathematics and Nonlinear Sciences, 9(1).

Hirai, T. & Murase, T. (2020). Performance evaluations of pc5-based cellular-

v2x mode 4 for feasibility analysis of driver assistance systems with crash
warning. Sensors, 20(10), 2950.

Hoang, C. (2020). Analytical support for logistics processes and transport

management in smart city. Civil Engineering and Architecture, 8(6),
1390-1394.

Hu, M., Lin, H., Wang, J., Xu, C., Tatem, A., Meng, B., & Lai, S. (2020). Risk
of coronavirus disease 2019 transmission in train passengers: an
epidemiological and modeling study. Clinical Infectious Diseases, 72(4),
604-610.

Iacobescu, C., Oltean, G., Florea, C., & Burtea, B. (2021). Unified interplanetary
smart parking network for maximum end-user flexibility. Sensors, 22(1),
221.

Ibanez, J., Zeadally, S., & Contreras-Castillo, J. (2018). Sensor technologies for
intelligent transportation systems. Sensors, 18(4), 1212.

Jameel, F., Javed, M., Zeadally, S., & Jantti, R. (2021). Efficient mining cluster
selection for blockchain-based cellular v2x communications. Ieee
Transactions on Intelligent Transportation Systems, 22(7), 4064-4072.

Javed, A., Malhi, A., & Framling, K. (2020). Edge computing-based fault-tolerant
framework: a case study on vehicular networks. 2020 International
Wireless Communications and Mobile Computing (IWCMC), Limassol,
Cyprus, 1541-1548.

Jayaprakash, P. O., Babu, M. M., Reddy B. D. & Lessly S. H. (2023). An intelligent

traffic system based on internet of vehicles using adaptive traffic
 International Studies and Evaluations in the Field of City, Region and Planning 23

management algorithm. International Journal of Intelligent Engineering

and Systems, 16(6), 838-849.

Jiao, Z. & Huang, M. (2020). Traffic flow forecast through time series analysis
based on deep learning. Ieee Access, 8, 82562-82570.

Joshi, N., Mittal, M., Jayawant, Y., Salunkhe, P., & Hattarge, M. (2021). Applying
systems engineering framework for architecting a smart parking system
within a smart city. Incose International Symposium, 31(1), 16-30.

Khalid, I. (2024). Optimizing hybrid v2x communication: an intelligent

technology selection algorithm using 5g, c-v2x pc5 and dsrc. Future
Internet, 16(4), 107.

Khurana, N., Mittal, S., Piplai, A., & Joshi, A. (2019). Preventing poisoning
attacks on ai based threat intelligence systems. IEEE 29th International
Workshop on Machine Learning for Signal Processing, Pittsburgh, PA,
USA, 1-6.

Kim, J. & Hong, S. (2016). Study on the privacy-preserving vehicular pki in

autonomous driving environments. Contemporary Engineering Sciences,
9, 619-625.

Komarova, О. (2021). Methodological aspects of the institutional analysis of the

level of regional infrastructure development. E3s Web of Conferences,
296, 03009.

Kwon, S., Seo, J., Yun, J., & Kim, K. (2021). Driving behavior classification and
sharing system using cnn-lstm approaches and v2x communication.
Applied Sciences, 11(21), 10420.

Lile, D. (2023). Empowering recreation in urban parks using digital solutions: a

proposed framework for smart urban parks in the city of tirana.

Liu, Y. (2023). Swin-apt: an enhancing swin-transformer adaptor for intelligent

transportation. Applied Sciences, 13(24), 13226.

Maaroufi, M., Stour, L., & Agoumi, A. (2021). Striving for smart mobility in
morocco: a case of lanes designated to heavy goods vehicles in casablanca.
Engineering Management in Production and Services, 13(1), 74-88.

Maciejewski, M., Bischoff, J., & Nagel, K. (2016). An assignment-based approach

to efficient real-time city-scale taxi dispatching. Ieee Intelligent Systems,
31(1), 68-77.

Maksimchuk, O., Borisova, Н., Panov, D., & Voronkova, G. V. (2022). Sustainable
development of road transport infrastructure by including elements of
reverse traffic in the “smart city” program. Iop Conference Series Earth
and Environmental Science, 988(2), 022072.
24 İlker ATMACA

Mane, D., Kumbharkar, P., Earan, N., Patil, K., Bonde, S., & Uke, N. (2023).
A research survey on real-time intelligent traffic system. International
Journal of Emerging Technology and Advanced Engineering, 13(4), 36-
51.

Merkert, R., Bushell, J., & Beck, M. (2020). Collaboration as a service (caas) to
fully integrate public transportation – lessons from long distance travel
to reimagine mobility as a service. Transportation Research Part a Policy
and Practice, 131, 267-282.

Mustakim, H. (2020). 5g vehicular network for smart vehicles in smart city: a

review. Journal of Computer Electronic and Telecommunication, 1(1).

Naderpour, M., Meskanen, T., Paverd, A., & Niemi, V. (2017). Auditable de-
anonymization in v2x communication. Journal of Ict Standardization,
5(1), 51-106.

Nag, D., Bs, M., & Goswami, A. (2019). Framework for public transport
integration at railway stations and its implications for quality of life.
SSRN Electronic Journal.

Naik, G., Choudhury, B., & Park, J. (2019). Ieee 802.11bd & 5g nr v2x:
evolution of radio access technologies for v2x communications. Ieee
Access, 7, 70169-70184.

Namazi, E., Li, J., & Lu, C. (2019). Intelligent intersection management systems
considering autonomous vehicles: a systematic literature review. Ieee
Access, 7, 91946-91965.

Namazi, E., Mester, R., Lu, C., & Li, J. (2022). Geolocation estimation of
target vehicles using image processing and geometric computation.
Neurocomputing, 499, 35-46.

Ning, S. (2021). Intelligent transportation system planning in the age of artificial

intelligence. E3s Web of Conferences, 253, 01036.

Otchere, P. & Opare, W. (2019). Design and construction of an authorized car

parking alert system. Saudi Journal of Engineering and Technology,
04(12), 522-534.

Pan, H., Huang, Z., & He, Z. (2017). Research on architecture of intelligent
transportation cloud platform for guangxi expressway. 5th international
conference on computer-aided design, manufacturing, modeling and
simulation (cdmms 2017).

Petrov, V., Andreev, S., Gerla, M., & Koucheryavy, Y. (2018). Breaking the limits
in urban video monitoring: massive crowd sourced surveillance over
vehicles. Ieee Wireless Communications, 25(5), 104-112.
 International Studies and Evaluations in the Field of City, Region and Planning 25

Pham, T., Tsai, M., Nguyen, D., Dow, C., & Deng, D. (2015). A cloud-based smart-
parking system based on internet-of-things technologies. Ieee Access, 3,
1581-1591.

Premkumar, S. & Saminadan, V. (2015). Performance analysis of wlan-wimax

smart distribution grid. International Journal of Engineering Research
And, V4(08).

Raja, G., Manaswini, Y., Vivekanandan, G., Sampath, H., Dev, K., & Bashir,
A. (2020). Ai-powered blockchain - a decentralized secure multiparty
computation protocol for iov, IEEE Conference on Computer
Communications Workshops (INFOCOM WKSHPS), Toronto, ON,
Canada, 865-870.

Rakhmatulloh, A., Dewi, D., & Andini, T. (2021). Route integration of semarang
trans feeder for the city of semarang and surroundings. Astonjadro, 10(2),
249.

Razizadeh, S. (2023). The adoption of artificial intelligence-based systems in

the development of public transportation in auckland. Rere Āwhio - The
Journal of Applied Research and Practice, (3), 19-25.

Rifai, A. & Arifin, F. (2020). Analysis of the level of passenger satisfaction with
services and transport facilities-based integration in jakarta. Journal of
World Conference (Jwc), 2(2), 66-73.

Riki, R., Widyarto, S., Kahar, S., & Hariyanto, S. (2020). Smart city comprehension
in community development. Proceedings of the Proceedings of the Third
Workshop on Multidisciplinary and Its Applications, WMA-3 2019, 11-14
December 2019.

Ristama, H. (2023). The utilization of intelligent traffic systems for managing

traffic problems in tourism areas: a literature review. Deviance Jurnal
Kriminologi, 7(1),

Rizvi, S., Zehra, S., & Olariu, S. (2019). Aspire: an agent-oriented smart parking
recommendation system for smart cities. Ieee Intelligent Transportation
Systems Magazine, 11(4), 48-61.

Rizwan, M., Khalid, N., Ahmed, M., & Gilani, A. (2019). Park my ride: your
true parking companion. Lahore Garrison University Research Journal
of Computer Science and Information Technology, 3(1), 57-62.

Roman, C., Liao, R., Ball, P., Ou, S., & Heaver, M. (2018). Detecting on-street
parking spaces in smart cities: performance evaluation of fixed and
mobile sensing systems. Ieee Transactions on Intelligent Transportation
Systems, 19(7), 2234-2245.
26 İlker ATMACA

Sani, N. (2020). Integration of unimap’s student bus routing towards free city-
bus service local people in perlis. Quantum Journal of Social Sciences and
Humanities, 1(5), 55-68.

Savin, G. (2021). Information support of the flow processes in the smart

city transport and logistics system: methodology basis. E3s Web of
Conferences, 270, 01017.

Scarinci, R., Markov, I., & Bierlaire, M. (2017). Network design of a transport
system based on accelerating moving walkways. Transportation Research
Part C Emerging Technologies, 80, 310-328.

Sen, P. (2021). Optimization of traffic flow using intelligent transportation

systems. Mathematical Statistician and Engineering Applications, 70(1),
720-727.

Shawly, T., Alsheikhy, A., Said, Y., & Lahza, H. (2022). An effective approach
for smart parking management. Ingenierie Des Systèmes D Information,
27(5), 783-789.

Sangeetha, R. G., Hemanth, C., Dipesh, R., Samriddhi, K., Venetha, S., Abbas
Alif, M., Arjun, S. & Varshithram, K. S. (2024). Density Based Real-
time Smart Traffic Management System along with Emergency Vehicle
Detection for Smart Cities. Int. J. ITS Res. 22, 328–338.

Smith, G., Sochor, J., & Karlsson, M. (2018). Mobility as a service: development
scenarios and implications for public transport. Research in
Transportation Economics, 69, 592-599.

Sochor, J. (2015). Implementing mobility as a service. Transportation Research

Record Journal of the Transportation Research Board, 2536(1), 1-9.

Su, X. (2023). A latent-factor-model-based approach for traffic data imputation

with road network information. Isprs International Journal of Geo-
Information, 12(9), 378.

Sumalatha, V., Dingari, M., & Jayalakshmi, P. (2019). Short term road traffic
flow forecasting using artificial neural network. International Journal of
Recent Technology and Engineering, 8(3), 7998-8000.

Tal, I., Zhu, T., & Muntean, G. (2013). Short paper: on the potential of v2x
communications in helping electric bicycles saving energy. IEEE
Vehicular Networking Conference, Boston, MA, 218-221.

Tzvetkova, S. (2020). Increasing the social effectiveness of public transport in

the city of sofia. International Journal for Quality Research, 14(1), 79-92.

Vallati, M., Scala, E., & Chrpa, L. (2021). A hybrid automated planning approach
for urban real-time routing of connected vehicles. IEEE International
Intelligent Transportation Systems Conference (ITSC), Indianapolis, IN,
USA, 3821-3826.
 International Studies and Evaluations in the Field of City, Region and Planning 27

Wang, J., Jiang, C., Han, Z., Ren, Y., & Hanzo, L. (2018). Internet of vehicles:
sensing-aided transportation information collection and diffusion. Ieee
Transactions on Vehicular Technology, 67(5), 3813-3825.

Yang, H., Dillon, T., & Chen, Y. (2015). Evaluation of recent computational
approaches in short-term traffic forecasting., 108-116.

Yun, H., Lee, E., Moon, S., & Kim, D. (2023). Data-driven approach for
measuring and managing physical distancing in subways during
pandemic conditions. Transportation Research Record Journal of the
Transportation Research Board, 2678(5), 588-600.

Zhang, X. (2014). Path reconstruction of intelligent traffic based on positive

feedback system. Applied Mechanics and Materials, 513-517, 3160-3164.

Zheng, S., Wang, J., Zhang, Y., Yin, H., Hua, H., & Zhong, Y. (2023). Locality-
aware spatial-temporal attention neural network with contrast learning
for traffic flow prediction. Proc. SPIE 12708, 3rd International Conference
on Internet of Things and Smart City.

Zheng, Z. (2019). Blockchain intelligence: when blockchain meets artificial

intelligence. Cryptography and Security, ArXiv:1912.06485.

