Fog and Edge Computing: A Comparative

Analysis of Approaches and Researches

Wan Aiman bin Ahmad Nazrin Nurizuan Nazrin Muhammad Irfan Umair bin Mohd
Wan Ibrahim Izwan bin Ahmad bin Komori Zaki bin Mohd Yaziz
2113615 Khalil 2113021 Razali 2123467
2112343 2119015

Abstract – Fog and edge computing have

emerged as pivotal technologies for handling II. FOG COMPUTING
the massive influx of data from IoT devices.
By processing data closer to its source, these A. Definition and Architecture
technologies mitigate bandwidth issues,
reduce latency, and enhance privacy. This
paper provides a detailed comparison of fog
and edge computing, discusses their
applications, and evaluates their impact
through research findings and case studies.
This article presents a comparative analysis
of fog and edge computing approaches,
highlighting their distinct characteristics,
applications, and research advancements. It
includes a detailed comparison table,
explanations of figures and diagrams.

Index Terms – Fog Computing, Edge

Computing, Internet of Things (IoT), Figure 1.0 : Fog Computing
Distributed Computing, Scalability, Latency
Fog computing, also known as edge computing,
decentralizes data processing by moving
I. INTRODUCTION resources closer to the source of the data. It's
comparable to having small data centers near
In today's digital environment, fog and edge devices, which reduces delays and security
computing are changing the way we manage issues.
data. Fog computing moves processing closer to In this architecture, edge devices collect data
where data is generated, and edge computing and do preliminary processing. Then, fog nodes,
takes it a bit further by placing it right on the which are closer to the devices, do more
devices themselves. This paper contrasts these complex tasks and may connect to the cloud for
two approaches, looking at how they work and further processing.
what research has been done. We can predict Overall, fog computing distributed computing
how they will influence computers in the future capability from the cloud to the edge, enabling
by recognizing their differences and similarities. faster and more efficient data processing.
B. Key Features and Benefits 9. Near the End User: Fog computing makes
computing more accessible to users, improving
speed and dependability while also enhancing
user experience.

C. Challenges and Future Directions

For the challenges of the fog computing,

there are the certain challenge:

Figure 2.0 : Fog Computing Benefits 1. Security Concerns: Preserving dispersed

systems at the edge has unique challenges,
1. Geographical Distribution: Fog computing including risks related to unauthorized
brings computation closer to where data is access, privacy of data, and vulnerabilities in
utilized, making processes faster by lowering devices.
trip time. 2. Interoperability: Ensuring seamless
communication and integration between
2. Decentralization: Rather of depending on a disparate devices and platforms can be
single large center, fog computing distributes challenging due to their disparate hardware,
duties over numerous smaller devices, protocols, and standards.
increasing system reliability. 3. Resource management: Taking into account
a variety of settings and fluctuating
3. Location Awareness: Fog systems understand workloads, it might be challenging to divide
where devices are and make judgments workloads and distribute resources among
depending on how near they are to one another. fog nodes in the best possible way.
This makes things more efficient.
And for the future directions:
4. Real-Time Interaction: Fog computing allows
devices to communicate with one another in real 1. Enhanced Security: Future developments in
time, which is useful for applications that fog computing will focus on strengthening
require quick responses such as self-driving cars. security measures including intrusion
detection, authentication, and encryption in
5. Storage Space Savings: By keeping data order to address new threats and
closer to where it is used, fog computing saves vulnerabilities.
space and improves networks by reducing data 2. Standardization and Interoperability: Efforts
transfer. should be made to define common
protocols, interfaces, and interoperability
6. Low Latency: Fog computing works in real standards to provide seamless integration
time, reducing delays by running operations and communication across diverse fog
close together. This is particularly useful for platforms and devices.
time-sensitive tasks like online gaming. 3. Integrating AI and machine learning
capabilities into fog computing
7. Heterogeneity Support: Fog systems provide infrastructures will enable intelligent
easy communication across a wide range of decision-making, predictive analytics, and
devices, including factories and phones. autonomous operation at the network edge.

8. Mobility Support: Fog systems have the

ability to track the movements of devices,
allowing you to maintain connectivity and
functionality even when you're on the go.
 III. EDGE COMPUTING
1. Speed: Edge computing could drastically cut
A. Definition and Architecture down and reduce latency and the time
required for data processing and response
delivery, which is very crucial for any
real-time applications such as autonomous
vehicles and smart grids.

