energies

Article
The Role of AI in Improving Environmental Sustainability: A
Focus on Energy Management
Naiara Uriarte-Gallastegi 1 , Germán Arana-Landín 2, * , Beñat Landeta-Manzano 1 and Iker Laskurain-Iturbe 2

1 Faculty of Engineering from Bilbao, Department of Business Management, University of the Basque Country,
Ingeniero Torres Quevedo Square, 1, 48013 Bilbao, Spain; [email protected] (N.U.-G.);
[email protected] (B.L.-M.)
2 Faculty of Engineering from Gipuzkoa, Department of Business Management, University of the Basque
Country, Europa Square 1, 20018 San Sebastian, Spain; [email protected]
* Correspondence: [email protected]; Tel.: +34-943018674

Abstract: This research addresses the increasing importance of understanding how Artificial Intelli-
gence can facilitate the transition of companies to a Circular Economy model. This study focuses on
energy management, examining its impact on efficiency and emissions across a multi-case analysis of
18 projects in diverse sectors. The findings indicate that Artificial Intelligence positively influences
both variables, with variations across applications and sectors. Notably, Artificial Intelligence signifi-
cantly enhances energy efficiency in four out of six sectors, achieving over 5% improvement in half of
the projects. Regarding emissions, positive effects are observed in 15 out of 18 projects, resulting in
over 5% reductions in seven cases. Artificial Intelligence plays a pivotal role in emissions reduction in
the Design and Energy sectors, with some projects achieving over 20% reductions. Additionally, this
study explores how improved energy efficiency positively affects strategic business variables, such as
cost, quality, and delivery time. The impact on emissions contributes to reducing occupational risks,
particularly those associated with chemical and biological agents. Although managers are satisfied,
measures need to be taken to overcome the lack of employee acceptance. These findings are of great
interest to the stakeholders involved in the integration of Artificial Intelligence into companies.

Citation: Uriarte-Gallastegi, N.; Keywords: Artificial Intelligence; Circular Economy; Industry 4.0; energy management; energy
Arana-Landín, G.; Landeta-Manzano, efficiency; emissions; non-energy benefits
B.; Laskurain-Iturbe, I. The Role of AI
in Improving Environmental
Sustainability: A Focus on Energy
Management. Energies 2024, 17, 649. 1. Introduction
https://doi.org/10.3390/
Humanity is currently facing an unprecedented challenge: climate change, an environ-
en17030649
mental crisis that threatens not only the planet’s biological diversity but also the survival
Academic Editor: Brian D. Fath and wellbeing of its inhabitants. This global challenge, exacerbated by decades of unsus-
tainable industrial and economic practices, has prompted an urgent call for collective action
Received: 30 December 2023
and innovation in all sectors of society, with sustainable economic practices including areas
Revised: 26 January 2024
such as health and clean primary resources [1].
Accepted: 26 January 2024
Published: 30 January 2024
Climate change, driven primarily by greenhouse gas emissions, has generated scientific
and political consensus on the need for a profound transformation in the way we produce
and consume energy [2]. In this context, energy efficiency measures are considered a first
line of action and presented as a financially viable and necessary solution for an energy
Copyright: © 2024 by the authors. transition [3]. The “2023 United Nations Climate Change Conference”, more commonly
Licensee MDPI, Basel, Switzerland. known as COP28, has highlighted the need for deepening, rapid, and sustained reductions
This article is an open access article in greenhouse gas emissions. The agreement reached by the 190 delegations included a
distributed under the terms and 43% reduction by 2030 and a 60% reduction by 2035 compared to 2019 and reaching net
conditions of the Creative Commons zero emissions by 2050, among other measures set out to triple global renewable energy
Attribution (CC BY) license (https:// capacity and double the global average annual rate of energy efficiency improvements by
creativecommons.org/licenses/by/ 2030 [4].
4.0/).

Energies 2024, 17, 649. https://doi.org/10.3390/en17030649 https://www.mdpi.com/journal/energies

Energies 2024, 17, 649 2 of 15

Businesses, as central players in the global economy, have a crucial role to play
in this transformation. Adopting sustainable business practices is not only an ethical
responsibility but also an economically smart strategy. Studies have shown that companies
that incorporate sustainable practices improve their long-term financial performance and
strengthen their resilience to the risks associated with climate change [5].
A key strategy in the pursuit of corporate sustainability is improving energy efficiency.
Reducing energy consumption not only lowers operating costs but also contributes sig-
nificantly to reducing carbon emissions [6]. In parallel, emission reduction has become a
priority for companies, not only due to increasing regulation and social pressure but also to
anticipate and adapt to an increasingly sustainability-oriented market [7].
In several regions of the world, especially in those countries that are leaders in climate
change policies, these strategies are being supported by governmental programs such as the
EUs “Horizon Europe”, the US “Better Buildings Initiative” [8], China’s “National Action
Program on Climate Change” [9], Japan’s “Strategic Energy and Environmental Innovation
Program” [10], and the "Clean Energy Finance and Investment Mobilisation Programme”
of the OECD [11]. With financial backing from Australia, Canada, Denmark, Egypt, and
Germany, the program proposes to strengthen national conditions for attracting investment
and financing in energy efficiency and clean industry. These programs provide financial
incentives, technical support, and regulatory frameworks that facilitate the adoption of
energy efficiency and clean industry.
These programs provide financial incentives, technical support, and regulatory frame-
works that facilitate the adoption of sustainable practices by companies.
In the area of technology, Industry 4.0 (I4.0) has introduced several innovative tools
that can play a key role in the transformation towards more sustainable business practices.
Among these, Artificial Intelligence (AI) stands out for its ability to optimize processes,
reduce resource consumption, and improve decision-making [12].
AI, applied to energy management, has the potential to transform radically the way
companies consume and manage energy. By analyzing large volumes of data and Machine
Learning (ML), AI can identify consumption patterns, predict future needs, and optimize
resource utilization, resulting in greater energy efficiency [13]. Moreover, in the field of
emissions reduction, AI can play a crucial role by facilitating the monitoring and manage-
ment of carbon emissions as well as contributing to the development of more sustainable
products and processes [14].
Numerous studies have been conducted that have examined the potential applications
of AI in these areas. These studies usually identify impact, which is mostly derived from
cited literature or anecdotal evidence and is rarely measured or quantified. Therefore, a
comprehensive and systemic assessment of the actual impact of AI, as well as best practices
for its implementation, is still needed [15]. There is a notable gap in our knowledge at the
business level about AIs real and quantifiable contribution to energy efficiency and emis-
sions reduction. Nevertheless, this knowledge gap is more noticeable when we introduce
into the equation strategic dependent variables for companies, such as product cost, prod-
uct quality, lead-time, risks, and satisfaction of managers, employees, and customers [13].
However, it is critical to understand the derivative in relation to these variables from the
point of view of decision-making on the consequences of AI, energy efficiency, and emis-
sion reduction at the industrial operational level. Energy allocations and environmental
regulations directly influence energy efficiency, thus seeking improvements in these aspects
and contributing to the country moving towards a low-carbon energy transition [16].
This study aims to address this gap at the micro-level and through real-life cases
deployed in large industrial corporations with AI developed by highly innovative com-
panies. This study explores whether this technology can aid the business and social shift
towards a more circular model. Its focus is on enhancing energy management efficiency
and reducing emissions, ultimately aiming to achieve the decarbonization of companies.
In addition, it is shown how key variables for business management are affected in this
Energies 2024, 17, 649 3 of 15

adoption process so that managers can then effectively apply AI to drive the transition to a
more sustainable model.
To achieve these objectives, this paper is structured in six sections. After this introduc-
tion, the Section 2 presents a review of the relevant academic literature on the subject. The
Section 3 describes the methodology used in this research. Then, in the Section 4, the results
obtained are presented. Finally, Section 5 discusses the findings and Section 6 presents
the conclusions of this research, as well as the limitations identified and possible areas for
future research.