Internet References

URL1:https://www.lta.gov.sg/content/dam/ltagov/getting_around/driving_in_
singapore/intelligent_transport_systems/pdf/smartmobility2030.pdf

URL2:http://www.libelium.com/saving-water-with-smart-irrigation-system-in-
barcelona/
 Chapter 2
SCENARIO WRITING AS A STRATEGIC TOOL IN
URBAN AND REGIONAL PLANNING

Mehmet Akif SAĞ1

1 Assistant Professor, Konya Technical University, Faculty of Architecture and Design,

Department of Urban and Regional Planning, Konya, Turkey, [email protected], ORCID:
0000-0002-6669-4012"
30 Mehmet Akif SAĞ

INTRODUCTION
The future is fraught with uncertainties for urban planning. Factors
such as economic fluctuations, demographic shifts, technological
innovations, and environmental challenges make it difficult to accurately
predict the shape of cities to come. Despite these uncertainties, strategic
spatial planning requires flexible and comprehensive approaches to help
cities achieve their long-term goals. Scenario writing plays a critical role in
addressing these challenges within urban and regional planning processes.
More than merely analyzing current conditions, scenario writing
enables the development of strategic predictions about potential
opportunities and threats. Planners use this method to navigate
uncertainties by combining different types of scenarios, normative,
exploratory, and predictive with forecast techniques such as the Delphi
method, trend analysis, and STEEP/PESTLE. Additionally, templates
like Four Square Scenarios, Cross-Impact Matrix, Timeline Scenarios,
and Stakeholder Scenarios serve as essential tools for data analysis and
scenario structuring.
This paper aims to provide a comprehensive understanding of scenario
writing in both theoretical and practical contexts. By systematically
discussing the main scenario types, forecast methods, and scenario
templates, it seeks to highlight the importance of scenario writing in
managing uncertainty and supporting planners’ decision-making
processes.
CHAPTER ONE: MAIN SCENARIO TYPES
1. Normative Scenarios
Normative scenarios are created to realize a desired future state or
goal (Börjeson, Höjer, Dreborg, Ekvall, & Finnveden, 2006). Developed
in alignment with a specific vision, policy, or objective, these scenarios
provide strategic guidance on how to achieve an ideal outcome in planning
processes (Godet, 2000). Their defining characteristic is their forward-
looking nature. rather than describing current conditions, they outline
what we aspire to achieve. For instance, a city might create a normative
scenario to envision a carbon-neutral future, identifying actionable steps
to reduce emissions and promote renewable energy. Such scenarios are
particularly useful in aligning efforts with long-term strategic objectives
(Van Notten, Rotmans, Van Asselt, & Rothman, 2003).
In urban and regional planning, normative scenarios contribute to
the development of local and regional development strategies in line with
long-term goals (Albrechts, 2004). For example, specific goals, such as
a city becoming carbon neutral by 2050, are shaped through normative
 International Studies and Evaluations in the Field of City, Region and Planning 31

scenarios (Bulkeley & Betsill, 2005). These scenarios guide planners and
decision makers in determining the necessary steps to achieve the desired
future and increase the effectiveness of strategic planning (Hopkins &
Zapata, 2007).
2. Exploratory Scenarios
Exploratory scenarios aim to investigate possible future situations
based on current trends and conditions (Börjeson et al., 2006). Developed
to explore uncertainties and risks, these scenarios address the question of
what outcomes might occur under varying future conditions (Van Notten
et al., 2003). The exploratory approach adopts an open-ended perspective,
enabling predictions about potential futures by analyzing trends and
variables (Godet, 2000).
In urban and regional planning, exploratory scenarios identify risks
and opportunities, allowing planners to develop proactive strategies for
managing uncertainty (Rotmans et al., 2000). For instance, by analyzing
factors like population growth, economic development, and climate
change, planners can envision how these elements might shape a region
over the next 20 years. This approach helps planners anticipate potential
challenges, such as infrastructure demands, or capitalize on opportunities,
such as sustainable economic growth (Glenn & Gordon, 2009). Ultimately,
exploratory scenarios equip planners to respond strategically to future
changes and uncertainties (Wack, 1985).
3. Predictive Scenarios
Predictive scenarios forecast future developments based on current
data and past trends (Börjeson et al., 2006). Their primary aim is to predict
the future as accurately as possible. These scenarios are typically created
using analytical models and statistical data, operating on the assumption
that current conditions will persist (Godet & Roubelat, 1996). By examining
historical and present data, predictive scenarios allow planners to make
data-driven forecasts and more precise decisions in the planning process
(Bradfield, Wright, Burt, Cairns, & Van Der Heijden, 2005).
In urban and regional planning, predictive scenarios help planners
anticipate future needs. For instance, they can project changes in
population growth rates by 2030 based on existing trends. These forecasts
are essential for estimating future infrastructure, transportation, and
housing requirements, providing planners with concrete data to guide
strategic decisions. By assuming the continuation of current trends,
predictive scenarios play a crucial role in strategic planning, offering
valuable insights for addressing future challenges and opportunities.
32 Mehmet Akif SAĞ

4. Possibilities of Using Scenario Types Together

In planning processes, it is important to not only predict future
developments but also to determine which objectives can be achieved
under specific conditions and to address uncertainties (Börjeson et al.,
2006). In this context, combining predictive, exploratory, and normative
scenarios often leads to more comprehensive and effective results. For
instance, when planning a city’s transport system, predictive scenarios
can use available data to estimate future population and traffic growth.
Simultaneously, exploratory scenarios can assess how unexpected events,
such as climate change or technological advancements, might impact
transport infrastructure (Banister & Hickman, 2013). Finally, normative
scenarios can outline a vision for a desired future state, such as sustainable
and green transportation, and develop strategies to achieve that goal
(Hickman & Banister, 2007).
Combining just two types of scenarios can also be effective. For
example, a predictive scenario can identify future trends based on existing
data, while a normative scenario can outline strategies to achieve specific
goals (Alcamo, 2008). Consider a city’s energy infrastructure: a predictive
scenario might forecast energy consumption trends, while a normative
scenario could demonstrate how the city might achieve carbon neutrality
by 2050 by increasing renewable energy usage (Grübler, Nakićenović, &
Victor, 1999).
In some cases, using a single scenario method may be appropriate. A
normative scenario, for instance, is ideal when the focus is on achieving a
specific target (Robinson, 1990). Similarly, a predictive scenario is sufficient
when assuming current trends will continue (Makridakis, Hogarth, &
Gaba, 2009). However, when uncertainty is high, an exploratory scenario is
best suited for identifying possible paths and providing strategic flexibility
(Van Notten et al., 2003).
The integration of multiple scenario methods offers significant
advantages in urban and regional planning. It enables planners to
understand current trends while preparing for uncertainties (Börjeson et
al., 2006). By combining approaches, planners can create flexible strategies
to achieve desired outcomes, resulting in a more comprehensive planning
process (Van Notten et al., 2003).
Although the future is inherently uncertain, combining scenario
methods creates more resilient and adaptive plans to address different
possibilities (Rotmans et al., 2000). This flexibility strengthens strategic
decision-making and equips planners to respond effectively to complex
challenges. In urban planning, relying on a single method is often
insufficient (Godet & Roubelat, 1996). By integrating predictive,
 International Studies and Evaluations in the Field of City, Region and Planning 33

exploratory, and normative scenarios, planners can develop more realistic,

actionable, and future-ready solutions (Alcamo, 2008). This multifaceted
approach enhances the effectiveness and outcome orientation of planning
processes (Robinson, 1990).
5. Benefits of Writing the Same Topic According to Different
Scenario Types (Creating Alternative Scenarios)
Analyzing the same topic through different scenario types offers
several significant benefits. Firstly, this approach provides a multifaceted
perspective, allowing each scenario type to explore a unique aspect of
the issue and offer distinct strategic approaches (Van Notten et al., 2003).
This enables planners and decision-makers to develop deeper and more
comprehensive insights.
Secondly, the planning process often involves high levels of uncertainty
and numerous variables. Addressing the same issue with different scenario
types helps manage these uncertainties and prepare for the future
(Börjeson et al., 2006). By addressing risks and unknowns from multiple
perspectives, these scenario types enhance flexibility and contribute to
more resilient planning processes (Rotmans et al., 2000).
Finally, integrating different scenario types fosters a more inclusive
planning approach. For example, a normative scenario may focus on
achieving specific goals, while exploratory and predictive scenarios
account for future risks and possibilities, increasing the likelihood of
success (Alcamo, 2008). This combined method effectively balances goal-
oriented planning with the need to navigate uncertainties, resulting in
flexible and adaptive strategies for the future (Robinson, 1990).
In summary, analyzing the same planning issue through multiple
scenario types offers a more nuanced and comprehensive approach. Each
scenario type addresses the issue from a unique perspective, creating a
robust framework for managing uncertainties and developing resilient,
adaptive plans for the future (Godet & Roubelat, 1996).
 HAPTER TWO:
C METHODOLOGIES FOR FUTURE
FORECASTING
1. Delphi Method: A Systematic Approach to Predicting the
Future
The Delphi method is a systematic, interactive approach to forecasting
that addresses uncertainties and complex problems (Linstone & Turoff,
2002). Its primary purpose is to gather expert opinions from diverse fields
and build consensus to make informed predictions about the future. This
method is particularly valuable in long-term strategic planning and policy
34 Mehmet Akif SAĞ

development processes (Okoli & Pawlowski, 2004).

A key feature of the Delphi method is that experts express their views
independently and anonymously, minimizing biases and pressures caused
by group dynamics (Rowe & Wright, 1999). Through this process, experts
share their insights, receive feedback, and refine their perspectives based
on the collective input. This iterative cycle enhances the accuracy and
effectiveness of the final results (Hsu & Sandford, 2007).
1.1. Expert Selection and Criteria
The success of the Delphi method depends on selecting the right
experts. Experts’ knowledge and experience directly influence the
accuracy of predictions and the efficiency of the process (Adler & Ziglio,
1996). Selection should be guided by specific criteria.
Firstly, relevant expertise is essential. Most studies require input
from multiple disciplines. For instance, an urban planning project
benefits from the perspectives of urban planners, sociologists, economists,
environmental scientists, and engineers (Day & Bobeva, 2005). This
interdisciplinary diversity ensures the issue is examined from multiple
angles, leading to more comprehensive results. Experts should possess
deep knowledge and substantial experience in their field, demonstrated
through academic research, industry experience, or contributions to past
projects (Skulmoski, Hartman, & Krahn, 2007). Combining theoretical
expertise from academics with practical insights from professionals
enhances the validity of results by balancing diverse perspectives.
Independence and impartiality are also critical. Experts must be free
to express their opinions without influence from personal or institutional
interests. The anonymity of the Delphi process minimizes group pressure
and promotes honest, unbiased judgments (Rowe & Wright, 2011).
Furthermore, diversity in perspectives enriches the process. Including
experts from various sectors such as academics, local government, public
institutions, and private organizations can provide a well-rounded
understanding of different needs and expectations (Powell, 2003). For
example, in urban planning, incorporating voices from both policymakers
and community representatives ensures the process addresses a broad
spectrum of concerns.
Finally, experts must be willing and motivated to actively participate
throughout the Delphi process, which typically involves multiple rounds.
They should be able to commit sufficient time to provide thoughtful
feedback, as the quality of input directly affects the accuracy of the results
(Okoli & Pawlowski, 2004).
 International Studies and Evaluations in the Field of City, Region and Planning 35