2. Energy Savings: Edge computing will help

in saving bandwidth and reduce any
associated costs by minimizing the amount
of data that needs to be sent over into the
networks. Less data transmission means less
power is used in the network infrastructure
which leads to overall energy savings.
Figure 3.0 : Edge Computing Architecture
3. Enhanced Privacy and Security: Local data
Edge computing refers to the practice of processing could help enhance data security
processing data close to the source of data and privacy since sensitive information does
generation rather than far away from centralized not have to travel over the network anymore,
data centers or clouds. Simply said, this method hence reducing exposure to potential
brings the processing resources on the network's breaches.
edge, close to the device, in which it could be a
smartphone, sensor, or connected car. It is 4. Reliability: Decentralizing processing tasks
designed to reduce latency, sustain bandwidth across multiple edge nodes will enhance the
and also make applications and services as overall reliability of the system because
responsive as possible [1]. problems in one node would not affect the
entire network.
Edge computing architecture involves
processing data locally on edge devices or 5. Scalability: Organizations could easily scale
adjacent computers/servers, which will eliminate their operations by adding more edge
the need to transport data to remote data centers. devices and nodes that are based on the
This could cover a variety of edge devices that demand without any significant changes to
collect and process any preliminary data. These their infrastructure.
devices are linked to local edge nodes, which
would perform more complicated processing 6. Versatility: Wide application range from IoT
activities. Although these nodes periodically devices in smart homes to industrial IoT
connect to the cloud for additional processing or (IIoT) in manufacturing plants, edge
storage, they mostly would rely on local computing can be implemented across in a
computation to ensure fast response times. variety of different environments due to its
adaptability to various scales and
B. Key Features and Benefits requirements.

Figure 4.0 : Edge Computing Benefits

C. Challenges and Future Directions IV. COMPARISON TABLE

There are several challenges that could Feature Fog Computing Edge Computing
occur in edge computing: Multi-layered, Single-layer,
Architecture intermediate processing at the
1. Security Concerns: Although edge processing source
computing can improve security in certain Layered approach, Simple, direct
Deployment
contexts, it however would introduce complex deployment on
Models
challenges related to protecting numerous configurations devices
edge devices and defending against attacks Lower than cloud, Minimal, processed
Latency
higher than edge at the edge
in a distributed network.
Highly scalable across Scalable, limited by
2. Management Complexity: Managing a Scalability
multiple nodes device capabilities
diverse array of edge devices, ensuring they
Enhanced by localized High, but dependent
are all properly functioning and up-to-date Security
processing on device security
will bring very significant challenges
Figure 4.0. Fog vs Edge Computing
especially across complex environments and
technologies.
The table above compares fog and edge
3. Interoperability Issues: Ensuring that
computing based on several key elements which
devices from different manufacturers work
are the architecture, deployment models, latency,
together seamlessly would remain a
scalability, and security.
challenge due to the varying of standards
and technologies.
1. Architecture: Fog computing utilizes multi
layered architecture acting as intermediate
And for the future directions:
processing between cloud and edge which
allows for a distributed processing
1. Enhanced Security Measures: Future
framework. Edge computing on the other
advancements in edge computing will most
hand, features a single-layer processing
likely focus on developing a robust security
which occurs on the edge itself which means
framework tailored to its distributed nature.
the data will be handled locally without
These will include the new encryption
needing to traverse to a centralized or
methods and also smarter anomaly
intermediate nodes allowing for faster
detection systems.
response times and reduced data
2. Greater Standardization: There will be
transmission.
increased efforts toward standardizing the
protocols and interfaces to ensure better
2. Deployment Model: Fog computing uses a
interoperability and ease of management
layered approach with complex
across every different manufacturers'
configurations. The complexity will not only
devices.
add more flexibility but also adds control on
3. AI Integration and Machine Learning:
where and how data is processed. Edge
Integrating AI at the edge will significantly
computing on the other hand is deployed
enhance the capability of edge devices to
directly, reducing the need for extensive
make intelligent decisions, learn from new
modifications or additions which means it is
data and autonomously improve operations.
less complex as it aims to leverage the
existing capabilities of edge devices to their
fullest.