2. Literature Review
The impact of I4.0 and digital technologies (DTs) on the Circular Economy (CE) is a
complex and multifaceted issue. These technologies can improve circulation processes, con-
tributing to sustainability by achieving economic, social, and environmental benefits [17],
mainly when integrated with the principles of reuse, remanufacturing, and recycling [18,19].
However, a more nuanced understanding is needed, as certain technologies, such as sen-
sors, RFID, and AI, turned out to be the most relevant, while others had a negative or no
effect [20]. Although, on the other hand, if we take AI, for example, academic literature has
highlighted the importance of combining AI with other technologies (IoT, Big Data (BD),
robotics, among others) in areas such as economics, quality, design, energy management,
and safety in an industrial environment and applied in various sectors [21].
The use of AI enables the adoption of practices that take advantage of data availability
and can also maximize the use of available resources, minimize emissions, and contribute
to energy management [22]. Precisely, AI has a significant impact on the energy efficiency
of companies, particularly in the manufacturing sector [23,24]. AI, in combination with
other I4.0Ts, can improve efficiency in an average range of 1525% in the processes in which
it is implemented [12]. Liu and Liu [25] found that AI, particularly in the form of industrial
robots, could improve energy efficiency and reduce energy intensity in manufacturing firms.
The literature highlights the relevance of AI for energy optimization, business internation-
alization, and resource efficiency, with AI applications such as predictive maintenance and
production planning contributing to increased energy efficiency [26]. This effect is more
pronounced in labor and technology-intensive sectors [27], demonstrating how the IoT and
ML can be used to create autonomous energy management systems, improving energy effi-
ciency in smart environments. In the current industrial context, such applications in many
small and medium-sized enterprises can help achieve a shift towards more sustainable
and circular business practices that promote more efficient and intelligent use of energy
resources [28]. However, the effectiveness of AI in reducing energy intensity varies across
industries, with a greater impact observed in capital-intensive sectors [29].
In addition, important barriers to AI implementation, such as the digitization process
and semantic interoperability, both crucial to achieving CE, need to be considered [30].
On the other hand, the implementation of AI technologies to monitor energy param-
eters could reduce each country’s greenhouse gas emissions by 70% [31] and, in turn,
improve productivity [32]. In line with these findings, the World Economic Forum stated
that AI-based systems, thanks to the growing amount of data and increasingly advanced
algorithms, could contribute to a 4% reduction in global emissions by 2030 [33].
In this regard, several studies have explored the impact of AI on the intentionality of
carbon emissions, particularly in the context of China and especially in large cities with
advanced technology and infrastructure [34]. Chen et al. [35] found that AI significantly
reduces carbon intensity, with the latter emphasizing the role of AI in optimizing industrial
structure and fostering green technology innovation. To achieve these results, the need
to innovate in emission-intensive sectors such as waste management or public transport
to make them energy efficient, smart, sustainable, and cost-effective is highlighted [32].
The complexity of the relationship between AI and carbon emissions intensity highlights
the importance for regulators and policymakers to consider the impact of AI technologies
Energies 2024, 17, 649 4 of 15

on the energy system as a fundamental systemic element for decreasing greenhouse gas
emissions [36].
These studies highlight the potential of AI to improve energy efficiency and mitigate
corporate carbon emissions, but also the need for more research to understand the com-
plexities of this relationship and to design industry-specific strategies. There are hardly
any articles quantifying the real influence of AI on the transformation of energy and emis-
sions management to achieve an EC model with the goal of decarbonizing companies.
Consequently, this research questions (RQ) are formulated as follows:
− RQ 1: To what extent can AI contribute to the improvement of energy efficiency?
− RQ 2: To what extent can AI contribute to the reduction of emissions?
Based on these questions, at the beginning of this research, in the field phase, unantici-
pated evidence was collected, evaluated, and assessed as being of great interest to academia
but also to business managers. Specifically, some data and information were found that
pointed to the influence of AI on energy efficiency and emissions, with an impact on some
strategic business variables.
Energy efficiency provides a broad set of benefits in multiple areas (e.g., cost, quality,
lead-time, risks, and satisfaction) beyond mere energy savings, known as non-energy
benefits (NEB) [37]. Therefore, considering “energy” in a broader spectrum within the
industry and assessing the implications of energy efficiency on relevant strategic variables
can be crucial for business sustainability [12,38].
The literature has highlighted the potential benefits of I4.0Ts, such as AI, the Internet of
Things (IoT), Cyber-Physical Systems, and Industrial Robots, to improve the performance
of manufacturing industries. However, they forget about the NEBs. Katic et al. [39] and
Nota et al. [40] both emphasize the importance of understanding the impact of Energy
Efficiency Measures (EEMs) on production resources and the potential for reducing energy
consumption in manufacturing processes using I4.0Ts. Medojevic et al. [41] further support
this, highlighting the correlation between I4.0 concepts and manufacturing energy and
environmental management systems. Ghobakhloo et al. [42] expand on this by explaining
how I4.0 contributes to energy sustainability, particularly through the digitalization of the
energy demand sector and the introduction of smarter and more sustainable products.
However, despite the many benefits attributed to energy efficiency, the pace of NEB-
related improvements has slowed down in recent times [43], which has contributed to an
“energy efficiency gap” [32]. This fact applies equally to the case of emissions that, although
closely linked to energy consumption, have not been the subject of research in academia,
although energy efficiency and emissions may have profound implications for strategic
variables and may be of great interest to key industrial decision-makers [44].
In agreement with the proposal presented by Hassan and Trianni [13], these researchers
propose a framework to evaluate the contribution of I4.0Ts to the increase of operational per-
formance in energy efficiency initiatives. Hassan and Trianni [13] argue that technologies
such as AI and IoT positively influence the operational performance of energy efficiency
measures, generating improvements in Overall Equipment Effectiveness (OEE), productiv-
ity, and reduced operating costs. The promotion of sustainable development (SD) globally
involves adopting undertakings and production processes focused on environmental sus-
tainability [45].
Thus, according to the literature review conducted on the subject, the following
questions can be posed, with the aim of providing proven evidence of the influence of
energy efficiency and emissions through AI on some strategic business variables:
− RQ 3: To what extent can AI contribute through energy efficiency improvements on cost,
quality, lead time, risk, and satisfaction variables?
− RQ 4: To what extent can AI contribute through the reduction of generated emissions in
the variables of cost, quality, lead-time, risks, and satisfaction?
Table 1 below presents a review of the main academic articles and studies on the
influence of AI on carbon emissions and energy efficiency in industry. It shows the knowl-
Energies 2024, 17, 649 5 of 15

edge gap they detected, the objectives they pursued, the methodology they used, the main
contributions they made, and the main limitations, weaknesses, or unresolved aspects of
their research.