The number of participants depends on the scope and complexity of

the topic, with 10–50 experts considered ideal (Okoli & Pawlowski, 2004).
For interdisciplinary projects, ensuring balanced representation from
each field is crucial. Inviting additional experts at the start can account for
potential dropouts, maintaining the robustness of the process (Skulmoski
et al., 2007).
1.2. Delphi Method Stages
The Delphi method is a systematic, anonymous tool for collecting and
analysing expert opinions in strategic future planning (Linstone & Turoff,
2002). The method involves several critical stages to ensure the accuracy
and validity of the results.
In the first stage, experts from various disciplines are selected to
analyse the city’s future development. For example, predicting how a city’s
transport system will evolve by 2050 requires the input of urban planners,
transport engineers, environmental scientists, economists, and sociologists
(Rowe & Wright, 1999). This interdisciplinary approach ensures the issue
is examined from multiple perspectives, leading to more comprehensive
results (Okoli & Pawlowski, 2004).
The first round involves experts answering open-ended questions,
such as: “What will the structure of the city’s transport system look like in
2050?” or “How might current transport challenges evolve in the future?”
Participants independently share their forecasts and detailed insights
(Hsu & Sandford, 2007). The data collected in this round are analysed
to identify common themes, key trends, and possible contradictions. For
instance, while some experts predict the widespread adoption of electric
and autonomous vehicles, others may highlight the growing importance of
public transport systems. Environmental scientists might draw attention
to the sustainability challenges of future transport options (Skulmoski et
al., 2007).
In the second round, results from the first round are compiled
anonymously and shared as feedback. Experts then re-evaluate their views
based on the insights of other participants. For example, an expert initially
skeptical about the adoption of electric vehicles due to infrastructure
challenges might revise their opinion after reviewing optimistic forecasts
from other participants. At this stage, experts are asked to provide more
specific answers to focused questions, such as: “What percentage of urban
transport will consist of electric and autonomous vehicles by 2050?” (Adler
& Ziglio, 1996).
The second round data are analysed again to identify areas of emerging
consensus and remaining disagreements. For instance, while most experts
36 Mehmet Akif SAĞ

may agree on increased demand for public transport by 2050, opinions

might still differ on the future of private car use (Rowe & Wright, 2011).
In the third round, experts respond to questions based on the
finalised results and provide their final judgements on unresolved issues.
For example: “Will public transport or individual transport dominate urban
mobility in 2050? Please justify.” This stage aims to build consensus on as
many issues as possible while understanding the reasons for any remaining
differences.
At the end of all rounds, the collected data are compiled into a
comprehensive report. The report highlights consensus points, such as
the expected increase in public transport demand and the importance of
sustainable transport systems. It also details ongoing disagreements, such
as the prevalence of autonomous vehicles or the future of private car use.
Recommendations from experts such as strengthening public transport
infrastructure, expanding bike and pedestrian routes, and increasing
charging stations for electric vehicles are included. This report provides
decision-makers with a clear guide for future planning and supports the
development of proactive strategies for the city’s transport system.
1. Trend Analysis
Trend analysis is a key method for forecasting in urban planning
and regional development (Armstrong, 2001). By examining past data,
it identifies future trends and provides valuable insights for strategic
decision-making. Factors such as urban growth dynamics, population
changes, economic developments, and environmental conditions form the
foundation of this approach, enabling planners to create more informed
and effective plans.
This method aids in forecasting housing, infrastructure, and service
needs by analysing demographic trends, population growth, and migration
patterns. For instance, in a city where the population grows by 2% annually,
future needs can be anticipated if this growth continues. Economic trend
analysis helps guide planning for economic growth and infrastructure
investments by examining sectoral developments and employment rates.
For example, in a city with rising industrial production, planners can
anticipate future demands for industrial areas and workforce.
Tracking environmental trends helps cities identify potential
challenges, such as climate change and water resource depletion. For
example, in a city with increasing water consumption, trend analysis can
guide water resource management to mitigate risks of scarcity. Similarly,
analysing transport and infrastructure trends allows planners to address
future needs by evaluating traffic density and infrastructure capacity.
 International Studies and Evaluations in the Field of City, Region and Planning 37

For instance, traffic data can inform improvements in public transport

systems or direct investments in infrastructure upgrades.
1.1. Trend Analysis Process
Trend analysis begins by collecting historical data to project future
developments. Accurate and timely data are critical for ensuring reliable
analysis results (Makridakis et al., 2009).
The collected data are analysed using graphs and visualization
techniques. For instance, population growth trends can be visualized
with line graphs to create future projections. Similarly, economic and
environmental changes can be represented graphically, offering decision-
makers a clear framework for analysis (Batty, 2008).
These analyses enable future projections based on past trends. For
example, a city’s population in 2030 can be estimated by examining
current population growth rates. Similarly, economic trend analysis can
highlight sectors with growth potential, allowing planners to anticipate
future industrial or employment needs (Armstrong, 2001).
Traditional trend analysis is an effective method for forecasting future
developments using existing data. However, when integrated into a vision-
driven framework, it becomes a more targeted and meaningful tool for
planning. The next section will explore the added value of vision-driven
trend analysis.
1.2. Importance of Vision-Driven Trend Analysis
Strategic spatial planning is based not only on analyzing the current
situation but also on interpreting existing data in line with the vision set
for the area being planned. This approach goes beyond predicting the
future based on past trends by linking data to the planned vision. In this
context, vision-oriented trend analysis involves not just analyzing data but
also reinterpreting it to align with the planned future (Healey, 1997).
While all available data are valuable, certain data become more
strategic based on their relevance to the vision. For example, if a city’s
vision includes sustainability, population growth data should be analyzed
in relation to environmental indicators such as water consumption,
energy use, and green space requirements, rather than being considered
in isolation. This process can be seen as a form of data processing, where
relevant data are selected strategically, and indirect links are made between
them and other data that may seem unrelated to the vision. For instance,
analyzing the relationship between energy consumption and public
transport use can help understand the impact of sustainable transport
policies on urban energy (Batty, 2008).
38 Mehmet Akif SAĞ

Vision-driven trend analysis is not just a passive assessment of

existing data. It generates new, vision-oriented insights and makes
strategic inferences for the future based on this information. For example,
although there may be no direct relationship between economic growth
and environmental sustainability, establishing connections between
these dynamics within the vision framework enables the development of
economic strategies that support sustainable development (Armstrong,
2001).
As a result, vision-driven trend analysis strengthens strategic
planning by uncovering relationships between data and aligning them
with planned goals. This approach allows planners to generate meaningful,
vision-oriented information from large data sets. Thus, analyzing past and
current data becomes a powerful tool for making informed and sustainable
decisions for the future (Wegener, 2004).
2. STEEP/PESTLE Analysis and its Importance in Strategic
Spatial Planning
STEEP and PESTLE analyses are effective tools for developing future
foresight and understanding complex systems in strategic planning
(Yüksel, 2012). These methods identify opportunities and threats that
cities and regions may face by systematically assessing social, political,
economic, environmental, technological, and legal factors. In projects with
high levels of uncertainty, STEEP/PESTLE analyses enable the creation of
sustainable, long-term strategies.
STEEP analysis focuses on social (S), technological (T), economic (E),
environmental (E), and political (P) factors, while PESTLE expands this
framework by including legal (L) dimensions. Both approaches are powerful
for assessing future trends and potential impacts. For example, they allow
planners to analyze how social changes influence economic growth and
environmental sustainability in a more holistic way (Christodoulou &
Cullinane, 2019).
These methods play a critical role in shaping long-term visions
and preparing city administrations for the future. By systematically
analyzing the economic effects of political decisions, the social impacts of
technological advancements, and the legal implications of environmental
changes, planners can make well-informed decisions. This creates a robust
framework for managing risks and capitalizing on opportunities in
strategic planning.
The first step in the process is identifying key social, technological,
economic, environmental, political, and legal variables that may
influence future urban development. Social factors include demographic
 International Studies and Evaluations in the Field of City, Region and Planning 39

changes, population growth, migration dynamics, and the state of

social services. Technological factors involve innovations in transport
systems or energy efficiency. Economic indicators such as sectoral
development and unemployment rates reveal the city’s economic
structure, while environmental factors address climate change, resource
use, and sustainability. Political and legal factors examine the influence of
government policies, local decisions, and regulations on urban planning.
The second step involves assessing how these variables directly and
indirectly affect the planning process. For example, analyzing demographic
trends helps planners address pressures from rapid population growth,
such as housing demand and the need for education and healthcare
infrastructure. Technological analysis identifies steps for integrating smart
city technologies or automated transport systems. Environmental analysis
supports the development of sustainable solutions to climate change and
resource protection, while political and legal analysis evaluates the impact
of public policies and regulations on project feasibility (Gupta, 2013).
Finally, the third step develops scenarios based on these analyses to
prepare for future uncertainties. For instance, technological scenarios may
predict widespread adoption of electric vehicles, requiring adjustments in
energy infrastructure and urban design. Environmental scenarios might
address water scarcity caused by climate change, prioritizing policies for
resource management and green space preservation. Social scenarios could
involve planning for increased housing, transport, and social services
in response to migration. These scenarios enhance flexibility and enable
cities to proactively achieve their long-term goals (Perera, 2017).
In conclusion, STEEP/PESTLE analyses strengthen planning
processes by building scenario writing on a scientific foundation. In
uncertain environments, these methods enable the development of flexible,
sustainable strategies while enriching strategic planning and offering
insights into future dynamics.
CHAPTER THREE: SCENARIO TEMPLATES
1. Four Square Scenario Template
The Four Square Scenario method is a key tool for addressing
uncertainties in scenario writing. This approach identifies major
uncertainties in planning processes and analyzes their interactions. The
method’s core concept is to create four distinct future scenarios by placing
two independent key variables on a coordinate system (Schwartz, 1997).
The Four Square Scenario process typically involves the following
steps:
40 Mehmet Akif SAĞ

i. Selection of Key Variables:

The first step is identifying the two most critical and uncertain
factors relevant to the issue being studied. These factors should have a high
degree of uncertainty and a significant impact on planning outcomes. For
example, in urban planning, key variables might include economic growth
and the pace of technological progress.
ii. Polarity of Variables:
Each selected variable is assigned two opposing extremes. For
instance, economic growth could be categorized as “high growth” and
“low growth,” while technological progress might be defined as “rapid
progress” and “slow progress.”
iii. Constructing the Four Square Matrix:
The two variables are plotted on a matrix, forming four intersection
points. Each quadrant represents a unique future scenario. For example:
• High Economic Growth / Rapid Technological Progress: An
innovative, fast-growing city.
• High Economic Growth / Slow Technological Progress: A city
where economic growth outpaces technological development.
• Low Economic Growth / Rapid Technological Progress: A scenario
where technological innovation is insufficient to support economic growth.
• Low Economic Growth / Slow Technological Progress: A static
city structure with limited development.
iv. Development of Scenarios:
Each quadrant is developed into a detailed story or scenario. These
scenarios provide insights into future opportunities, risks, and strategic
requirements.
The Four Square Scenario approach helps planners and decision-
makers prepare for multiple potential futures. By clarifying the impact
of critical variables, it supports more informed and strategic decision-
making. Additionally, understanding the probabilities and potential
effects of different scenarios enables planners to craft flexible, adaptive
plans to navigate uncertainty (Bradfield et al., 2005).
2. Cross-Effect Matrix Scenario Template
The Cross-Effect Matrix is a method used to analyze the relationships
between future events and the likelihood of these relationships occurring.
Rather than evaluating events in isolation, this approach examines how
they influence one another. It is particularly effective for understanding
 International Studies and Evaluations in the Field of City, Region and Planning 41

the dynamics of complex systems and creating comprehensive future

scenarios (Godet, 2000).
This method is typically applied in the following steps:
i. Identification of Critical Events:
The first step is to identify key events or variables to be analyzed.
These may include critical uncertainties or trends that significantly impact
planning. For example, in urban planning, events such as an economic
crisis, population growth, energy demand, or technological innovations
can be selected.
ii. Creating the Cross-Impact Matrix:
The selected events are arranged in rows and columns within a matrix.
Each cell represents the potential impact of one event on another. These
impacts are measured using either numerical values (e.g., a scale from 0 to
5) or qualitative ratings (e.g., ‘high,’ ‘medium,’ ‘low’).
iii. Assessment of Impacts:
The relationships between events are analyzed to identify their
interactions. For instance:
• Could an economic crisis reduce energy demand?
• Does population growth increase pressure on infrastructure?
• Do technological innovations enhance environmental
sustainability by improving energy efficiency?
iv. Determination of Probabilities:
Each relationship in the matrix is evaluated based on probabilities,
helping to estimate the likelihood of events and their cascading effects.
For example, if an economic crisis occurs, there is a high probability that
infrastructure investment will slow down.
v. Interpretation of Results:
The analyzed relationships generate data for scenario development,
enabling planners to understand how future events may unfold and
interact.
By clarifying the complex interactions between future events, the
Cross-Effect Matrix supports the creation of realistic and actionable
scenarios. This method systematically assesses the links between events,
helping planners not only anticipate individual outcomes but also
understand their chain effects.
42 Mehmet Akif SAĞ