3. Latency: Fog computing offers lower

latency compared to traditional cloud as it is
closer to the data source but it has a higher
latency than edge as it involves some
 intermediate steps and travels. Edge The paper “All one needs to know about Fog
computing has the least latency as data does and Edge Computing Paradigms” [1] provides
not need to travel far from its origin the most comprehensive take on fog and edge
eliminating almost all transmission delays. computing, from theoretical frameworks up to
practical implementations and diversified
4. Scalability: Fog computing is highly applications like IoT, smart cities, and
scalable due to its ability for integration healthcare. This source will help readers
allowing for large-scale networks and understand the relationship of these two
handling vast amounts of data. Edge computing paradigms holistically and how they
computing is also scalable but is generally interrelate in all the three sectors.
limited by the computing power and storage
capabilities of individual devices. “Fog and Edge Computing: Concepts, Tools,
and Focus Areas” [2], on the other hand, is
5. Security: Fog computing will have enhanced highly customized for the readers with an
security due to localized data processing as interest in technological tools and trends
the data is less exposed to long distance accompanying the development of fog and edge
transmission. Edge computing has high computing. This source would be most useful
security as well, mainly because less data is for practitioners and scholars, oriented to
transmitted over networks, reducing the keeping up to date with new developments and
attack surface but in the end it all comes applied tools in this discipline.
down to each device’s individual security
measures. “Simulating Fog and Edge Computing
Scenarios” [4] focuses on the methodological
side, particularly discussing techniques of
All One Needs Fog and Edge simulation used in the analysis of performance
to Know about Computing: Simulating fog and scalability of fog and edge computing
Fog and Edge Concepts, and Edge systems. The book is tailor-made for researchers
Computing Tools, and Computing who are interested in simulation techniques from
Criteria Paradigms Focus Areas Scenarios the perspective of technical aspects associated
Extensive with modeling and simulation under these
Overview of Focus on
survey of fog
current simulation computing paradigms.
Scope and edge
concepts, tools, techniques and
computing
and trends challenges
paradigms
Architecture, Fundamental Scenario
Key deployment concepts, modeling,
Topics models, emerging tools, performance,
applications focus areas scalability
Researchers Practitioners
Researchers in
Target looking for a and researchers
simulation and
Audience comprehensive interested in
modeling
overview latest tools
Applicati IoT, smart Not
General IoT
on cities, application-spec
and real-time
Example healthcare, and ific, focuses on
data scenarios
s more scenarios
Theoretical and
practical Insight into Methodological
Contribu
distinctions technological advancements in Figure 6.0 Cloud vs Fog vs Edge Computing
tions
between advancements simulation
paradigms In the hierarchy of cloud, fog, and edge
Figure 5.0 Comparative Analysis of General Researches computing, each layer represents a distinct
 approach to data processing across the network, computing takes cloud computing to the network
optimized for tasks at varying distances from edge. The study highlights fog computing's
data sources. ability to reduce latency, optimize network
Cloud Computing Layer: Positioned as a layer of capacity, and improve data privacy and security
high centralization, cloud computing conducts by processing data closer to its source.
large data processing and complex analytics in
remote data centers. It offers excellent 2. The Advent of Edge Computing
computational resources, including great storage
ability, but with the disadvantage of greater In their foundational book, "Edge Computing:
latency, which is less favorable for real-time Vision and Challenges," W. Shi and coworkers
usage[1][6]. (2016) present edge computing as a disruptive
Fog Computing Layer: This is an intermediate technique that places data processing capabilities
layer acting as an interface. Fog computing at the network's edge. The findings show that
happens closer to the data sources than the cloud edge computing dramatically reduces latency
servers, but it is further or less decentralized and bandwidth utilization, allowing for efficient
than the edge devices. Processes data from handling of the massive data generated by edge
several edge devices with minimal latency into devices.
most substantial data processing and temporary
storage before transferring the data to the cloud 3. Impact on IoT Efficiency and Security
for further analysis or long-term storage. This
layer provides the local applications with faster A more recent study of edge computing's
response local to the user and higher local impact, focusing on IoT efficiency and security,
applications' privacy, but the layer is of found that local data processing not only
semi-centralized nature and hence involves improves system responsiveness but also
complex management[2][7]. strengthens security measures by reducing data
Edge Computing Layer: This is the second layer exposure to potential external threats.
of the architecture. It works with or near the
source of data and processes the data on the 4. Addressing Security Challenges in Fog
devices at the. Thus, it reduces latency since it Computing
processes data at the point of generation, which
is critical for real-time decision-making and The special security issues raised by the
bandwidth saving. However, its main limitation architecture of fog computing are mapped out in
comes from the computing and storage the paper "Security and Privacy in Fog
capabilities of the individual edge devices[3][8]. Computing: Challenges" by Y. Yang et al. (2017)
[22]. It draws attention to the problems of
V. RESEARCH APPROACHES & FINDINGS heterogeneity, decentralization, and
multi-tenancy, which call for creative solutions
In this section, we discuss the significant to protect privacy and data integrity.
research studies on fog and edge computing,
synthesize their findings, and implications of 5. Industrial IoT and Market Perspectives
these studies on technology, operations, and
strategic decision-making within the industry. A. Al-Fuqaha et al. (2015) [21] conducted a
thorough analysis that looks at IoT from the
A. Significant Research Contributions in standpoint of the industrial market. The survey
Fog and Edge Computing highlights the crucial role that fog and edge
computing play in controlling the flood of data
1. Understanding Fog Computing’s Role coming from industrial sources, which improves
the scalability and dependability of IoT
One of the seminal studies, "Fog Computing and solutions.
Its Role in the Internet of Things" by M. Chiang
and T. Zhang (2016), articulates how fog
Discussion of Findings and Their Implications staff to physically visit each unit and read the
meters in order to collect customers’ usage data.
1. Technological Implications
The research collectively demonstrates that both In a study by Rajaguru and Johansson, Opris et.
fog and edge computing effectively mitigate the al. debated whether the use of smart meters
latency issues prevalent in traditional cloud could lead to energy efficiency [24], by
setups. This is particularly vital for applications providing data in real time for demand-response