Table 1. Main academic contributions on the role of AI in the transformation of energy and emissions
management to achieve a CE model.

Ref. Knowledge Gap Methodology Main Contributions Main Limitations

Systematic Literature Review
AI helps to close the loop in Cover a broader range of
The adoption of EEMs in (SLR), N = 25, Scopus and
[13] water management (improving I4.0 cross-cutting
industry and I4.0Ts WoS (2022–2023).
OEE, productivity, and costs). technologies
Semi-struct. Interviews.
Findings may not be
generalizable to other
Multiple case studies, N = 27, AI ↓ material, energy use, waste,
[23] Influence of I4.0Ts on CE. contexts.
(2018–2021). and emissions.
Limit quantitative
analysis
Integrate I4.0 and CE into SLR, N = 90, Scopus and I4.0Ts help to transform waste The model overlooks
[24]
the supply chain network. WoS (2011–2020) into new products in a circle. social responsibility.
Panel data, 13 industries in AI has an inhibitory effect on Limited data sources.
Impact of AI on the carbon
[25] China (2005–2016). carbon. Robots can contribute to Other technologies can
footprint.
STIRPAT model method. GDP growth. distort the data.
A holistic perspective
SLR, N = 581, WoS and AI intelligent algorithms enable I4.0Ts must be developed
[27] integrates I4.0Ts in the
Scopus (2017–2022). prediction and trading. jointly to avoid obstacles.
energy sector.
Ambiguous
Firms’ limited non-energy
Measuring and monetizing Survey, N = 31, face-to-face understanding of
benefits to profitability in
[37] non-energy benefits and interviews, qualitative and potential non-energy
energy-efficiency investments
sustainability performance statistical analysis. benefits among
and investment decisions.
respondents.
EEMs influence production Limited Sample Size.
Incorporating AI into
Survey, N = 217 SMEs resources, irrespective of energy Self-reported data.
companies’ existing
BISNODE GVIN database, intensity. There is a varied Limited Slovenian
[38] information systems for
Slovenia. perception of resource manufacturing sector.
integrated energy
Cluster analysis. importance and management EEM adoption is not
management.
efficiency. discussed in detail.
Limited generalizability.
Impact of EEMs on The preliminary conceptual
Current research offers
shop-floor operations and framework combines EEM
incomplete advice.
[39] the operational Theory building. adoption, production resources,
Difficulty in
performance of industrial and operational performance for
comprehensively
organizations. a structured assessment.
assessing EEMs.
Application of the OEE SLR, N = 128, Elsevier, Integrating OEE with I4.0Ts
Limited data sources.
indicator, its relationship to Emerald, IEEE, Springer and improves accuracy, enables
Only articles with more
[40] I4.0, and its contribution to Taylor and Francis real-time production monitoring
than 20 citations were
value co-creation in the (2015–2020). PRISMA and control, and involves
considered.
industry method. stakeholders.
Nominal Group Technique, Demonstrates how I4.0
Primarily focusing on
I.0 technology trends and N = 8 experts, Matrice contributes to energy
European experts limits
industry digitization d’Impacts Croisées sustainability.
[42] the generalizability.
enhance energy Multiplication Appliquée I4.0 promotes energy
Need of further research
sustainability. aun Classement analysés sustainability, including the
to address these gaps.
(MICMAC). digitalization of energy.
The impact of AI on SLR, N = 482, Scopus A lack of legislation does
[43] sustainable (1994–2022). AI and ML stand out in SD. not promote sustainable
entrepreneurship PRISMA method. business.
Energies 2024, 17, 649 6 of 15

Table 1. Cont.

Ref. Knowledge Gap Methodology Main Contributions Main Limitations

Data from publicly
AI for energy efficiency Smart networks ↑ energy
Statistical regression, N = 30, traded energy companies
[44] and the impact on management and ↑ total factor
Panel data (2006–2019). could facilitate a detailed
productivity. efficiency.
analysis.
Note: ↑ increase, ↓ decrease.

Our research helps fill a vital gap by focusing specifically on the role of AI within the
broader spectrum of I4.0Ts to achieve a CE. It provides more specific knowledge on how
AI can be effectively applied for decarbonization and sustainable practices in companies.
The solid foundation established by existing studies, combined with our specific focus on
AI, brings valuable new insights to the field, informing future research and policy.

3. Materials and Methods

This research process was conducted in four phases. The initial phase consisted of
a literature review on the relationship between I4.0Ts and CE, more specifically, on the
potential of I4.0 to influence positively the transition process of companies towards a circular
business model. This analysis detected the growing interest in recent years in exploring
the capabilities of AI and its potential to support and drive change. We also noted and
were surprised by the scarcity of studies addressing the role of AI in energy and emissions
management, being two fundamental interrelated areas in the environmental behavior
of companies and their circular performance, strongly linked to climate change and UN
Sustainable Development Goals 7, “Affordable and Clean Energy”, and 13, “Climate
Action” [4]. Because of this initial analysis, we set out to find out what role AI is playing
in companies and how they are managing the energy they consume and the emissions
they emit directly. However, because of its ambition, this objective required us to assess
the possibilities of achieving it, first by accessing the necessary, sufficient, and reliable
direct sources of data and information. If we wanted such data and information, we
had to look to the source, the company, in action. Nevertheless, access to such data and
information and the availability of resources to do so had to be considered because the
sources of evidence for such data and information were also different in nature, quantity,
and quality. Their collection and subsequent analysis would require resources, and the
scope of the investigation therefore had to be determined. Taking all this into account,
the case study was considered the most appropriate methodology for approaching this
study phenomenon, given that it is a methodology that “investigates a contemporary
phenomenon in its real context” [46] and allows the use of multiple sources of evidence,
quantitative and/or qualitative, simultaneously [47].
Once the methodological design of this research was defined, the unit of analysis and
the selection of cases were defined. This was based on a theoretical and logical, rather than
statistical and random, sampling, trying to choose those cases that could offer a greater
opportunity for learning [45]. Accessibility to the information required (although in some
cases it have been required to keep some of the data collected confidential), an adequate
willingness on the part of the people most directly involved with the management of AI
projects, and the implementation of AI-based solutions with at least 1 year of experience in
full operation were required conditions.
The selected cases correspond to companies participating in the BIND 4.0 program
to support technology-based start-ups organized by the Basque Business Development
Agency (SPRI) in the Basque Country, Spain. It is a program recognized as “Exceptional
Industry Support” in the start-up Ecosystem Stars 2023 awards given by the International
Chamber of Commerce and Mind the Bridge in Europe. The program, acknowledged as
“Exceptional Industry Support” in the 2023 start-up Ecosystem Stars awards by the Interna-
tional Chamber of Commerce and Mind the Bridge in Europe, facilitates the connection and
advancement of open innovation projects for disruptive start-ups, primarily within venture
 Agency (SPRI) in the Basque Country, Spain. It is a program recognized as “Exceptional
Industry Support” in the start-up Ecosystem Stars 2023 awards given by the International
Chamber of Commerce and Mind the Bridge in Europe. The program, acknowledged as
“Exceptional Industry Support” in the 2023 start-up Ecosystem Stars awards by the Inter-
Energies 2024, 17, 649 national Chamber of Commerce and Mind the Bridge in Europe, facilitates the connection 7 of 15
and advancement of open innovation projects for disruptive start-ups, primarily within
venture clients, notably large tractor companies. A study was conducted on 18 projects
covering a variety
clients, notably of tractor
large AI applications across
companies. sectors
A study wassuch as Health,
conducted Energy,
on 18 Maintenance
projects covering a
and Security, Supply and Distribution, Image Processing, and Design, as detailed in Fig-
variety of AI applications across sectors such as Health, Energy, Maintenance and Security,
ure 1. and Distribution, Image Processing, and Design, as detailed in Figure 1.
Supply

sectors of
Figure 1. AI projects conducted in 6 sectors of business
business activity.
activity.