The Cross-Effect Matrix’s greatest strength is its ability to

systematically analyze interdependencies between events. By offering a
comprehensive perspective, it enhances planning flexibility and helps
anticipate unexpected consequences, making it a valuable tool in managing
uncertainties and risks.
3. Timeline Scenario Template
Timeline Scenarios analyze the temporal progression of events and
their impact on planning processes. This method enhances planning by
examining how future changes unfold within specific timeframes, making
planning processes more realistic and strategic. By predicting the sequence
and timing of future events, it helps planners create flexible, effective
strategies to achieve long-term goals (Bradfield et al., 2005).
Timeline Scenarios typically involve the following steps:
i. Determining the Temporal Framework:
The first step is to define the timeline the scenario will address,
typically divided into short, medium, and long term periods. For example,
in an urban planning project, timelines might span 5, 15, and 30 years.
ii. Identification of Critical Events and Trends:
Key events, trends, and changes expected within the defined time
periods are identified. These may include population growth, economic
development, technological advancements, or environmental changes.
iii. Creation of Temporal Development Scenarios:
The identified events and trends are placed on a timeline. For instance:
• Short term: Completion of a new transport project.
• Medium term: Increased pressure on infrastructure due to
population growth.
• Long term: Decline in water resources due to climate change
impacts.
iv. Identification of Temporal Priorities and Strategies:
These scenarios help establish priorities and strategies for achieving
long-term goals. For example, infrastructure projects might be prioritized
in the short term, while long-term investments focus on innovative
technologies to ensure environmental sustainability.
Timeline Scenarios provide a roadmap for how events and changes
may unfold over time. By integrating the time dimension, this method
helps planners prioritize and allocate resources more effectively. It also
 International Studies and Evaluations in the Field of City, Region and Planning 43

clarifies not only the impacts of events but the timeframes over which
these impacts occur, enabling decision-makers to balance short, medium
and long-term objectives.
By incorporating the temporal evolution of events, Timeline Scenarios
add a realistic perspective to long-term planning processes. This dynamic
framework allows planners to address uncertainties more effectively,
ensuring that strategies remain adaptive and forward looking.
4. Stakeholder Scenarios Template
The Stakeholder Scenarios method develops inclusive and flexible
scenarios by incorporating the perspectives of various stakeholder groups
into the planning process. This approach integrates the perceptions,
expectations, and priorities of diverse interest groups, enriching the
scenario-writing process and increasing the applicability and acceptance
of resulting strategies (Wright, Cairns, & Goodwin, 2009).
The Stakeholder Scenarios method typically includes the following
steps:
i. Identification of Stakeholders:
The first step is to identify the stakeholder groups that will participate in
the scenario writing process. These groups may include public institutions,
private sector representatives, non-governmental organizations, local
communities, and academia. Each stakeholder contributes unique
expertise, interests, or experiences to the process.
ii. Ensuring Stakeholder Participation:
Once stakeholders are identified, they are engaged through workshops,
surveys, or focus group meetings. These activities allow participants to
share their needs, expectations, and perspectives on potential challenges.
iii. Integration of Stakeholder Perspectives:
After collecting stakeholder input, these perspectives are synthesized
into the development of scenarios. For example, in a transport plan,
public institutions may prioritize infrastructure requirements, while local
communities focus on accessibility. Combining these views leads to more
comprehensive and realistic scenarios.
iv. Creating Different Scenarios:
Multiple scenarios are developed to reflect the priorities and
perspectives of various stakeholder groups. This approach ensures that
strategic options address the needs of different groups, enhancing the
inclusivity and adaptability of the scenarios.
44 Mehmet Akif SAĞ

Stakeholder Scenarios are particularly valuable in addressing complex,

multidimensional planning issues. Engaging diverse stakeholders not only
broadens the knowledge base but also makes decision-making processes
more democratic. Moreover, incorporating stakeholder perspectives
enhances the practicality of plans and ensures they are tailored to the local
context.
The key advantage of this method is its promotion of plurivocality,
ensuring diverse perspectives are included in the planning process. Active
stakeholder involvement makes scenarios more comprehensive, widely
accepted, and helps identify and resolve potential conflicts in advance.
5. Distinguishing Scenario Templates from Scenario Techniques
and Their Role in the Planning Process
Distinguishing between scenario templates and techniques is crucial
for effective scenario writing and analysis. While these concepts are closely
related, clarifying their roles ensures a more structured and efficient
approach to planning. Scenario templates provide structured frameworks
for analyzing data and applying it to scenarios, whereas scenario techniques
define how the scenarios are designed for specific purposes.
Scenario templates systematically structure data analysis, facilitating
the integration of diverse perspectives into the planning process. Templates
like Four-Quadrant Scenarios, Cross-Impact Matrix, Timeline Scenarios,
and Stakeholder Scenarios help planners identify critical uncertainties,
examine event relationships, and assess temporal changes. Importantly,
these templates are versatile and can be applied within any scenario
technique. For example, in exploratory scenario writing, the Four-
Quadrant Scenarios template can explore possible future situations, while
in normative scenario writing, the Cross-Impact Matrix helps evaluate
critical interactions needed to achieve a specific goal.
Scenario techniques, on the other hand, establish the philosophy and
methodology of the writing process. Normative scenarios outline steps
to achieve a specific objective, exploratory scenarios investigate potential
future possibilities, and predictive scenarios focus on forecasting future
outcomes based on current trends and data. Templates complement these
techniques by enabling data to be used more meaningfully. For instance,
in predictive scenario writing, the Timeline Scenarios template models
temporal changes more clearly.
In conclusion, templates and techniques are distinct yet
complementary elements of scenario writing. Templates provide the tools
for applying techniques, enabling multidimensional analysis and scenario
development. By distinguishing the analytical role of templates from the
 International Studies and Evaluations in the Field of City, Region and Planning 45

methodological role of techniques, planning processes can become more

flexible, strategic, and effective.
CONCLUSION
Scenario writing in urban and regional planning is a vital tool for
strengthening strategic decision making through a systematic approach
to future uncertainties. This method helps planners understand future
possibilities, manage risks, and seize opportunities. However, creating
effective scenarios requires not only technical expertise but also rigorous
analysis and the integration of diverse stakeholder contributions.
The flexibility and foresight of scenario writing provide an essential
framework for cities to develop sustainable, adaptable strategies. This
meticulous approach lays the foundation for resilient, future-ready
planning.
46 Mehmet Akif SAĞ

REFERENCES

Adler, M., & Ziglio, E. (1996). Gazing into the oracle: The Delphi method and its
application to social policy and public health: Jessica Kingsley Publishers.

Albrechts, L. (2004). Strategic (spatial) planning reexamined. Environment and

Planning B: Planning and design, 31(5), 743-758.

Alcamo, J. (2008). Environmental futures: the practice of environmental scenario

analysis (Vol. 2): Elsevier.

Armstrong, J. S. (2001). Principles of forecasting: a handbook for researchers and

practitioners (Vol. 30): Springer.

Banister, D., & Hickman, R. (2013). Transport futures: Thinking the unthinkable.
Transport Policy, 29, 283-293.

Batty, M. (2008). The size, scale, and shape of cities. Science, 319(5864), 769-771.

Börjeson, L., Höjer, M., Dreborg, K.-H., Ekvall, T., & Finnveden, G. (2006).
Scenario types and techniques: towards a user’s guide. Futures, 38(7),
723-739.

Bradfield, R., Wright, G., Burt, G., Cairns, G., & Van Der Heijden, K. (2005).
The origins and evolution of scenario techniques in long range business
planning. Futures, 37(8), 795-812.

Bulkeley, H., & Betsill, M. (2005). Rethinking sustainable cities: Multilevel

governance and the’urban’politics of climate change. Environmental
politics, 14(1), 42-63.

Christodoulou, A., & Cullinane, K. (2019). Identifying the main opportunities

and challenges from the implementation of a port energy management
system: A SWOT/PESTLE analysis. Sustainability, 11(21), 6046.

Day, J., & Bobeva, M. (2005). A generic toolkit for the successful management
of Delphi studies. Electronic Journal of Business Research Methods, 3(2),
pp103‑116-pp103‑116.

Glenn, J. C., & Gordon, T. J. (2009). Futures research methodology: version 3.0:
Millennium Project Washington, DC.

Godet, M. (2000). The art of scenarios and strategic planning: tools and pitfalls.
Technological Forecasting and Social Change, 65(1), 3-22.

Godet, M., & Roubelat, F. (1996). Creating the future: the use and misuse of
scenarios. Long range planning, 29(2), 164-171.

Grübler, A., Nakićenović, N., & Victor, D. G. (1999). Dynamics of energy

technologies and global change. Energy policy, 27(5), 247-280.
 International Studies and Evaluations in the Field of City, Region and Planning 47

Gupta, A. (2013). Environment & PEST analysis: an approach to external business

environment. International Journal of Modern Social Sciences, 2(1), 34-43.

Healey, P. (1997). Traditions of planning thought. Collaborative planning:

Shaping places in fragmented societies, 7-30.

Hickman, R., & Banister, D. (2007). Looking over the horizon: Transport and
reduced CO2 emissions in the UK by 2030. Transport Policy, 14(5), 377-
387.

Hopkins, L. D., & Zapata, M. (2007). Engaging the future: Forecasts, scenarios,
plans, and projects. (No Title).

Hsu, C.-C., & Sandford, B. A. (2007). The Delphi technique: making sense of
consensus. Practical assessment, research, and evaluation, 12(1).

Linstone, H. A., & Turoff, M. (2002). Delphi bibliography. Journal Articles,

14(54), 68.

Makridakis, S., Hogarth, R. M., & Gaba, A. (2009). Forecasting and uncertainty
in the economic and business world. International Journal of Forecasting,
25(4), 794-812.

Okoli, C., & Pawlowski, S. D. (2004). The Delphi method as a research tool:
an example, design considerations and applications. Information &
management, 42(1), 15-29.

Perera, R. (2017). The PESTLE analysis. Nerdynaut: Avissawella, Sri Lanka.

Powell, C. (2003). The Delphi technique: myths and realities. Journal of Advanced
Nursing, 41(4), 376-382.

Robinson, J. B. (1990). Futures under glass: a recipe for people who hate to
predict. Futures, 22(8), 820-842.

Rotmans, J., van Asselt, M., Anastasi, C., Greeuw, S., Mellors, J., Peters, S., . . .
Rijkens, N. (2000). Visions for a sustainable Europe. Futures, 32(9-10),
809-831.

Rowe, G., & Wright, G. (1999). The Delphi technique as a forecasting tool: issues
and analysis. International Journal of Forecasting, 15(4), 353-375.

Rowe, G., & Wright, G. (2011). The Delphi technique: Past, present, and future
prospects—Introduction to the special issue. Technological Forecasting
and Social Change, 78(9), 1487-1490.

Schwartz, P. (1997). Art of the long view: planning for the future in an uncertain
world: John Wiley & Sons.

Skulmoski, G. J., Hartman, F. T., & Krahn, J. (2007). The Delphi method for
graduate research. Journal of Information Technology Education: Research,
6(1), 1-21.
48 Mehmet Akif SAĞ

Van Notten, P. W., Rotmans, J., Van Asselt, M. B., & Rothman, D. S. (2003). An
updated scenario typology. Futures, 35(5), 423-443.

Wack, P. (1985). Scenarios: uncharted waters ahead. Harvard business review,

63(5), 72-89.

Wegener, M. (2004). Overview of land use transport models. In Handbook of

transport geography and spatial systems (pp. 127-146): Emerald Group
Publishing Limited.

Wright, G., Cairns, G., & Goodwin, P. (2009). Teaching scenario planning:
lessons from practice in academe and business. European Journal of
Operational Research, 194(1), 323-335.

Yüksel, I. (2012). Developing a multi-criteria decision making model for PESTEL

analysis. International Journal of Business and Management, 7(24), 52.
 Chapter 3
COMPARISON OF TEXT-TO-IMAGE GENERATIVE
AI TOOLS FOR URBAN PORTRAITS: A CASE
STUDY OF INVISIBLE CITIES BY USING STABLE
DIFFUSION, DALL-E AND MIDJOURNEY

Özlem Kevseroğlu1
Rifat Kurban2

1 Assist. Prof. Dr. Özlem Kevseroğlu, Abdullah Gül University, Department of Architecture,
Kayseri, Turkey, [email protected], 0000-0003-1828-225
2 Assoc. Prof. Dr. Rifat Kurban, Abdullah Gül University, Department of Computer Engi-
neering, Kayseri, Turkey, [email protected], 0000-0002-0277-2210
50 Özlem Kevseroğlu, Rifat Kurban

1. Introduction
Generative artificial intelligence (AI) including text-to-image tools, may
become one of the core technologies for architects and urban planners. In
this manner, they are foreseen to deal with how urban portraits are navigated
and visualized. Practicing planners will benefit from visualizations as these
are valuable tools for communication, comprehension, and decision making
by eliminating gaps created by language over images. Such technologies
allow planners and designers through the use of artificial intelligence
algorithms to quickly create multiple artistic representations of any given
text input, which assists the design stage. They also enhance the inclusion of
users’ needs and other disciplines’ views in the design process which boosts
the and inclusiveness of the cities’ built environment. Hence the application
of text-to-image tools and generative artificial intelligence has the potential
to change the way experts carryout urban design by providing abundant
resources which are limited by time and works themselves.
The field of architecture and urban planning is currently witnessing
a growing integration of generative artificial intelligence (AI), which is
fundamentally transforming conventional design methods and expanding
the realm of possibilities. Generative AI in architecture refers to the
application of AI techniques, such as generative adversarial networks
(GANs) and deep learning models, to autonomously create designs, explore
complex architectural forms, and optimize building functioning (As et
al., 2018). Architects can efficiently explore numerous design alternatives
using AI’s ability to generate novel design concepts through the application
of predetermined rules or datasets (Li et al., 2023). In addition to basic
automation, generative AI is being employed in architecture and urban
planning research to facilitate a significant expansion of architectural
practice by enhancing both creativity and efficiency in the design process
(Hegazy & Saleh, 2023). Architects may leverage the capabilities of AI to
create intricate and innovative architectural solutions through the use of
generative grammar, exploration of parametric design possibilities, and
investigation of deep learning methodologies (Sadek, 2023). AI-powered
technologies streamline the integration of intelligent design aspects,
facilitate the creation of architectural ideas and shapes, and promote
technical innovation in the sector.
No matter how promising generative AI in architecture may be, ethical
considerations must always take precedence. It is critical to ethically employ
AI in order to avoid negative consequences for human creativity and design
ethics. The integration of generative AI into these domains signifies a
significant transformation in the design process, enabling architects and
planners to explore innovative methods, boost their creative output, and
optimize their designs for maximum efficiency. With the help of AI, experts
 International Studies and Evaluations in the Field of City, Region and Planning 51

can discover and test the boundaries of architectural innovation.