that depend on swift data analysis and programs. These programs have several
decision-making, such as in healthcare purposes; to efficiently control energy
monitoring and autonomous driving. Moreover, generation plants, and as a way for customers to
the local processing of data inherently improves regulate their use of energy. Through the
security and privacy, a boon for industries program, energy companies can tune their
handling sensitive information. generation plants to reduce their carbon
emissions which results in lesser pollution, while
2. Operational Implications customers can take matters into their own hands
From an operational standpoint, the shift by saving their energy consumption.
towards decentralized computing frameworks
like fog and edge computing introduces As the technology evolved, so did the strategies
complexities in data management and resource for managing energy distribution. Initially,
distribution. Efficient strategies for dynamic technical operations rely on predictive analysis
resource allocation and real-time processing are in their decision-making process, which is
imperative to harness the full potential of these vulnerable to errors and could lead to disastrous
technologies. outcomes such as blackouts. The introduction of
smart meters paved a way of enhanced
3. Strategic Implications telemetry, measuring not only energy
For businesses, the adoption of fog and edge consumption, but also other important
computing technologies means recalibrating parameters such as power factor, phase load,
investment strategies towards infrastructure that frequency among other things. Smart meters are
supports decentralized data processing. also equipped with radio-frequency transmitters,
Additionally, the growth of these technologies eliminating the need for manual meter readings
calls for the development of new standards and and allowing energy companies to acquire usage
regulatory frameworks to address emerging information in real time.
challenges in security, privacy, and
interoperability. Smart energy meters, as an edge device,
performs basic tasks such as monitoring the
energy usage and parameters among others.
VI. CASE STUDIES These meters collect data continuously and then
transmit it to the smart meter pole, the fog
Case Study 1: Smart meters device, that then aggregates this information.
The smart meter, in the case of TNB, is
One of the most notable examples of fog and connected to the central database either via fiber
edge computing uses is in the smart energy optic connection or through cellular connection
meters that phased out the old energy meters. In such as 4G [23], and sends the collected data in
Malaysia, the primary energy company, Tenaga bulk into storage for billing and analytical
Nasional Berhad (TNB) implemented the use of purposes. This structured integration not only
smart energy meters through their Advanced simplifies operations by automating data
Metering Infrastructure [23]. Traditionally, collection and transmission, but also enhances
energy meters operate on the electromechanical reliability and resiliency of energy distribution
principle, utilizing a rotating disc to measure and systems, ensuring continuous supply of energy
record the total energy consumption used in a for the wellbeing of people.
building unit. Energy companies used to deploy
Case Study 2: Autonomous vehicles
Furthermore, fog and edge computing enhance
Autonomous vehicles represent a significant privacy and security by processing sensitive data
advancement in the transportation industry, as it locally, reducing the risk of data breaches
used to pose extreme issues such as how a associated with transmitting data over long
system could process real-time data at high distances. The distributed nature of fog
speed, and also how a system could perform computing also adds an extra layer of resilience,
decision-making. At the core of an autonomous as data processing is spread across multiple
vehicle network is the fog and edge computing nodes, thereby mitigating the impact of a single
model. The vehicle itself acts as the edge device. point of failure.
Sensors, cameras and other onboard systems
continuously gather vast amounts of data Scalability is another significant advantage, as
regarding the vehicle state and the environment. for an example organizations can expand their
This data includes spatial information such as computing infrastructure by simply adding more
camera vision, radar distance measurement and edge devices and fog nodes based on demand
velocity information. Processing this without extensive modifications to the central
information directly on the vehicle could allow infrastructure. This flexibility allows for
for split-second decision-making, which is an cost-effective scaling while maintaining
essential factor for an autonomous system to performance.
navigate around safely, without relying on high
latency links to servers to determine decisions. Overall, the integration of fog and edge
computing into modern digital infrastructures
Further enhancing this architecture is the role of supports a more efficient, secure, and responsive
fog computing, which serves as an intermediate data processing landscape. These technologies
layer between the vehicles and the central cloud are crucial for advancing real-time analytics and
servers. In the case of autonomous vehicles, decision-making capabilities, paving the way for
roadside units (RSUs) act as fog nodes. These innovative applications in various sectors, from
nodes perform additional processing tasks that utility to automotive and beyond.
are not as time-sensitive as vehicle on-board
units (OBUs), but still require quick turnarounds
such as aggregating data from nearby vehicles to REFERENCES
manage traffic flow or updating navigation maps
with real-time road conditions. By doing this, [1] A. Yousefpour et al., “All one needs to know
fog computing reduces the bandwidth load on about fog computing and related edge
the core network, and speeds up the response computing paradigms: A complete survey,”
time for collective driving insights. Journal of Systems Architecture, vol. 98, pp.
289–330, Sep. 2019, doi:
10.1016/j.sysarc.2019.02.009.
VII. CONCLUSION
[2] V. Hurbungs, V. Bassoo, and T. P. Fowdur,
Fog and edge computing offer significant “Fog and edge computing: concepts, tools and
benefits across a range of applications by focus areas,” International Journal of
optimizing data processing capabilities near the Information Technology, vol. 13, no. 2, pp.
source of data generation. This proximity to data 511–522, Jan. 2021, doi:
sources greatly enhances the efficiency and 10.1007/s41870-020-00588-5.
responsiveness of systems that require real-time
data analysis, such as autonomous vehicles and [3] Y. Kalyani and R. W. Collier, “A systematic
IoT devices. By minimizing the distance data survey on the role of cloud, fog, and edge
must travel for processing, these computing computing combination in smart agriculture,”
models drastically reduce latency, improve Sensors, vol. 21, no. 17, p. 5922, Sep. 2021, doi:
system reliability, and conserve bandwidth. 10.3390/s21175922.
[4] S. Svorobej et al., “Simulating fog and Edge [12] Q. Qi and F. Tao, “A smart manufacturing
Computing Scenarios: An Overview and service system based on edge computing, fog
Research challenges,” Future Internet, vol. 11, computing, and cloud computing,” IEEE
no. 3, p. 55, Feb. 2019, doi: Access, vol. 7, pp. 86769–86777, Jan. 2019, doi:
10.3390/fi11030055. 10.1109/access.2019.2923610.