This
This study
study aimed
aimed for
for aa balance
balance between
between thethe number
number of of cases
cases analyzed
analyzed and
and the
the depth
depth
and
and breadth
breadth of
of the
the analysis,
analysis, prioritizing
prioritizing aa cross-case
cross-case (comparative)
(comparative) analysis
analysis based
based onon the
the
information
information provided
provided byby start-ups
start-ups and
and venture
venture clients.
clients. This
This approach
approach sought
sought to
to maintain
maintain
constructive
constructive validity
validity and
and reliability
reliability without
without compromising
compromising the the evidence
evidence chain
chain [46].
[46]. While
While
considering
considering a larger number of cases would enhance certainty, it could sacrifice depth
a larger number of cases would enhance certainty, it could sacrifice depth and
and
breadth. Notably, there is no consensus on the ideal number of cases [46,48]. Following
Eisenhardt’s [49], the target was set at three cases, reaching theoretical saturation in all
except the “Energy” and “Maintenance and Safety” sectors. An additional case was
conducted in the “Energy” sector due to its unique nature. However, constraints prevented
further cases in the “Maintenance and Safety” sector. Despite not meeting the target number,
the two cases in this sector offered valuable insights into high-interest AI applications,
emphasizing the significance of the findings.
Subsequently, the third phase of this research, the field phase, began with the collec-
tion of evidence. To ensure conclusive findings, the data collection phase employed the
Energies 2024, 17, 649 8 of 15

concept of triangulation [50,51]. This involved utilizing various methods and sources for
evidence, both quantitative and qualitative, gathered from start-ups and venture clients.
Documentation, including reports and internal studies, archives containing presentation
files, image and sound files, in-depth interviews, and direct observation through visits to
venture client facilities, were conducted.
Once the characterization of the cases investigated was completed, the last and fourth
phases were carried out. The individual analysis of the cases and the cross-case analysis
were carried out following the methodological indications of Miles et al. [51]. It consisted of
examining, categorizing, and tabulating the evidence collected, trying to identify common
patterns of behavior between the cases and the sectors, and determining the connection
between the data and this research questions at the company and sector level. The results
of the cross-case analysis are summarized in Section 4, “Results”. This section shows the
degree of influence of the AI with respect to energy efficiency and emissions generated
from the venture clients (per case and per sector) as a five-point Likert type item ranging: 0,
no influence; 1, low influence (less than 1%); 2, medium influence (1% or more and less
than 5%); 3, high influence (5% or more and less than 20%); and 4, very high influence (20%
or more). These are numerical values calculated based on empirical evidence collected
in the field phase, shared with the main actors in the cases, and contrasted with those of
two external academic evaluators. In no case were data or information about a negative
influence collected.
On the other hand, the main results in aggregate form on how the influence of AI
on these energy efficiency and emissions generated affects key performance indicators
related to cost, quality, lead-time, risks, and company satisfaction are presented. In this
case, it was agreed to also employ a five-point Likert-type scale, with ‘0’ as the central value,
with ‘−2’ representing a generalized negative impact and ‘2’ representing a generalized
positive impact. Given the exploratory nature of our purpose and the complex nature of
the phenomenon, it was agreed to perform the cross-case analysis at a general, non-sectoral
level in order to produce a sufficient number of replications in the search for a common
behavioral pattern [51,52]. Each new case provides us with an independent test of the
hypothesized relationships by comparing the expected pattern of behavior followed by the
dependent variables as a function of the independent variables with the actual pattern [53].

4. Results
When assessing the main influences of AI in relation to energy management, the
results show that AI has a high or very high positive influence on energy efficiency in 9 of
the 18 cases. This indicates that more than 5% efficiency improvements have been achieved
in half of the projects. Only in three projects was sufficient evidence found to prove any
influence of AI, specifically on energy efficiency. These are projects in three completely
different sectors (Supply and Distribution, Image Processing, and Design). In contrast, in
the other projects in these sectors, evidence was found to corroborate the influence of AI on
the energy management performance of companies.
In the case of emissions, in 15 of the 18 projects it is evident that there have been posi-
tive influences, allowing in 7 of these cases reductions of more than 5% (see Figure 2). As in
the case of energy efficiency, only three projects in three sectors (Maintenance and Security,
Supply and Distribution, and Image Processing) are considered to have no improvement
due to AI implementation. The rest of the projects in these sectors show positive results.
However, in the Health sector, two projects were found in which the influence of AI was
relatively low. Nevertheless, this contrasts with the third project analyzed in the sector,
where a very high influence was observed.
 Security, Supply and Distribution, and Image Processing) are considered to have no im-
provement due to AI implementation. The rest of the projects in these sectors show posi-
tive results. However, in the Health sector, two projects were found in which the influence
Energies 2024, 17, 649 of AI was relatively low. Nevertheless, this contrasts with the third project analyzed in the 9 of 15
sector, where a very high influence was observed.

Figure2. 2.
Figure Influence
Influence of AIofonAI on energy
energy efficiency
efficiency and emissions
and emissions generated
generated in in the cases conducted.
the cases conducted.

AtAtthethe sectoral
sectoral level,
level, AI has
AI has a medium,
a medium, high, high,
or very orhigh
veryinfluence
high influence
on energyoneffi-
energy efficiency
in 4 ofinthe
ciency 4 of6the
sectors analyzed
6 sectors analyzed(see (seeFigure
Figure 3).3). InInthe
theImage
Image Processing
Processing and Design sectors,
and Design
sectors,
there isthere
onlyis aonly a moderately
moderately low low positive influence.
positive influence. InIn other words,
other in the
words, insectors of
the sectors of Health,
Health, Energy, Maintenance and Security, and Supply and
Energy, Maintenance and Security, and Supply and Distribution it is necessary Distribution it is necessary to to consider
consider the influence of AI on energy efficiency. This could be because AI has proven to
the influence of AI on energy efficiency. This could be because AI has proven to be a useful
be a useful tool to improve the energy efficiency of clean energy production processes.
Energies 2024, 17, x FOR PEER REVIEW 10 of In
16
tool to improve the energy efficiency of clean energy production processes. In this sense, it
this sense, it is worth noting that the Energy and Maintenance and Security sectors are
is worth
those noting
in which AI hasthata the
highEnergy and influence
or very high Maintenance and efficiency.
on energy Security sectors are itthose in which
Specifically,
AI has
should
cases beainfluence
the high or
noted that very
of in high
AI2on
of the influence
4 companies
energy efficiency on isenergy
in the energy
reported efficiency.
to sector, theSpecifically,
be related to quality anditcost
improvements should be noted
ob-
that in
tained in2terms
improvement, of the 4not
companies
itofisefficiency
fully were
shared.in However,
thethan
higher energy it issector,
20%. stronglythe improvements
linked to reductions obtained
in lead- in terms of
Onmainly
times, the were
efficiency other
in the hand,
higher a high
Design, than influence
Supply20%. of AI in relation
and Distribution to emissions is evident in 2 of
sectors.
the 6 sectors: Energy and Design (see Figure 3). This could be attributed to the fact that AI
has proven to be a valuable tool for improving emissions monitoring and applying ML
models to measure key performance indicators. Significant reductions of more than 20%
of the emissions generated have been corroborated in 3 of the 4 projects analyzed in the
energy sector.
In relation to the third and fourth research questions, see Figure 4, the relationship
between the influence of AI on energy efficiency and emissions and strategic business var-
iables (cost, quality, lead-time, risks, and satisfaction) was contrasted. Although in some

Figure
Figure3.3.Influence of AI
Influence ofon
AIenergy efficiency
on energy and emissions
efficiency generated in
and emissions the sectorsin
generated conducted.
the sectors conducted.