Generative AI models like DALL-E, Stable Diffusion, and Midjourney,
which form visuals from text prompts, have greatly influenced architectural
visual culture. DALL-E, Stable Diffusion, and Midjourney are the leaders in
text-to-image production. They utilize state-of-the-art artificial intelligence
algorithms to generate visually realistic graphics of superior quality
based on textual inputs (Buehler, 2022). The models have demonstrated
the potential to promote architectural creativity by assisting architects in
rapidly and efficiently visualizing their concepts (Jeong et al., 2023). These
models are transforming the architectural design process by allowing
architects to translate verbal explanations of architectural concepts into
visual representations (Steinfeld, 2023). Architects and planners now have
a new tool at their disposal for effectively exploring design concepts and
communicating architectural ideals thanks to the development of these
text-to-image generative models (Adetayo, 2024). Architects can employ
transformer models such as DALL-E and diffusion-based buildings to
generate distinctive and realistic architectural renderings based on textual
instructions (Ma et al., 2023). Architects can utilize generative diffusion
models like DALL-E not only for image generation but also for investigating
various design possibilities, providing visually coherent narratives, and
expediting the prototyping of architectural ideas (Singh et al., 2021). The
field of architectural visualization has made significant progress due to these
models’ impressive capabilities in creating pictures based on composition
and systematic generalization (Singh et al., 2021). Moreover, the capacity to
scale up and use large-scale text-to-image generation models, like DALL-E
(Ko et al., 2022), has highlighted their importance in various fields such as
fashion, fine art, industrial design, urban planning, and architecture. These
models possess the potential to completely transform the manner in which
architects articulate and conceptualize design ideas, opening up fresh
possibilities for the representation and examination of architecture (Çelik,
2023). The emergence of AI art platforms like DALL-E, Midjourney, and
Stable Diffusion has further propelled this exploration, raising questions
about the role of AI in architectural representation and its potential to
augment human creativity (Ploennigs and Berger, 2023). Research has
investigated the practical applications of these tools, identifying both
benefits, such as accelerated design exploration and enhanced visual
communication, and limitations, including the need for robust prompt
engineering and concerns about bias and job displacement (Albaghajati
et al., 2023; Rane et al., 2023). A quantitative method focused on certain
criteria related to the design, taking inspiration from the Shah Mosque by
using MidJourney to provide the pros and cons is proposed (Maksoud et
al., 2024). Notably, Sukkar et al. (2024) highlighted the challenges of using
52 Özlem Kevseroğlu, Rifat Kurban

AI to represent Islamic architectural heritage, emphasizing the need for

culturally sensitive and contextually aware AI models. Yildirim (2022)
explored the text-to-image generation AI, tracing its history and evaluated
the algorithms that underpin its functionality and, this study analyzes
popular platforms like DALL-E 2, Midjourney, and others, ultimately
showcasing the potential of this technology to revolutionize architectural,
interior, and urban design processes. Zeytin et al. (2024) examined the
impact of Generative Design Assistants (GDAs), especially ArchiGAN,
on architectural design practice, revealing both their potential for
innovation and the need for a balanced approach to avoid overshadowing
human creativity. Overall, DALL-E, Stable Diffusion, and Midjourney are
advanced tools that have revolutionized the interaction between AI and
architecture. These instruments provide architects with innovative methods
to generate visual content based on written descriptions. The potential of
these generative AI models to expand architectural visualization, enhance
architectural innovation, and optimize the design process is immense.
Released in 1972, Italo Calvino’s postmodern novel Invisible Cities
presents a series of strange depictions of fictional places, deviating from
traditional narrative standards (Calvino, 2010). The book, including a
conversation between the adventurer Marco Polo and the emperor Kublai
Khan, presents fifty-five distinct urban locations, each possessing unique
qualities, atmospheres, and social structures. The fictional cities shown in
this work sometimes surpass physical limitations and delve into universal
concepts such as language, memory, desire, mortality, and the urban
experience. Invisible Cities is significant for urban studies because of its
poetic and evocative examination of urban imaginaries. The book challenges
the conventional portrayal of cities by examining not only their physical
architecture and demography but also delving into the psychological
and emotional components of urban life. Each city assumes the function
of a metaphor, symbolizing a specific aspect of the human situation or a
philosophical concept. For example, the city of Zaira explores the concept
of language and its influence on our perception of the external world,
while the city of Isaura symbolizes the relationship between memory and
reality. The reason why this book was chosen in this study is that it contains
descriptions that will enable the cities to be written down and imagined in
the human mind, and therefore it is very suitable to be used as a prompt.
The novel has served as a catalyst for architectural contemplation, providing
valuable perspectives on current architectural dilemmas and intricate
societal matters pertaining to urban growth (Bajçinovci et al., 2017).
Furthermore, the artistic partnership resulting from “Invisible Cities” has
given rise to the development of imperceptible auditory elements akin to
the delicacy of the fortepiano, highlighting the multidisciplinary impact
 International Studies and Evaluations in the Field of City, Region and Planning 53

of Calvino’s writings on the fields of architecture and music (Manzolli &

Marinho, 2020). Moreover, scholars have employed the novel’s portrayal of
the interaction and conversation between order and chaos to comprehend
and tackle the intricate organizational and sociological aspects of urban
areas. This showcases the wide-ranging uses of Calvino’s concepts in the
field of urban studies (Mukhija, 2015; Case & Gaggiotti, 2014). Calvino’s
work offers a distinctive perspective for examining and interpreting
urban environments, providing a comprehensive theoretical structure for
comprehending and tackling diverse architectural and urban issues.
Calvino’s writings present a diverse range of urban ideas that can be
transformed into visual representations, making them particularly relevant
for studying AI-generated images of cities. The abundance of symbolic
characters and emotive language in the book provides AI algorithms
with ample information to comprehend and generate original, thought-
provoking pictures. These renderings can initiate discussions on the
capabilities and limitations of artificial intelligence in understanding and
depicting the intricate nature of urban life.
The focus of this study is to explore the possible uses of AI tools in
architecture and urbanism, the new domain of text-to-image generation
using MidJourney V6, Microsoft Designer Image Creator sculpting tool
(DALL-E 3), Stable Diffusion XL Turbo, Open-AI GPT4 (DALL-E 3) as
artificial intelligence (AI) media and their possible usage in architecture
and urbanism by means of visual questionnaire applied to 20 subjects
located in five cities of Calvino’s Invisible Cities (Argia, Diomira, Ersilia,
Isidora, Sophronia) and by comparative assessment of the AI tools in
accordance with what the subjects provided. This paper intends to include
the features and limitations of these tools, and make propositions about the
different future uses of these tools in the field of architectural design and
urban studies.
The rest of this paper is organized as follows: the cities selected from
the “Invisible Cities,” and the AI tools used in the study are given in Section
2. Section 3 outlines extensive experiments with urban representation
prompts of selected cities and the subjective ranking results. Finally, Section
4 concludes the paper.
2. Material & Methods
This section presents a selection of urban images intentionally
highlighting the great cities which are mirrored in some of the pages of
Calvino’s book, Invisible Cities. These passages, rich with imagery, further
consider the process of image creation through sophisticated technology
and its implementation. In the materials part, the selected extracts are
presented more artistically to help the reader become familiar with the
54 Özlem Kevseroğlu, Rifat Kurban

rich description of the cities that Calvino envisioned. At the same time,
in the methods part, we present intelligent techniques that were used for
image making and provide the background context in which the image was
generated.
2.1 Material
Argia, featured in “Cities & Memory 1”; Diomira, also in “Cities &
Memory 1”; Ersilia, highlighted in “Trading Cities 4”; Isidora, showcased
in “Cities & Memory 2”; and Sophronia, celebrated within the pages of
“Thin Cities,” constitute the cities chosen for examination in this study.
The selection criteria were carefully curated to ensure that the number of
prompts remained concise while retaining the capacity to be translated into
tangible expressions. Table 1 outlines the prompts provided to the artificial
intelligence models, derived directly from the original texts and descriptions
encapsulating the essence of each of these distinctive cities. These city
portraits have been converted into a number of prompts for each program.
In the table below you can see both the paragraphs and the prompts for the
cities. Since the programs have a limited input context window (token size),
the city portraits have been shortened. When abbreviating, care was taken
to keep the expressions that would become tangible and to remove the rest
of the intangible expressions. The reduction process aimed to maintain
the essence of each city’s unique character within the constraints of the
program’s input context window.
2.2 Methods
Text-to-image generation has attracted a lot of interest because of its
potential uses in a number of fields, including virtual environments, design
automation, and content creation. In recent years, advances in generative
models have made it easier to create complex systems that can convert verbal
descriptions into visually consistent visuals. DALL-E, Stable Diffusion, and
MidJourney are three of these models that have become front-runners,
each pushing the limits of what is possible in text-to-image synthesis. In
this work, we compare these models and provide insights into their designs.
 International Studies and Evaluations in the Field of City, Region and Planning 55

Table 1. Selected urban portrait texts and prompts for the questionaries.

Text of the City in the Book (Calvino, 2010) Prompt of the City (Short-
ened Version)
“What makes Argia different from other cities is that it “What makes Argia different
has earth instead of air. The streets are completely filled from other cities is that it has
with dirt, clay packs the rooms to the ceiling, on every earth instead of air. The streets
stair another stairway is set in negative, over the roofs are completely filled with dirt,
of the houses hang layers of rocky terrain like skies clay packs the rooms to the
with clouds. We do not know if the inhabitants can ceiling, on every stair another
move about the city, widening the worm tunnels and stairway is set in negative, over
Argia

the crevices where roots twist: the dampness destroys the roofs of the houses hang
people’s bodies and they have scant strength; everyone layers of rocky terrain like skies
is better off remaining still, prone; anyway, it is dark. with clouds. We do not know if
From up here, nothing of Argia can be seen; some say, the inhabitants can move about
“It’s down below there,” and we can only believe them. the city, widening the worm
The place is deserted. At night, putting your ear to the tunnels and the crevices where
ground, you can sometimes hear a door slam.” roots twist.”
“Leaving there and proceeding for three days toward “Leaving there and proceeding
the east, you reach Diomira, a city with sixty silver for three days toward the east,
domes, bronze statues of all the gods, streets paved you reach Diomira, a city with
with lead, a crystal theater, a golden cock that crows sixty silver domes, bronze stat-
each morning on a tower. All these beauties will al- ues of all the gods, streets paved
ready be familiar to the visitor, who has seen them with lead, a crystal theater, a
Diomira

also in other cities. But the special quality of this golden cock that crows each
city for the man who arrives there on a September morning on a tower and the
evening, when the days are growing shorter and the multicolored lamps are lighted
multicolored lamps are lighted all at once at the doors all at once at the doors of the
of the food stalls and from a terrace a woman’s voice food stalls.”
cries ooh!, is that he feels envy toward those who now
believe they have once before lived an evening identi-
cal to this and who think they were happy, that time.”
“In Ersilia, to establish the relationships that sustain the “In Ersilia, the inhabitants
city’s life, the inhabitants stretch strings from the cor- stretch strings from the corners
ners of the houses, white or black or gray or black-and- of the houses, white or black
white according to whether they mark a relationship of or gray or black-and-white ac-
blood, of trade, authority, agency. When the strings be- cording to whether they mark
come so numerous that you can no longer pass among a relationship of blood, of
them, the inhabitants leave: the houses are dismantled; trade, authority, agency. When
only the strings and their supports remain. From a the strings become so numer-
mountainside, camping with their household goods, ous, the inhabitants leave: the
Ersilia’s refugees look at the labyrinth of taut strings houses are dismantled; only
Ersilia

and poles that rise in the plain. That is the city of Ersil- the strings and their supports
ia still, and they are nothing. They rebuild Ersilia else- remain. From a mountainside,
where. They weave a similar pattern of strings which camping with their household
they would like to be more complex and at the same goods, refugees look at the lab-
time more regular than the other. Then they abandon it yrinth of taut strings and poles
and take themselves and their houses still farther away. that rise in the plain.”
Thus, when traveling in the territory of Ersilia, you
come upon the ruins of the abandoned cities, without
the walls which do not last, without the bones of the
dead which the wind rolls away: spiderwebs of intricate
relationships seeking a form.”
 56 Özlem Kevseroğlu, Rifat Kurban