[5] G. Mujica et al., “Edge and Fog computing [13] J. Gedeon, F. Brandherm, R. Egert, T.
platform for data fusion of complex Grube, and M. Mühlhäuser, “What the fog?
heterogeneous sensors,” Sensors, vol. 18, no. 11, Edge Computing revisited: promises,
p. 3630, Oct. 2018, doi: 10.3390/s18113630. applications and future challenges,” IEEE
Access, vol. 7, pp. 152847–152878, Jan. 2019,
[6] J. P. L. Pérez, J. Díaz, J. Berrocal, R. doi: 10.1109/access.2019.2948399.
López-Viana, and Á. González-Prieto, “Edge
computing,” Computing, vol. 104, no. 12, pp. [14] J. C. Guevara, R. Da Silva Torres, and N. L.
2711–2747, Jul. 2022, doi: S. Da Fonseca, “On the classification of fog
10.1007/s00607-022-01104-2. computing applications: A machine learning
perspective,” Journal of Network and Computer
[7] J. M. G. Sánchez et al., “Edge Computing for Applications, vol. 159, p. 102596, Jun. 2020,
Cyber-physical Systems: A Systematic mapping doi: 10.1016/j.jnca.2020.102596.
study emphasizing trustworthiness,” ACM
Transactions on Cyber-physical Systems, vol. 6, [15] K. Dolui and S. K. Datta, “Comparison of
no. 3, pp. 1–28, Jul. 2022, doi: edge computing implementations: Fog
10.1145/3539662. computing, cloudlet and mobile edge
computing,” Dwasd, Jun. 2017, doi:
[8] K. Dolui and S. K. Datta, “Comparison of 10.1109/giots.2017.8016213.
edge computing implementations: Fog
computing, cloudlet and mobile edge [16] W. Shi, G. Pallis, and Z. Xu, “Edge
computing,” Global Internet of Things Summit Computing [Scanning the issue],” Proceedings
(GIoTS), Jun. 2017, doi: of the IEEE, vol. 107, no. 8, pp. 1474–1481,
10.1109/giots.2017.8016213. Aug. 2019, doi: 10.1109/jproc.2019.2928287.