Negative impact / Positive impact

-2 -1 0 1 2

Product cost
Energies 2024, 17, 649 10 of 15

On the other hand, a high influence of AI in relation to emissions is evident in 2 of

the 6 sectors: Energy and Design (see Figure 3). This could be attributed to the fact that
AI has proven to be a valuable tool for improving emissions monitoring and applying ML
models to measure key performance indicators. Significant reductions of more than 20%
of the emissions generated have been corroborated in 3 of the 4 projects analyzed in the
energy sector.
In relation to the third and fourth research questions, see Figure 4, the relationship
between the influence of AI on energy efficiency and emissions and strategic business
variables (cost, quality, lead-time, risks, and satisfaction) was contrasted. Although in
some cases the influence of AI on energy efficiency is reported to be related to quality and
cost improvement, it is not fully shared. However, it is strongly linked to reductions in
Figure 3. Influence
lead-times, of AI in
mainly on the
energy efficiency
Design, and emissions
Supply generated in sectors.
and Distribution the sectors conducted.

Negative impact / Positive impact

-2 -1 0 1 2

Product cost

Product quality

Lead time

Risk

Workers satisfaction

Managers satisfaction

Energy efficiency Emissions

Figure 4. Influence of energy efficiency and emissions through AI on some business strategic variables.
Figure
Note: 4.There
Influence
is noofclear
energy efficiency
evidence of and emissions through
the influence AI on on
of emissions some
thebusiness strategic
dependent vari- of product
variables
ables. Note: There is no clear evidence of the influence of emissions on the dependent variables of
cost and lead-time, but they do improve the product and the company’s image, although not across
product cost and lead-time, but they do improve the product and the company’s image, although
theacross
not board,thewhich
board,iswhich
why isthe netthe
why valuation takestakes
net valuation the central value
the central of of
value zero.
zero.

However,
However, thethe
linklink between
between this influence
this influence and satisfaction
and satisfaction is noteworthy.
is noteworthy. Some cus-Some cus-
tomersand
tomers and managers
managers are are satisfied,
satisfied, butisthis
but this not is not widespread.
widespread. EmployeesEmployees are generally
are generally
dissatisfied.
dissatisfied. They
They feelfeel
thatthat
this this influence
influence affectsaffects them negatively
them negatively because because
it creates it creates a sense
a sense
ofofinsecurity
insecurity associated
associated withwith changing
changing ways ways of working
of working to achieve
to achieve these efficiencies,
these efficiencies, a a
perception
perception that
thatsome
some of their work
of their is notisnecessary,
work and a fear
not necessary, and of possible
a fear job cuts. job cuts.
of possible
Except for the case of a small influence on product and company image, no evidence
has been found regarding the influence of AI on emissions and variations in quality, cost,
or lead-time. However, impacts on occupational risks are reported in more than half of
the cases. Generally, improvements are linked to a reduction of chemical and biological
occupational hazards. However, employees are not satisfied and consider that AI affects
them negatively. They do not feel safe because they lose some process control.

5. Discussion
The evidence of the influence of AI has been very varied, but, in general, this research
has empirically confirmed the contribution of AI in improving energy efficiency and
emissions generated, in line with Ghobakhloo et al. [42], Entezari et al. [54], or Khan
et al. [55]. The extent of its effects varies significantly depending on the application and
sector, as already noted by Chauhan et al. [56], which is a result of the wide variety of
current AI applications.
In the analysis of the cases, the improvement of these indicators in the application
of AI has not been a priority. The main objective of the companies is to improve their
economic indicators, leaving environmental and social indicators in the background. This is
Energies 2024, 17, 649 11 of 15

in line with the trend in business and academia, which has focused primarily on economic
competitiveness. The control of energy consumption and emissions and the use of ML
models are aspects that are considered a priority in the industrial sector [57,58]. This implies
a delay in the consideration of environmental and social aspects [38].
When assessing the main influences of AI in relation to energy management, the
results show that AI has a high or very high positive influence on energy efficiency. This is
consistent with findings in the literature [42].
In the Health, Energy, Maintenance and Security, and Supply and Distribution sectors,
it is necessary to consider the influence of AI on energy efficiency. This could be because AI
has proven to be a useful tool to improve the energy efficiency of clean energy production
processes. In this sense, it is worth noting that the Energy and Maintenance and Security
sectors are those in which AI has a high or very high influence on energy efficiency. In the
case of the Energy sector, there is evidence proven by several studies [59,60], but none have
been found for the Maintenance and Security sector, where energy efficiency or emissions
may not have been a central issue so far.
On the other hand, in some cases, the influence of AI on energy efficiency is reported
to be related to quality and cost improvement, but it is not fully shared, which contrasts
with Hasan and Triani [13]. Although the authors apply the framework, they develop it in
three different sectors. However, it does seem to be strongly linked to the reduction of lead
times, mainly in the design, supply, and distribution sectors.
On the human aspect, some customers and managers are satisfied, but this is not
widespread, as AI may intensify the issue of digital skill gaps [61]. In this respect,
Ghobakhloo et al. [62] talk about the benefits of management digitalization competency
as one of the factors influencing the implementation of AI. On the other hand, employees
are generally dissatisfied, which can make it difficult to implement AI [61]. Employees feel
that AI influence affects them negatively because it creates a sense of insecurity associated
with changing ways of working to achieve these efficiencies, a perception that some of their
work is not necessary, and a fear of possible job cuts. This confirms the findings of previous
studies such as those of Leesakul et al. [63], Malik et al. [64], or Palos- Sánchez et al. [65].
Regarding emissions, it is evident that there have been positive influences from AI. A
high influence of AI in relation to emissions is evident in the Design and Energy sectors,
especially in the last one. This could be attributed to the fact that AI has proven to be a
valuable tool for improving emissions monitoring and applying ML models to measure
key performance indicators [25], as in the case of smart energy meters. The data collected
by these devices can help accurately predict electricity and natural gas consumption to
better plan and manage the energy supply system and reduce associated emissions [59].
Concerning the Design sector, there is a lack of prior research either confirming or refuting
the impact of AI on emissions. It is perhaps because it lacks any a priori relevance, although
the sector does have a delayed impact with the generation of emissions in other phases of
the product cycle, and AI has the potential to reduce and even prevent them, as it does in
problem-solving tasks [59,66].
The literature highlights that innovation obtained from AI applications is much needed
in sectors with high levels of emissions, such as Supply and Distribution or Maintenance
and Safety, so that they can be energy efficient, smart, sustainable, and cost-effective [32].
However, except for the case of a small influence on product and company image, no
evidence has been found regarding the influence of AI on emissions and variations in
quality, cost, or lead-time. There is also no evidence in the literature to provide further
clarity on this aspect. Nevertheless, impacts on occupational risks are reported in more
than half of the cases. Generally, improvements are linked to a reduction of chemical and
biological occupational hazards, which has already been reported in recent studies [67].
However, employees are not satisfied and consider that AI affects them negatively. They
do not feel safe because they lose some process control. This novel finding perfectly
complements the profile that the worker adopts with respect to AI implementation.
Energies 2024, 17, 649 12 of 15