“When a man rides a long time through wild regions “Finally, he comes to Isido-
he feels the desire for a city. Finally, he comes to ra, a city where the buildings
Isidora, a city where the buildings have spiral stair- have spiral staircases encrust-
cases encrusted with spiral seashells, where perfect ed with spiral seashells, where
telescopes and violins are made, where the foreigner perfect telescopes and violins
hesitating between two women always encounters are made, where the foreigner
a third, where cockfights degenerate into bloody hesitating between two wom-
Isidora

brawls among the bettors. He was thinking of all en always encounters a third,
these things when he desired a city. Isidora, there- where cockfights degenerate
fore, is the city of his dreams: with one difference. into bloody brawls among the
The dreamed-of city contained him as a young man; bettors. In the square there is
he arrives at Isidora in his old age. In the square the wall where the old men sit
there is the wall where the old men sit and watch the and watch the young go by; he
young go by; he is seated in a row with them. Desires is seated in a row with them.
are already memories.” Desires are already memories.”
“The city of Sophronia is made up of two half-cities. “The city of Sophronia is made
In one there is the great roller coaster with its steep up of two half-cities. In one
humps, the carousel with its chain spokes, the Ferris there is the great roller coaster
wheel of spinning cages, the death-ride with crouch- with its steep humps, the car-
ing motorcyclists, the big top with the clump of tra- ousel with its chain spokes, the
pezes hanging in the middle. The other half-city is Ferris wheel of spinning cages,
of stone and marble and cement, with the bank, the the death-ride with crouching
factories, the palaces, the slaughterhouse, the school, motorcyclists, the big top with
and all the rest. One of the half-cities is permanent, the clump of trapezes hanging
the other is temporary, and when the period of its in the middle. The other half-
Sophronia

sojourn is over, they uproot it, dismantle it, and take city is of stone and marble and
it off, transplanting it to the vacant lots of another cement, with the bank, the fac-
half-city. And so every year the day comes when tories, the palaces, the slaugh-
the workmen remove the marble pediments, lower terhouse, the school, and all the
the stone walls, the cement pylons, take down the rest.”
Ministry, the monument, the docks, the petroleum
refinery, the hospital, load them on trailers, to fol-
low from stand to stand their annual itinerary. Here
remains the half-Sophronia of the shooting-galler-
ies and the carousels, the shout suspended from the
cart of 63 the headlong roller coaster, and it begins
to count the months, the days it must wait before the
caravan returns and a complete li~ can begin again.”

2.2.1 DALL-E
OpenAI’s DALL-E is a generative model that turns written descriptions
into visuals. It utilizes a transformer-based architecture to encode the text
input into embeddings. These embeddings condition the image generation
process, which is handled by a separate transformer model. The transformer
is adept at modeling long-range dependencies and global relationships
within the data. It uses its self-attention mechanism to attend to different
parts of the text embedding and various stages of the image generation,
guiding the creation of the final image. Through this process, DALL-E
produces images that exhibit impressive levels of detail and coherence
while remaining semantically consistent with the text prompt. DALL-E
 International Studies and Evaluations in the Field of City, Region and Planning 57

promises to be a powerful text-to-image creation tool for various urban

studies applications, including urban planning and exploring potential
design solutions.
The development of generative AI has revolutionized the field of image
generation, with DALL-E at the forefront of this innovation. OpenAI
introduced DALL-E, a state-of-the-art text-to-image model that has garnered
significant attention for its advanced and diverse capacities for combinational
image generation with specific textual prompts (Wang et al., 2023). The
emergence of surreal and grotesque image sets created using DALL-E
Mini has prompted analysis of their aesthetic content and connections to
preexisting media forms and trends in digital culture (O’Meara & Murphy,
2023). Furthermore, DALL-E 2 has been shown to significantly underperform
when tasked with generating specific images, indicating the need for further
research and development in this area (Mbalaka, 2023). Additionally,
DALL-E has demonstrated an impressive ability for composition-based
systematic generalization in image generation, albeit with the requirement
of a dataset of text-image pairs (Singh et al., 2021). Moreover, Ramesh et
al. (2022) have compared DALL-E with other text-to-image systems, such
as GLIDE, showcasing comparable quality to the latter but with greater
diversity in its image generation. Adams et al. (2022) have highlighted the
potential of DALL-E 2 as a tool for image generation, augmentation, and
manipulation in the field of radiology, provided that these models possess
sufficient medical domain knowledge. Additionally, compared to traditional
DALL-E, the quality of generated images has significantly improved, leading
to the introduction of DALL-E 2 with significantly better output generation
(Poredi, 2023). These advancements in generative AI, particularly with
DALL-E, have paved the way for the use of AI tools to generate high-quality
images from text prompts and reference images, indicating its potential for
various applications (Moral-Andrés et al., 2022).
DALL-E’s development and capabilities, as well as its subsequent
iterations, have significantly advanced the field of generative AI,
particularly in text-to-image generation. However, there are still challenges
and limitations that need to be addressed, such as the need for domain-
specific knowledge and the performance of the model when tasked with
generating specific images. Despite these challenges, DALL-E and its
iterations represent a promising future tool for image generation, with
potential applications across various domains, including radiology, design,
and content generation.
2.2.2 Stable Diffusion
Stable Diffusion, a latent text-to-image diffusion model created by
Stability AI, excels at generating high-quality images from natural language
58 Özlem Kevseroğlu, Rifat Kurban

descriptions (Podell, et al., 2023). Unlike pixel-based diffusion models, Stable

Diffusion applies the diffusion process within a compressed latent space,
enabling efficient training and image generation with reduced computational
requirements. Using a U-Net architecture with cross-attention layers, the
model effectively extracts both global and local information from the text
prompt, guiding the generation process. During the diffusion phase, the
model gradually denoises a latent representation based on the text prompt,
ultimately producing a coherent image consistent with the given description.
Stable Diffusion’s flexibility makes it a valuable tool for various applications
within urban studies, including urban design, architectural visualization, and
other creative domains. This approach not only delivers visually impressive
results but also allows for control over the image generation process through
text prompts and multiple sampling methods.
Stable Diffusion is a cutting-edge text-to-image generative AI model
that has gained significant attention in the field of artificial intelligence.
This model has been widely recognized for its ability to generate high-
quality images in a matter of seconds based on textual descriptions (Du et
al., 2023). The Stable Diffusion model has been implemented across various
applications, including text-to-image generation, image-to-video synthesis,
model-agnostic zero-shot classification, and unsupervised semantic
correspondence investigation (Siemens, 2023). This demonstrates the
versatility and broad utility of Stable Diffusion in the realm of generative AI.
Moreover, the text-guided diffusion model GLIDE, which is based on Stable
Diffusion, has been acknowledged as the state of the art in text-to-image
generative AI, further highlighting the significance of Stable Diffusion in
the domain of generative AI (Kather et al., 2022). Additionally, the model
has been leveraged in diverse contexts, such as in the development of
co-creative text adventure games and in enhancing social virtual reality
experiences, showcasing its adaptability and relevance in various domains
(Sun et al., 2022; Numan et al., 2023). Furthermore, the emergence of Stable
Diffusion has been attributed to the advancements in learning techniques,
particularly the utilization of generative adversarial networks (GANs) and
diffusion models using Contrastive Language-Image Pre-training (CLIP)
(Brisco et al., 2023). This underscores the technological underpinnings
and sophisticated methodologies that contribute to the efficacy of Stable
Diffusion in text-to-image generation.
Stable Diffusion has emerged as a pivotal and influential text-to-
image generative AI model, offering rapid image generation capabilities,
versatility across applications, and a foundation built on advanced learning
techniques. Its impact extends beyond traditional text-to-image generation,
encompassing diverse domains and applications, thereby solidifying its
position as a leading model in the field of generative AI.
 International Studies and Evaluations in the Field of City, Region and Planning 59

2.2.3 MidJourney
Users can access the text-to-image AI model known as Midjourney
through its Discord channel, website, and API. While the specific design
of Midjourney remains undisclosed, it is widely believed to be based on
diffusion models, similar to Stable Diffusion. To interact with the model,
users input commands into the chosen platform, prompting the generation
of images that correspond to the provided descriptions. Midjourney
distinguishes itself through its emphasis on artistic expression and the
creation of visually captivating imagery in diverse styles and aesthetics.
This capability likely stems from a combination of its model architecture,
potentially incorporating mechanisms for artistic interpretation and style
transfer, and its training data, which presumably encompasses a wide array
of artistic works. Although the technical details of Midjourney remain
largely unknown, its outputs suggest an approach that transcends basic
text-to-image translation, venturing into the realms of artistic co-creation
and creative exploration. This positions Midjourney as a potentially
valuable tool for urban studies, particularly in disciplines where aesthetic
considerations and visual communication play a crucial role.
Text-to-image generation has gained significant attention in recent
years due to its potential applications in various fields such as content
marketing, medical education, design, and journalism. Midjourney, a
generative AI tool, has emerged as a prominent player in this domain,
alongside other models like DALL-E 2 and Disco Diffusion (Brisco et
al., 2023; Kenig et al., 2023; Lyu et al., 2022). These AI systems have
demonstrated the capability to produce images based on text prompts,
mimicking the cognitive processes of the human mind (Kenig et al., 2023).
However, developers may intentionally limit the use of such AI programs in
certain domains, such as medical material creation (Youssef, 2023). Despite
this, Midjourney and similar AI generators have shown promise in creating
fully rendered images solely based on user prompts, offering intricate
and impressive results (Newton & Dhole, 2023). The potential of text-to-
image generation models, including Midjourney, extends to image editing
capabilities and the development of new tools for creative practitioners
(Saharia et al., 2022).

Table 2. Comparison of state-of-the-art text-to-image generation AI tools.

Feature MidJourney V6 Stable Diffusion XL Turbo DALL-E 3

Release December, 2023 July, 2023 August, 2023
Date
Architec- - Diffusion-based Transform-
ture er-based
60 Özlem Kevseroğlu, Rifat Kurban

Accessi- Subscrip- Open-source, widely accessible Subscrip-

bility tion-based, access tion-based, ac-
through using cess through web
Discord web in- interface, API
terface support
Pricing (as 8USD/mo (Basic 12USD/mo (Clipdrop Pro) 20USD/mo
of May, plan) (OpenAI GPT4)
2024) Free (MS Design-
er Img. Creator)
Web Site https://midjourney. https://clipdrop.co/stable-diffu- https://chatgpt.
com sion-turbo com
https://designer.
microsoft.com/
image-creator
More than that, people have looked into how AI systems like Midjourney
affect the environment. For example, Tomlinson et al. (2023) compared the
carbon emissions that AI systems and humans produce when they do the
same tasks. As such, Midjourney and similar AI image generators have the
potential to revolutionize various industries, offering efficient and creative
solutions for image generation and content creation. Table 2 shows a brief
comparison of DALL-E, Stable Diffusion and Midjourney AI tools.
3. Experimental Results
This section explores the potential of text-to-image artificial intelligence
(AI) tools for creating urban portraits. Three AI tools – DALL-E, Stable
Diffusion, and Midjourney – were used to visualize five city descriptions
selected from Calvino’s seminal work, Invisible Cities. Two methods were
employed for DALL-E 3: the first utilized OpenAI’s ChatGPT4, which
restructures prompts before transmitting them to DALL-E 3, while the
second used Microsoft Designer’s Image Creator tool. Stable Diffusion’s XL
Turbo model was accessed via the ClipDrop website, and Midjourney’s v6
version was accessed through its Discord channel. Each AI tool generated
four 1024×1024 resolution color images per prompt. These images are
presented in the Visual Results section and visually evaluated. To assess
the quality of the generated images, 20 domain experts were tasked with
ranking the outputs of each tool from best to worst based on the provided
prompt (artistic style, adherence to the text and overall impression). The
ranking results are presented in the Subjective Evaluation section, offering
a statistically informed perspective on the comparative performance of
each AI tool.
3.1 Visual Results
In this section, the renderings created with the given prompts are
given. For each prompt, AI tools were asked to generate 4 different images.
 International Studies and Evaluations in the Field of City, Region and Planning 61

Figure 1 shows the generated artificial images for the city of Argia.
In the prompt, important key features can be listed as: earth, dirt, clay,
stairways (inverted), roofs (layered), worm tunnels, crevices, subterranean,
confinement and claustrophobic. Overall, Argia’s visual identity is one
of subterranean confinement, filled with earth and featuring unusual
architectural elements like inverted stairways and layered roofs. This
creates a stark contrast with typical cities, emphasizing its otherworldly
and claustrophobic nature. An analysis of the Midjourney outputs reveals
a progression in the depiction of soil texture, moving from an urban scale
to a finer street-level representation. Additionally, these images generally
exhibit a layered morphological structure, encompassing both underground
and aboveground elements, particularly in the staircase formations.

MidJourney V6 MS Designer (DALL-E 3)

Stable Diffusion XL Turbo GPT4 (DALL-E 3)

Figure 1. Visual representation of Argia’s urban portraits obtained by AI tools.