[9] A. Al-Qamash, I. Soliman, R. Abulibdeh, [17] “ICFEC 2024 - 8th IEEE International
and M. Saleh, “Cloud, Fog, and Edge Conference on Fog and Edge Computing 2024.”
Computing: A Software Engineering https://icfec2024.ontariotechu.ca/
Perspective,” International Conference on
Computer and Applications (ICCA), Aug. 2018, [18] A. Sunyaev, “Fog and Edge Computing,” in
doi: 10.1109/comapp.2018.8460443. Springer eBooks, 2020, pp. 237–264. doi:
10.1007/978-3-030-34957-8_8.
[10] P. PunithaIlayarani and M. M. Dominic,
“Anatomization of Fog Computing and Edge [19] I. Stojmenović, S. Wen, X. Huang, and H.
Computing,” IEEE International Conference on Luan, “An overview of Fog computing and its
Electrical, Computer and Communication security issues,” Concurrency and Computation,
Technologies (ICECCT), Feb. 2019, doi: vol. 28, no. 10, pp. 2991–3005, Apr. 2015, doi:
10.1109/icecct.2019.8869125. 10.1002/cpe.3485.

[11] “A smart manufacturing service system [20] F. Bonomi, R. A. Milito, J. Zhu, and S.
based on edge computing, fog computing, and Addepalli, “Fog computing and its role in the
cloud computing,” IEEE Journals & Magazine | internet of things,” MCC ’12, Aug. 2012, doi:
IEEE Xplore, 2019. 10.1145/2342509.2342513.
https://ieeexplore.ieee.org/document/8740963
[21] Ala Al-Fuqaha; Mohsen Guizani; Mehdi
Mohammadi; Mohammed Aledhari; Moussa
Ayyash, “Internet of Things: A Survey on
Enabling Technologies, Protocols, and
Applications,” Volume: 17, Issue: 4,
Fourthquarter 2015, doi:
10.1109/COMST.2015.2444095.

[22] Mithun Mukherjee, Rakesh Matam, Lei

Shu, Leandros Maglaras, Mohamed Amine
Ferrag, Nikumani Choudhury, and Vikas Kumar,
“Security and Privacy in Fog Computing:
Challenges”, VOLUME 5, 2017 doi:
10.1109/ACCESS.2017.2749422

[23] M H F Mohd Subhi, “Communication

Technology Options for Better Customer
Experience - The Case of Advanced Metering
Infrastructure (AMI) at TNB,” ILSAS
International Journal of Learning and
Development, vol. 1, no. 1, 2020.

[24] S. Rajaguru, B. Johansson, and M. Granath,

“Exploring Smart Meters: What We Know and
What We Need to Know,” in Lecture Notes in
Business Information Processing, Springer
Science and Business Media Deutschland
GmbH, 2023, pp. 105–120. doi:
10.1007/978-3-031-43126-5_