These results confirm the need raised in previous studies on the need to take measures
in the AI adoption process to minimize the frustration it generates among workers [59].
Employee empowerment is essential, as failure to address it properly can lead to unsuc-
cessful implementation and cause them to miss the opportunity to strategically align AI
with potential sustainability benefits [62]. By empowering employees, the positive implica-
tions of AI adoption in general and energy and emissions management in particular will
be maximized.

6. Conclusions
Companies must take advantage of the opportunities offered by technological devel-
opment, considering the threats and opportunities for SD generated by this technological
revolution. Despite AI being acknowledged for its impact on economic competitiveness, a
notable literature gap exists in understanding its implications for social and environmental
aspects within the industrial sector [38]. The primary objective of this study is to scrutinize
how the integration of AI shapes circularity in companies by enhancing energy efficiency,
reducing emissions, and influencing key strategic variables for economic performance.
Through a multiple case study approach, this research empirically affirms AIs positive
impact on energy efficiency and emissions reduction, aligning with Sustainable Develop-
ment Goals 7, 12, and 13. Case analyses reveal AIs influence varies across sectors and
applications, with a predominant focus on economic indicators in AI applications (rather
than environmental and social considerations), except in the energy sector, where im-
proved energy efficiency positively affects strategic variables such as product cost, quality,
and lead-time.
Concerning emissions, this study identifies the need for AI innovation in emission-
intensive sectors for enhanced energy efficiency, sustainability, and cost-effectiveness.
Although our findings do not establish a direct link between cost and emissions, they do
show a correlation with improved image and risk reduction, particularly in chemical and
biological aspects.
On a societal level, despite managers and customers expressing moderate satisfaction,
workers articulate significant dissatisfaction, attributing AI to increased job instability and
a feeling of insufficient control over processes. Consequently, actions should be taken in the
AI adoption process to minimize the frustration experienced by workers. This will enhance
the positive implications of AI adoption overall.
In light of the results, a collaborative effort by the public and private sectors, guided
by a roadmap, is recommended to formulate policies integrating AI for effective energy
and emissions management.
Turning to the limitations of this research, we underscore those derived from the case
study methodology. While offering a more detailed and realistic understanding than other
methodologies, there are limitations related to the generalisability of the results. These are
highly context-dependent, as stated by Yin [46]. Therefore, further research along these
lines would be of great interest, aiming to deepen and enhance the level of knowledge,
especially considering that companies can accrue higher experience levels over time and
work with more information on the obtained results.

Author Contributions: Conceptualization, N.U.-G. and G.A.-L.; methodology, B.L.-M. and G.A.-L.;
investigation and resources, N.U.-G., B.L.-M. and I.L.-I.; writing—original draft preparation, N.U.-G.,
G.A.-L. and B.L.-M.; writing—review and editing, N.U.-G., G.A.-L. and B.L.-M. All authors have read
and agreed to the published version of the manuscript.
Funding: The researchers are members of the research team IT169122 of the Basque University System.
Data Availability Statement: The data presented in this study are available upon request from
the corresponding author. The data are not publicly available because the original data may be
considered sensitive due to the increased risk of identification of sources, and therefore, the express
approval of the companies for the purposes requested will be required.
Energies 2024, 17, 649 13 of 15

Acknowledgments: The researchers are members of the research team IT169122 of the Basque
University System. This research work has had the technical and human support of the Environmental
Sustainability and Health Institute (ESHI) of the Technological University Dublin, the Circular
Economy University-Company Classroom (Faculty of Engineering Gipuzkoa, UPV/EHU, Provincial
Council of Gipuzkoa), the Centre de Documentation et de Recherches Europeennes (Universite de
Pau et des pays de l’Adour) and the Euskampus Foundation through the Missions 1.0 and Mission
2.0 programs.
Conflicts of Interest: The funders had no role in the design of the study, the collection, analysis, or
interpretation of data, the writing of the manuscript, or the decision to publish the results.

Abbreviations

AI Artificial intelligence
BD Big Data
CE Circular Economy
DTs Digital technologies
EEMs Energy Efficiency Measures
I4.0 Industry 4.0
I4.0Ts Industry 4.0 Technologiess
IoT Internet of Things
MICMAC Multiplication Appliquée aun Classement
ML Machine Learning
OEE Overall Equipment Effectiveness
PRISMA Preferred Reporting Items for Systematic Reviews and Meta-Analyses
RQ Research Question
SD Sustainable Development
SLR Systematic Literature Review

References
1. Kar, A.K.; Choudhary, S.K.; Singh, V.K. How can AI impact sustainability: A systematic literature review. J. Clean. Prod. 2022,
376, 134120. [CrossRef]
2. Boasson, E.L.; Tatham, M. Climate policy: From complexity to consensus? J. Eur. Public Policy 2023, 30, 401424. [CrossRef]
3. Campbell, N.; Ryan, L.; Rozite, V.; Lees, E.; Heffner, G. Capturing the Multiple Benefits of Energy Efficiency; ECEEE: Brussels launch of
the IEA Energy Efficiency Reports; IEA: Paris, France, 21 October 2014; Available online: https://www.eceee.org/static/media/
uploads/site-2/events/eceee_events/Brussels-launch-of-IEA-2014-report/nina-campbell-seminar-21October.pdf (accessed on
28 October 2014).
4. United Nations (UN). Climate Change. UN Climate Change Conference United Arab Emirates. Available online: https:
//unfccc.int/cop28 (accessed on 11 November 2023).
5. Chen, H.M.; Kuo, T.C.; Chen, J.L. Impacts on the ESG and financial performances of companies in the manufacturing industry
based on the climate change related risks. J. Clean. Prod. 2022, 380, 134951. [CrossRef]
6. Wang, X.; Zhong, M. Can digital economy reduce carbon emission intensity? Empirical evidence from China’s smart city pilot
policies. Environ. Sci. Pollut. Res. 2023, 30, 51749–51769. [CrossRef] [PubMed]
7. Epstein, M.J.; Roy, M.J. Sustainability in action: Identifying and measuring the key performance drivers. Long Range Plan. 2001,
34, 585–604. [CrossRef]
8. U.S. Departament of Energy. Better Buildings Initiative. Available online: https://betterbuildingssolutioncenter.energy.gov/
(accessed on 13 November 2023).
9. McDowall, W.; Geng, Y.; Huang, B.; Bartekova, E.; Bleischwitz, R.; Turkeli, S.; Kemp, R.; Domenech, T. Circular Economy Policies
in China and Europe. J. Ind. Ecol. 2017, 21, 651–661. [CrossRef]
10. International Energy Agency, Energy Policies of IEA Countries Japan 2016 Review. Available online: https://www.iea.org/
reports/energypoliciesofieacountriesjapan2016review (accessed on 27 December 2023).
11. OCDE, Clean Energy Finance and Investment Mobilisation. 2023. Available online: https://www.oecd.org/cefim/ (accessed on
27 December 2023).
12. Arana Landín, G.; Uriarte Gallastegi, N.; Landeta Manzano, B.; Laskurain Iturbe, I. The Contribution of Lean Management—I4.0Ts
to Improving Energy Efficiency. Energies 2023, 16, 2124. [CrossRef]
13. Hasan, A.M.; Trianni, A. Boosting the adoption of industrial energy efficiency measures through I4.0Ts to improve operational
performance. J. Clean. Prod. 2023, 425, 138597. [CrossRef]
14. Mariani, M.M.; Machado, I.; Nambisan, S. Types of innovation and AI: A systematic quantitative literature review and research
agenda. J. Bus. Res. 2023, 155, 113364. [CrossRef]
Energies 2024, 17, 649 14 of 15