62 Özlem Kevseroğlu, Rifat Kurban

The DALL-E 3 model accessed through MS Designer expresses these

underground and aboveground layers more prominently, with a notable
emphasis on worm-created tunnels. Conversely, images generated by Stable
Diffusion demonstrate the presence of soil texture and urban stratification
solely at the surface level, lacking detailed subterranean features. The
GPT4-powered DALL-E 3 images, however, successfully convey the
existence of distinct underground and aboveground layers, showcasing an
interconnected staircase system. Notably, these images depict signs of life
in both subterranean and surface environments.
MidJourney V6 MS Designer (DALL-E 3)

Stable Diffusion XL Turbo GPT4 (DALL-E 3)

Figure 2. Visual representation of Diomira’s urban portraits obtained by AI

tools.

Figure 2 presents a comparative analysis of artificially generated

images of the fictional city, Diomira, produced by various AI models. The
prompt provided to the models incorporated key descriptive elements,
 International Studies and Evaluations in the Field of City, Region and Planning 63

including: “sixty silver domes, bronze statues of all the gods, streets paved
with lead, crystal theatre, golden cock, crows each morning, multicolored
lamps, and food stalls.” This lexicon was designed to encapsulate the unique
characteristics of Diomira, enhancing its memorability for human readers
and facilitating distinctiveness in creative program inputs. An examination
of the outputs reveals distinct approaches to the prompt interpretation.
Midjourney renders a street view incorporating several identifying features,
such as a rooster, a dome, and a stall. DALL-E 3 similarly depicts a street scene
beneath the domes, but emphasizes the sunrise and prominently features
the rooster. In contrast, Stable Diffusion omits certain prompted elements,
including the rooster and a bench. Conversely, the GPT4-powered DALL-E
3 integrates nearly all descriptive elements, including bronze statues, silver
domes, lead-paved streets, and multicolored lamps, but notably excludes
the rooster. This comparative analysis underscores the diverse interpretative
capabilities and limitations of current AI image generation models in
translating textual descriptions into visual representations.
Figure 3 shows an analysis of AI-generated images depicting the
fictional city of Ersilia, characterized by its unique socio-spatial structure.
The prompt provided to the models highlighted key features: “ropes,
colors, relationships, dismantled houses, refugees, labyrinth of ropes.” This
lexicon encapsulates Ersilia’s defining element: a system where colorful
ropes physically manifest inter-household relationships, signifying kinship,
trade, power, and service. The city’s narrative revolves around the eventual
unsustainability of this complex web, as an overly dense network compels
citizens to abandon their homes, leaving behind a skeletal landscape of
interconnected poles - a poignant testament to societal complexity and its
potential pitfalls.
MidJourney V6 MS Designer (DALL-E 3)

Stable Diffusion XL Turbo GPT4 (DALL-E 3)

64 Özlem Kevseroğlu, Rifat Kurban

Figure 3. Visual representation of Ersilia’s urban portraits obtained by AI tools.

Former residents observe this tangled legacy from a distance, prompting

reflection on the intricate web they left behind. Analyzing the outputs, we
observe distinct interpretative approaches. Midjourney emphasizes the
perspective of the refugees, offering a close-up view of the abandoned
houses. DALL-E 3, in contrast, presents a multi-scalar and multi-angular
representation of the city, mirroring the labyrinthine network of ropes
that embodies Ersilia’s social fabric. Stable Diffusion focuses on the street
pattern, while the GPT4-powered DALL-E 3 model generates visuals similar
to those of DALL-E 3. This comparison illuminates the diverse capabilities
of current AI image generation models in translating textual descriptions
of complex socio-spatial systems into compelling visual narratives.
Figure 4 displays a compared look of AI-generated images that reflect
the fictitious town of Isidora. This city is known for its distinct architectural
and societal characteristics. The instruction given to the models
emphasized essential elements: “spiral staircases, seashells, telescopes,
violins.” The vocabulary used in this text creates a vivid image of a city that
is both intricately beautiful and filled with intense emotions, reflecting the
description of a traveler’s experience in Isidora.
 International Studies and Evaluations in the Field of City, Region and Planning 65

MidJourney V6 MS Designer (DALL-E 3)

Stable Diffusion XL Turbo GPT4 (DALL-E 3)

Figure 4. Visual representation of Isidora’s urban portraits obtained by AI tools.

Upon analyzing the outputs, we see clear and separate interpretations.

Midjourney provides a viewpoint at street level, capturing the city from
a position that is relatable to humans. DALL-E 3, on the other hand,
specifically emphasizes a grand structure featuring a notable helical staircase,
frequently depicted from a remote perspective. Stable Diffusion highlights
the urban vitality by focusing on the liveliness of street life. Finally, the
DALL-E 3 model, which is powered by GPT4, combines symbolic aspects.
It has a spiral staircase that resembles a seashell and a square formation in
front of it.
The square formation may represent passersby who are being
viewed, maybe alluding to the “telescopes” mentioned in the prompt. This
investigation highlights the different capacities of existing AI models in
66 Özlem Kevseroğlu, Rifat Kurban

converting written descriptions into various visual depictions of imaginary

urban landscapes.
MidJourney V6 MS Designer (DALL-E 3)

Stable Diffusion XL Turbo GPT4 (DALL-E 3)

Figure 5. Visual representation of Sophronia’s urban portraits obtained by AI

tools.

Figure 5 illustrates the synthetic images produced for the city of

Sophronia. The crucial distinguishing characteristics of the hint can be
enumerated as follows: The clue comprises various components, including
urban areas divided in half, the concept of duality, amusement park rides
such as roller coasters and carousels, circus apparatus like trapezes, and
various types of buildings such as stone, marble, and cement structures,
financial institutions, industrial facilities, grand residences, places where
animals are slaughtered, and educational institutions. Sophronia is a
city that is split into two distinct sections. 50% of the area is allocated to
 International Studies and Evaluations in the Field of City, Region and Planning 67

entertainment and grandeur, with exhilarating rides and attractions such

as roller coasters and Ferris wheels.
The remaining portion symbolizes the ordinary and pragmatic
facets of existence, constructed from durable substances and containing
establishments such as financial institutions, manufacturing facilities, and
educational institutions. The Midjourney outputs do not clearly observe the
distinction between the two cities, but they do have aspects that portray the
urban environment. The DALL-E 3 model exhibits two clear distinctions
but does not identify any difference in terms of function or language. The
Stable Diffusion model has generated a language that closely resembles the
output of The Midjourney. The GPT4 enabled DALL-E 3 images encode
the clear differentiation between the two cities using both functional and
chromatic elements.
3.2 Subjective Evaluation
In this section, the efficacy of different text-to-image generative AI
tools in creating visual representations of urban environments described
in literary texts are investigated. Specifically, we focus on five fictional cities
from Italo Calvino’s Invisible Cities: Argia, Diomira, Ersilia, Isidora, and
Sopronia.
The study employed a comparative framework, analyzing the outputs
of three prominent AI tools: Stable Diffusion XL Turbo, DALL-E 3 (accessed
via Microsoft Designer Image Creator and OpenAI GPT-4), and Midjourney
v6. A survey was conducted with a participant group of 20 individuals,
comprising 10 professional architects, urban design practitioners, planners
and 10 senior undergraduate students from the department of architecture.
Prior to the survey, participants were briefed on the content of
Invisible Cities and the objectives of the study. A questionnaire was then
distributed, featuring the name of each city, a concise explanation of the
research aim, and specific instructions for the participants. For each city,
the questionnaire presented with descriptive text from Invisible Cities and
visual representations, the images generated by the aforementioned AI
tools, labeled anonymously as A, B, C, and D to conceal the specific tool
used.
Participants were tasked with ranking the four visual representations
(A, B, C, and D) from most (1st) to least (4th) successful in capturing the
essence of the textual description for each city. This ranking system allowed
for a subjective assessment of how effectively each AI tool translated the
literary text into a visual representation.
This research aims to contribute to the growing body of knowledge
surrounding AI-driven image generation, focusing on its potential for
68 Özlem Kevseroğlu, Rifat Kurban

visualizing complex literary descriptions of urban portraits. The findings

will provide valuable insights into the strengths and weaknesses of different
AI tools, informing future applications in architectural design, urban
planning, and artistic explorations.

Figure 6. Subjective evaluation results for each city (1st ranking indicates the
best).

Figure 6 shows subjective evaluation results with boxplots for each city.
Boxplot is a visualization tool that summarizes the distribution of data in a
single graph. Shows the minimum, maximum, median, first quartile (Q1)
and third quartile (Q3) values of the data set. It also indicates the presence
of outliers. Blue box, in the figures, indicate the 1st and 3rd quartiles, black
lines indicate the minimum and maximum rankings, red line indicates the
median value, red cross indicates the outlier and green diamond indicates
the mean ranking value.
 International Studies and Evaluations in the Field of City, Region and Planning 69

Analysis of Figure 6 reveals distinct performance patterns across the

generative AI tools for each city depicted. For Argia, both Midjourney and
GPT4 DALL-E 3 demonstrate a median ranking of 2nd, while MS Designer
DALL-E 3 and SDXL Turbo fall at 3rd. Averaging the rankings reveals a
slight edge in performance for Midjourney over GPT4 DALL-E 3.
Similarly, in Diomira, Midjourney and GPT4 DALL-E 3 maintain
a median ranking of 2nd, with MS Designer DALL-E 3 and SDXL Turbo
again at 3rd. However, in this instance, average ranking indicates a marginal
advantage for GPT4 DALL-E 3, which emerges as the preferred tool for this
particular city, surpassing Midjourney.
A shift in performance emerges in the visualization of Ersilia. The
DALL-E 3 implementations, both MS Designer and GPT4, exhibit a
median ranking of 2nd, while Midjourney falls to 3rd and SDXL Turbo
ranks last at 4th. The average ranking for MS Designer DALL-E 3 slightly
exceeds that of GPT4 DALL-E 3. Notably, SDXL Turbo, identified as the
least successful tool for depicting Ersilia, presents closely aligned median
and average rankings.

Table 3. Overall subjective evaluation results (1st ranking indicates the best).

MidJourney DALL-E 3(De- SDXL Tur- DALL-E 3

V6 signer) bo (GPT4)
Median Rank-
3rd 2nd 4th 2nd
ing
Average Rank-
2.40th 2.30th 3.23th 2.07th
ing
Standard Dev. 0.92 1.02 1.04 1.11

Isidora echoes the pattern observed in Ersilia, with MS Designer

DALL-E 3 and GPT4 DALL-E 3 achieving a median ranking of 2nd,
Midjourney at 3rd, and SDXL Turbo at 4th. However, in this case, average
ranking favors GPT4 DALL-E 3 over MS Designer DALL-E 3.
Lastly, the visualization of Sophronia showcases a clear hierarchy in
median ranking. GPT4 DALL-E 3 secures the top position with 1st, followed
by MS Designer DALL-E 3 at 2nd, Midjourney at 3rd, and SDXL Turbo at
4th. A comparison of average rankings reveals comparable performance for
both DALL-E 3 implementations.
Overall, the subjective evaluations, given in Table 3, reveal a preference
for DALL-E 3, particularly the GPT4 implementation. DALL-E 3 (GPT4)
achieved a median ranking of 2nd and the best average ranking (2.07th),
indicating a tendency towards higher rankings. Midjourney secured a
70 Özlem Kevseroğlu, Rifat Kurban

median ranking of 3rd, with an average ranking of 2.40th. SDXL Turbo was less
favored, consistently receiving lower median and average rankings. These
results suggest that, based on subjective evaluations, DALL-E 3 (GPT4)
demonstrated stronger performance in visually translating the textual
descriptions of the selected cities from Italo Calvino’s Invisible Cities. The
standard deviation shows the scatter in the results, and if it is low, it means
that the results are clustered around the mean. From this perspective, the
standard deviation for all 4 tools is approximately close to 1 which means
there is a moderate degree of variation in the subjective rankings for each
AI tool. This suggests that while there is a general consensus regarding the
relative performance of the tools, individual preferences and interpretations
of the textual descriptions still play a significant role in the rankings.
4. Conclusions
In this study, the potential of text-to-image generative AI tools for
visualizing urban environments described in literary texts, specifically
focusing on five fictional cities from Italo Calvino’s Invisible Cities is
explored. The motivation stemmed from the increasing interest in utilizing
AI for creative applications in fields like architecture and urban planning.
We aimed to contribute to the growing body of knowledge surrounding AI-
driven image generation by comparing the performance of three prominent
tools: Stable Diffusion XL Turbo, DALL-E 3 (accessed through both
Microsoft Designer Image Creator and OpenAI GPT-4), and Midjourney
V6.
A subjective evaluation was conducted with 20 participants, including
architects, urban designers, planners, and architecture students. They
ranked images generated by each AI tool based on how effectively they
captured the essence of textual descriptions from Invisible Cities. Our
quantitative results revealed a preference for DALL-E 3, particularly the
GPT4 implementation. DALL-E 3 (GPT4) consistently achieved higher
rankings, indicating its superior ability to translate complex literary
descriptions into compelling visual representations.
The inclusion of text-to-image generative AI approaches in the field
of urban studies can create new opportunities in the dimensions of urban
planning, design, and research through the visualization of urban contexts,
the interaction with people’s views, and the improvement of participatory
planning processes. Urban studies suggest multiple new ideas presented in
Table 4 – these are the incorporation of visual elements to text documents,
in this case to enhance realistic portrayal of urban scenarios.
 International Studies and Evaluations in the Field of City, Region and Planning 71

Table 4. Possible usage areas of integrating text-to-image generative AI tools in

urban studies.

Idea Generation
Future Urban Landscapes
Visual Simulation and Impact Visualization
Scenario Planning Before and After Scenarios
Infrastructure Changes
Integration with GIS
Design Options
Visual Preferences Surveys
Enhanced Public En- Interactive Tools
gagement Awareness and Education Scenario Effects
Campaigns Digital Exhibitions
Facade Designs
Architectural Design
Design and Aesthetics Landscape Integration
Exploration Furniture and Layout
Public Space Design
Lighting and Atmosphere
Zoning Changes
Zoning Implications
Policy Development Density Studies
and Analysis Green Infrastructure
Environmental Impact
Sustainability Scenarios
Reconstruction
Research and Theoret- Historical Urban Changes
Change Over Time
ical Modeling
Theoretical Models Urban Theories Visualization
Plugin for Urban Design
Software and Tool Develop- Software (AutoCAD, Revit, or
Technical Integration ment SketchUp, ... etc.)
Strategies APIs for Custom Tools
Collaboration with AI Re-
Custom AI Models
searchers and Developer
The future of urban studies is set to change for the better with the
incorporation of text to image generative artificial intelligence tools which
provide a more graphic method of imagining the future, eliciting the
response of people,’ and creating evidence-based creative urban strategies.
In turn, the planning of ten morphologies and the design of cities become
less placid, and more firmly engaging and constructive which should result
in an improved environment.
AI content generators allow designers to think beyond the revised
design and have the ability to come up with different design options
that are acceptable and realistic also. It is possible that it can enhance
the initial stages in the architectural design process by providing both
stimulation as well as visualization. In this study, we attempt to contribute
to understanding of the developed nature of AI image in the immediate
environment for architectural practices. In addition, the study also provides
the possible future desire limit for the evolution of this technology and its
implementation in the design process. These further advances could help
architects and planners in production of more computerized images while
72 Özlem Kevseroğlu, Rifat Kurban

using less powerful hardware and fewer applications. Such development

would ease the design processes and compel us to major on the designing
and decision-making processes.
Acknowledgements
The authors would like to acknowledge the use of generative AI tools
to enhance the grammar and language quality of this study. These tools
were employed exclusively to refine the clarity and readability of the text,
without altering the content or the original findings presented.
 International Studies and Evaluations in the Field of City, Region and Planning 73

References

Albaghajati, Z. M., Bettaieb, D. M., & Malek, R. B. (2023). Exploring text-to-

image application in architectural design: insights and implications.
Architecture, Structures and Construction, 3(4), 475-497. https://doi.
org/10.1007/s44150-023-00103-x

Adams, L., Busch, F., Truhn, D., Makowski, M., Aerts, H., & Bressem, K. (2022).
What does DALL-E 2 know about radiology? https://doi.org/10.48550/
arxiv.2209.13696

Adetayo, A. (2024). Reimagining learning through ai art: the promise of DALL-E

and Midjourney for education and libraries. Library Hi Tech News.
https://doi.org/10.1108/lhtn-01-2024-0005

As, I., Pal, S., & Basu, P. (2018). Artificial intelligence in architecture: generating
conceptual design via deep learning. International Journal of Architectural
Computing, 16(4), 306-327. https://doi.org/10.1177/1478077118800982

Bajçinovci, B., Thaci, K., & Bajçinovci, B. (2017). Architectural reflection on Italo
Calvino’s Invisible Cities. Journal of Science Humanities and Arts - Josha,
4(1). https://doi.org/10.17160/josha.4.1.261

Brisco, R., Hay, L., & Dhami, S. (2023). Exploring the role of text-to-image AI
in concept generation. Proceedings of the Design Society, 3, 1835-1844.
https://doi.org/10.1017/pds.2023.184

Buehler, M. (2022). Generating 3D architectured nature-inspired materials and

granular media using diffusion models based on language cues. Oxford
Open Materials Science, 2(1). https://doi.org/10.1093/oxfmat/itac010

Calvino, I. (2010). Invisible cities. Random House.

Case, P. and Gaggiotti, H. (2014). Italo Calvino and the organizational

imagination: reading social organization through urban metaphors.
Culture and Organization, 22(2), 178-198. https://doi.org/10.1080/14759
551.2014.901325

Çelik, T. (2023). Architectural design method suggestion with machine learning

technologies based on voronoi diagram principle. Periodica Polytechnica
Architecture, 54(1), 12-28. https://doi.org/10.3311/ppar.21643

Du, H., Li, Z., Niyato, D., Kang, J., Xiong, Z., Huang, H., … & Mao, S. (2023).
Generative AI-aided optimization for AI-generated content (AIGC)
services in edge networks. https://doi.org/10.48550/arxiv.2303.13052

Hegazy, M. and Saleh, A. (2023). Evolution of AI role in architectural

design: between parametric exploration and machine hallucination.
74 Özlem Kevseroğlu, Rifat Kurban

MSA Engineering Journal, 2(2), 262-288. https://doi.org/10.21608/

msaeng.2023.291873

Jeong, H., Kwon, G., & Ye, J. (2023). Zero-shot generation of coherent storybook
from plain text story using diffusion models.. https://doi.org/10.48550/
arxiv.2302.03900

Kather, J., Laleh, N., Foersch, S., & Truhn, D. (2022). Medical domain knowledge
in domain-agnostic generative ai. NPJ Digital Medicine, 5(1). https://doi.
org/10.1038/s41746-022-00634-5

Kenig, N., Echeverria, J., & Vives, A. (2023). Human beauty according to
artificial intelligence. Plastic and Reconstructive Surgery Global Open,
11(7), e5153. https://doi.org/10.1097/gox.0000000000005153

Ko, H., Park, G., Jeon, H., Jo, J., Kim, J., & Seo, J. (2022). Large-scale text-to-
image generation models for visual artists’ creative works.. https://doi.
org/10.48550/arxiv.2210.08477

Li, H., Wu, Q., Xing, B., & Wang, W. (2023). Exploration of the intelligent-
auxiliary design of architectural space using artificial intelligence model.
Plos One, 18(3), e0282158. https://doi.org/10.1371/journal.pone.0282158

Lyu, Y., Wang, X., Lin, R., & Wu, J. (2022). Communication in human–AI
co-creation: perceptual analysis of paintings generated by text-to-
image system. Applied Sciences, 12(22), 11312. https://doi.org/10.3390/
app122211312

Ma, Y., Zhang, X., Sun, X., Ji, J., Wang, H., Jiang, G., … & Ji, R. (2023). X-mesh:
towards fast and accurate text-driven 3d stylization via dynamic textual
guidance.. https://doi.org/10.48550/arxiv.2303.15764

Manzolli, J. and Marinho, H. (2020). Invisible sounds for Calvino’s Invisible

Cities: collaboration in a creative process. Orfeu, 5(1). https://doi.
org/10.5965/2525530405012020554

Maksoud, A. (2024). Integrating an image-generative tool on creative design

brainstorming process of a Safavid mosque architecture conceptual form.
Buildings, 14(3), 843. https://doi.org/10.3390/buildings14030843

Mbalaka, B. (2023). Epistemically violent biases in artificial intelligence design:

the case of DALL-E 2 and starry AI. Digital Transformation and Society,
2(4), 376-402. https://doi.org/10.1108/dts-01-2023-0003

Moral-Andrés, F., Gómez, E., Reviriego, P., & Lombardi, F. (2022). Can artificial
intelligence reconstruct ancient mosaics?.. https://doi.org/10.48550/
arxiv.2210.06145

Mukhija, V. (2015). Learning from invisible cities: the interplay and dialogue of
order and disorder. Environment and Planning a Economy and Space,
47(4), 801-815. https://doi.org/10.1068/a130117p
 International Studies and Evaluations in the Field of City, Region and Planning 75

Newton, A. and Dhole, K. (2023). Is AI art another industrial revolution in the

making?.. https://doi.org/10.48550/arxiv.2301.05133

Numan, N., Giunchi, D., Congdon, B., & Steed, A. (2023). Ubiq-genie: leveraging
external frameworks for enhanced social VR experiences.. https://doi.
org/10.1109/vrw58643.2023.00108

O’Meara, J. and Murphy, C. (2023). Aberrant ai creations: co-creating surrealist

body horror using the DALL-E mini text-to-image generator. Convergence
the International Journal of Research into New Media Technologies,
29(4), 1070-1096. https://doi.org/10.1177/13548565231185865

Ploennigs, J. and Berger, M. (2023). AI art in architecture. AI in Civil Engineering,

2(1). https://doi.org/10.1007/s43503-023-00018-y

Podell, D., English, Z., Lacey, K., Blattmann, A., Dockhorn, T., Müller, J., ... &
Rombach, R. (2023). Sdxl: Improving latent diffusion models for high-
resolution image synthesis. arXiv preprint arXiv:2307.01952. https://doi.
org/10.48550/arXiv.2307.01952

Poredi, N. (2023). Ausome: authenticating social media images using frequency

analysis.. https://doi.org/10.1117/12.2663296

Ramesh, A., Dhariwal, P., Nichol, A., Chu, C., & Chen, M. (2022). Hierarchical
text-conditional image generation with clip latents.. https://doi.
org/10.48550/arxiv.2204.06125

Rane, N. L., Choudhary, S. P., & Rane, J. (2023). Integrating ChatGPT, Bard,
and leading-edge generative artificial intelligence in architectural
design and engineering: applications, framework, and challenges.
International Journal of Architecture and Planning, 3(2), 92-124. https://
doi.org/10.51483/ijarp.3.2.2023.92-124

Sadek, M. (2023). Artificial intelligence as a pedagogical tool for architectural

education: what does the empirical evidence tell us?. MSA Engineering
Journal, 2(2), 133-148. https://doi.org/10.21608/msaeng.2023.291867

Saharia, C., Chan, W., Saxena, S., Li, L., Whang, J., Denton, E., … & Norouzi, M.
(2022). Photorealistic text-to-image diffusion models with deep language
understanding. https://doi.org/10.48550/arxiv.2205.11487

Siemens, S. (2023). Synthetically generated microscope images

of microtopographies using stable diffusion.. https://doi.
org/10.1117/12.2673643

Singh, G., Deng, F., & Ahn, S. (2021). Illiterate DALL-E learns to compose.
https://doi.org/10.48550/arxiv.2110.11405

Steinfeld, K. (2023). Clever little tricks: a socio-technical history of text-to-image

generative models. International Journal of Architectural Computing,
21(2), 211-241. https://doi.org/10.1177/14780771231168230
76 Özlem Kevseroğlu, Rifat Kurban

Sukkar, A. W., Fareed, M. W., Yahia, M. W., Abdalla, S. B., Ibrahim, I., & Senjab,
K. A. K. (2024). Analytical evaluation of MidJourney architectural
virtual lab: defining major current limits in ai-generated representations
of Islamic architectural heritage. Buildings, 14(3), 786. https://doi.
org/10.3390/buildings14030786

Sun, Y., Ni, X., Feng, H., Lc, R., Lee, C., & Asadipour, A. (2022). Bringing stories
to life in 1001 nights: a co-creative text adventure game using a story
generation model., 651-672. https://doi.org/10.1007/978-3-031-22298-
6_42

Tomlinson, B., Black, R., Patterson, D., & Torrance, A. (2023). The carbon
emissions of writing and illustrating are lower for AI than for humans.
https://doi.org/10.21203/rs.3.rs-2724922/v1

Wang, B., Zhu, Y., Chen, L., Liu, J., Sun, L., & Childs, P. (2023). A study of the
evaluation metrics for generative images containing combinational
creativity. Artificial Intelligence for Engineering Design Analysis and
Manufacturing, 37. https://doi.org/10.1017/s0890060423000069

Yildirim, E. (2022). Text-to-image generation ai in architecture. Art and

Architecture: Theory, Practice and Experience, 97.

Youssef, A. (2023). Unleashing the ai revolution: exploring the capabilities and

challenges of large language models and text‐to‐image AI programs.
Ultrasound in Obstetrics and Gynecology, 62(2), 308-312. https://doi.
org/10.1002/uog.26297

Zeytin, E., Öztürk Kösenciğ, K., & Öner, D. (2024). The role of ai design
assistance on the architectural design process: an empirical research with
novice designers. Journal of Computational Design, 5(1), 1-30. https://doi.
org/10.53710/jcode.1421039