15. Piscicelli, L. The sustainability impact of a digital CE. In Current Opinion in Environmental Sustainability; Dube, O.P., Galaz, V.,
Solecki, W., Eds.; Elsevier: Amsterdam, The Netherlands, 2023; Volume 61, p. 101251. [CrossRef]
16. Wang, Y.; Deng, X.; Zhang, H.; Liu, Y.; Yue, T.; Liu, G. Energy endowment, environmental regulation, and energy efficiency:
Evidence from China. In Technological Forecasting and Social Change; Cunningham, S., Hu, M.-C., Eds.; Elsevier: Amsterdam, The
Netherlands, 2022; Volume 177, p. 121528. [CrossRef]
17. Uriarte-Gallastegi, N.; Landeta-Manzano, B.; Arana-Landín, G.; Laskurain-Iturbe, I. How Do Technologies Based on Cyber–
Physical Systems Affect the Environmental Performance of Products? A Comparative Study of Manufacturers’ and Customers’
Perspectives. Sustainability 2022, 14, 13437. [CrossRef]
18. Bag, S.; Yadav, G.; Dhamija, P.; Kataria, K.K. Key resources for industry 4.0 adoption and its effect on sustainable production and
circular economy: An empirical study. J. Clean. Prod. 2021, 281, 125233. [CrossRef]
19. Uçar, E.; Le Dain, M.A.; Joly, I. Digital technologies in circular economy transition: Evidence from case studies. In Proceedings
of the 27th CIRP Life Cycle Engineering Conference (LCE2020) Advancing Life Cycle Engineering: From Technological Eco-
Efficiency to Technology that Supports a World that Meets the Development Goals and the Absolute Sustainability, Grenoble,
France, 13–15 May 2020; Volume 90, pp. 133–136. [CrossRef]
20. Chiarini, A. Industry 4.0 technologies in the manufacturing sector: Are we sure they are all relevant for environmental perfor-
mance? Bus. Strategy Environ. 2021, 30, 3194–3207. [CrossRef]
21. Uriarte Gallastegi, N.; Landeta Manzano, B.; Arana Landin, G.; Laskurain Iturbe, I. Influence of AI on Resource Consumption. In
Proceedings of the IFIP International Conference on Advances in Production Management Systems, Trondheim, Norway, 17–21
September 2023; Springer Nature: Cham, Switzerland, 2023; Volume 690, p. 662673.
22. Godinho Filho, M.; Monteiro, L.; de Oliveira Mota, R.; dos Santos Leite Gonella, J.; de Souza Campos, L.M. The Relationship
between CE, I4.0 and Supply Chain Performance: A Combined ISM/Fuzzy MICMAC Approach. Sustainability 2022, 14, 2772.
[CrossRef]
23. Laskurain Iturbe, I.; Arana Landín, G.; Landeta Manzano, B.; Uriarte Gallastegi, N. Exploring the influence of I4.0Ts on the CE.
J. Clean. Prod. 2021, 321, 128944. [CrossRef]
24. Spaltini, M.; Poletti, A.; Acerbi, F.; Taisch, M. A quantitative framework for I4.0 enabled CE. Procedia CIRP 2021, 98, 115120.
[CrossRef]
25. Liu, J.; Liu, L.; Qian, Y.; Song, S. The effect of AI on carbon intensity: Evidence from China’s industrial sector. Socio-Econ. Plan. Sci.
2022, 83, 101002. [CrossRef]
26. Waltersmann, L.; Kiemel, S.; Stuhlsatz, J.; Sauer, A.; Miehe, R. AI Applications for Increasing Resource Efficiency in Manufacturing
Companies—A Comprehensive Review. Sustainability 2021, 13, 6689. [CrossRef]
27. Li, J.; Herdem, M.S.; Nathwani, J.; Wen, J.Z. Methods and applications for AI, Big Data, IoT, and Blockchain in smart energy
management. Energy AI 2023, 11, 100208. [CrossRef]
28. Wiegand, T.; Wynn, M. Sustainability, the CE and digitalisation in the German textile and clothing industry. Sustainability 2023,
15, 9111. [CrossRef]
29. Liu, L.; Yang, K.; Fujii, H.; Liu, J. Artificial intelligence and energy intensity in China’s industrial sector: Effect and transmission
channel. Econ. Anal. Policy 2021, 70, 276–293. [CrossRef]
30. Rajput, S.; Singh, S.P. Industry 4.0–challenges to implement circular economy. Benchmark. Int. J. 2021, 28, 1717–1739. [CrossRef]
31. Vimal, K.E.K.; Churi, K.; Kandasamy, J. Analysing the drivers for adoption of I4.0Ts in a functional paper–cement–sugar circular
sharing network. Sustain. Prod. Consum. 2022, 31, 459477. [CrossRef]
32. Mondal, S.; Singh, S.; Gupta, H. Green entrepreneurship and digitalization enabling the CE through sustainable waste manage-
mentAn exploratory study of emerging economy. J. Clean. Prod. 2023, 422, 138433. [CrossRef]
33. World Economic Forum. Harnessing Technology for the Global Goals: A Framework for Government Action. 2021. Avail-
able online: https://www3.weforum.org/docs/WEF_Harnessing_Technology_for_the_Global_Goals_2021.pdf (accessed on
12 November 2023).
34. Zhao, P.; Gao, Y.; Sun, X. The impact of AI on pollution emission intensity—Evidence from China. In Environmental Science and
Pollution Research; Garrigues, P., Ed.; Springer: Bordeaux, France, 2023; Volume 30, pp. 91173–91188.
35. Chen, P.; Chu, Z.; Zhao, M. The Road to corporate sustainability: The importance of AI. Technol. Soc. 2023, 76, 102440. [CrossRef]
36. Ahmad, T.; Zhang, D.; Huang, C.; Zhang, H.; Dai, N.; Song, Y.; Chen, H. AI in sustainable energy industry: Status Quo, challenges
and opportunities. J. Clean. Prod. 2021, 289, 125834. [CrossRef]
37. Nehler, T.; Rasmussen, J. How do firms consider non-energy benefits? Empirical findings on energy-efficiency investments in
Swedish industry. J. Clean. Prod. 2016, 113, 472–482. [CrossRef]
38. Trianni, A.; Cagno, E.; Dolšak, J.; Hrovatin, N. Implementing energy efficiency measures: Do other production resources matter?
A broad study in Slovenian manufacturing small and medium-sized enterprises. J. Clean. Prod. 2021, 287, 125044. [CrossRef]
39. Katic, M.; Trianni, A. Energy efficiency measures and production resources: Towards an integrative classification framework
for decision makers. In Proceedings of the 2020 IEEE International Conference on Industrial Engineering and Engineering
Management (IEEM), Singapore, 18–21 December 2023.
40. Nota, G.; Nota, F.D.; Peluso, D.; Toro Lazo, A. Energy efficiency in Industry 4.0: The case of batch production processes.
Sustainability 2020, 12, 6631. [CrossRef]
Energies 2024, 17, 649 15 of 15

41. Medojevic, M.; Medic, N.; Marjanovic, U.; Lalic, B.; Majstorovic, V. Exploring the Impact of I4.0 Concepts on Energy and Environ-
mental Management Systems: Evidence from Serbian Manufacturing Companies. In Advances in Production Management Systems.
Towards Smart Production Management Systems, Proceedings of the APMS 2019. IFIP Advances in Information and Communication
Technology, Austin, TX, USA, 1–5 September 2019; Ameri, F., Stecke, K., von Cieminski, G., Kiritsis, D., Eds.; Springer: Cham,
Switzerland, 2019; Volume 567, p. 567. [CrossRef]
42. Ghobakhloo, M.; Fathi, M. Industry 4.0 and opportunities for energy sustainability. J. Clean. Prod. 2021, 295, 126427. [CrossRef]
43. Gupta, B.B.; Gaurav, A.; Panigrahi, P.K.; Arya, V. Analysis of AI based technologies and approaches on sustainable entrepreneur-
ship. Technol. Forecast. Soc. Chang. 2023, 186, 122152. [CrossRef]
44. Zhou, Y.; Xia, Q.; Zhang, Z.; Quan, M.; Li, H. Artificial intelligence and machine learning for the green development of agriculture
in the emerging manufacturing industry in the IoT platform. Acta Agric. Scand. Sect. B Soil Plant Sci. 2022, 72, 284–299. [CrossRef]
45. Yin, Q.; Wang, D.; Wang, Y. Serial Mediation Model Linking Returnee Entrepreneurship Education and Green Returnee
Entrepreneurial Behavior: An Analysis of Environmental Improvement. Sustainability 2023, 15, 14044. [CrossRef]
46. Yin, R.K. Case Study Research and Applications: Design and Methods; Sage: Thousand Oaks, CA, USA, 2018.
47. Eisenhardt, K.M. Building Theories from Case Study Research. Acad. Manag. Rev. 1989, 14, 532–550. [CrossRef]
48. Patton, M.Q. Qualitative Research & Evaluation Methods: Integrating Theory and Practice; Sage Publications: London, UK, 2014.
49. Gummesson, E. Qualitative research in management: Addressing complexity, context and persona. Manag. Decis. 2006, 44, 167179.
[CrossRef]
50. Thurmond, V.A. The point of triangulation. J. Nurs. Scholarsh. 2001, 33, 253258. [CrossRef] [PubMed]
51. Miles, M.B.; Huberman, A.M.; Saldana, J. Qualitative Data Analysis: A Methods Sourcebook, 3rd ed.; Sage: Thousand Oaks, CA,
USA, 2014.
52. Rialp, A.; Rialp, J.; Urbano, D.; Vaillant, Y. The bornglobal phenomenon: A comparative case study research. J. Int. Entrep. 2005,
3, 133171. [CrossRef]
53. McCutcheon, D.M.; Meredith, J.R. Conducting case study research in operations management. J. Oper. Manag. 1993, 11, 239–256.
[CrossRef]
54. Entezari, A.; Aslani, A.; Zahedi, R.; Noorollahi, Y. AI and machine learning in energy systems: A bibliographic perspective.
Energy Strategy Rev. 2023, 45, 101017. [CrossRef]
55. Khan, I.S.; Ahmad, M.O.; Majava, J. I4.0 and sustainable development: A systematic mapping of triple bottom line, CE and
Sustainable Business Models perspectives. J. Clean. Prod. 2021, 297, 126655. [CrossRef]
56. Chauhan, C.; Parida, V.; Dhir, A. Linking CECE and digitalisation technologies: A systematic literature review of past achieve-
ments and future promises. Technol. Forecast. Soc. Chang. 2022, 177, 121508. [CrossRef]
57. Andeobu, L.; Wibowo, S.; Grandhi, S. AI applications for sustainable solid waste management practices in Australia: A systematic
review. Sci. Total Environ. 2022, 834, 155389. [CrossRef]
58. Patterson, D.; Gonzalez, J.; Le, Q.; Liang, C.; Munguia, L.M.; Rothchild, D.; So, D.; Texier, M.; Dean, J. Carbon emissions and large
neural network training. arXiv 2021, arXiv:2104.10350.
59. Smajla, I.; Sedlar, D.K.; Vulin, D.; Jukić, L. Influence of smart meters on the accuracy of methods for forecasting natural gas
consumption. Energy Rep. 2021, 7, 8287–8297. [CrossRef]
60. Singh, R.; Akram, S.V.; Gehlot, A.; Buddhi, D.; Priyadarshi, N.; Twala, B. Energy System 4.0: Digitalization of the energy sector
with inclination towards sustainability. Sensors 2022, 22, 6619. [CrossRef]
61. Ghosh, S. The Future Is Both Automated and Intelligent. Forbes Technology Council. Available online: https://www.forbes.com/
sites/forbestechcouncil/2021/04/08/the-future-is-both-automated-and-intelligent/?sh=45cdfcfa5664 (accessed on 16 July 2022).
62. Ghobakhloo, M.; Asadi, S.; Iranmanesh, M.; Foroughi, B.; Mubarak, M.F.; Yadegaridehkordi, E. Intelligent automation implemen-
tation and corporate sustainability performance: The enabling role of corporate social responsibility strategy. Technol. Soc. 2023,
74, 102301. [CrossRef]
63. Leesakul, N.; Oostveen, A.M.; Eimontaite, I.; Wilson, M.L.; Hyde, R. Workplace 4.0: Exploring the implications of technology
adoption in digital manufacturing on a sustainable workforce. Sustainability 2022, 14, 3311. [CrossRef]
64. Malik, N.; Tripathi, S.N.; Kar, A.K.; Gupta, S. Impact of artificial intelligence on employees working in industry 4.0 led
organizations. Int. J. Manpow. 2022, 43, 334–354. [CrossRef]
65. Palos-Sánchez, P.R.; Baena-Luna, P.; Badicu, A.; Infante-Moro, J.C. Artificial intelligence and human resources management: A
bibliometric analysis. Appl. Artif. Intell. 2022, 36, 2145631. [CrossRef]
66. Verganti, R.; Vendraminelli, L.; Iansiti, M. Innovation and design in the age of artificial intelligence. J. Prod. Innov. Manag. 2020,
37, 212–227. [CrossRef]
67. Arana Landín, G.; Laskurain Iturbe, I.; Iturrate, M.; Landeta Manzano, B. Assessing the influence of I4.0 technologies on
occupational health and safety. Heliyon 2023, 9, 13720. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.