South African Journal of Industrial Engineering November 2024 Vol 34(3), pp 195-207

THE IMPLICATIONS OF INTEGRATING ARTIFICIAL INTELLIGENCE INTO DATA-DRIVEN

DECISION-MAKING

J. Sutherns1* & G.B. Fanta1

ARTICLE INFO ABSTRACT

Article details Integrating artificial intelligence (AI) into data-driven decision-making

Presented at the 34th annual conference
of the Southern African Institute for offers advantages like increased performance, reduced costs and
Industrial Engineering, held from 14 to improved organisational efficiency; however, there are associated risks.
16 October 2024 in Vanderbijlpark,
South Africa
The study employs a PRISMA protocol to systematically review academic
articles from Scopus, ScienceDirect, and Web of Science databases to
Available online 29 Nov 2024 determine whether the risks AI pose are worth the rewards they offer.
Literature trends reveal a growing interest in AI-driven decision-making,
Contact details with significant research gaps in African contexts. The study indicates
∗ Corresponding author
jesssutherns@[Link]
that AI is highly utilized for decision-making to foster competitiveness in
manufacturing, finance, healthcare, education, and transport.
Identified risks include bias, discrimination, privacy issues, and
Author affiliations
1 Department of Engineering and cybersecurity threats. It is highlighted that businesses need to address
Technology Management, concerns about privacy, fairness, and transparency. Policymakers must
University of Pretoria, South
Africa
develop ethical and legal standards besides regular monitoring and
auditing of AI uses to mitigate risks.
ORCID® identifiers
J. Sutherns
OPSOMMING
[Link]

G.B. Fanta
[Link] Die integrasie van kunsmatige intelligensie (KI) in data-gedrewe
besluitneming bied voordele soos verhoogde werkverrigting, verlaagde
koste en verbeterde organisatoriese doeltreffendheid; daar is egter
DOI
[Link]
gepaardgaande risiko's. Die studie gebruik 'n PRISMA-protokol
stelselmatig om akademiese artikels van die Scopus-, ScienceDirect- en
Web of Science-databasisse te hersien om te bepaal of die risiko's wat KI
inhou die belonings werd is wat dit bied. Die literatuurtendense toon 'n
groeiende belangstelling in KI-gedrewe besluitneming, maar met
aansienlike navorsingsgapings in Afrika-kontekste. Die studie dui aan dat
KI baie gebruik word vir besluitneming om mededingendheid in
vervaardiging, finansies, gesondheidsorg, onderwys en vervoer te
bevorder. Geïdentifiseerde risiko's sluit in vooroordeel, diskriminasie,
privaatheidskwessies en kuberveiligheidsbedreigings. Die studie
beklemtoon dat besighede bekommernisse oor privaatheid,
regverdigheid en deursigtigheid moet aanspreek. Beleidmakers moet
etiese en wetlike standaarde ontwikkel, benewens die gereelde
monitering en ouditering van KI-gebruike om risiko's te versag.

195
1. INTRODUCTION

In today’s era, marked by Industry 4.0 and characterised by a combination of digital technologies with
physical systems, the basis of and driving force behind growth and change is incorporating artificial
intelligence into data-driven decision-making [1]. Indeed, significant progress in cyber-physical systems,
the Internet of Things, and cloud computing are at the root of fundamentally changing the dynamics of how
businesses operate, how governments function, and how humans interact with technology [2], [3]. At its
core, AI is a revolutionary machine – a machine that can quickly process vast quantities of data to unlock
critical insights and drive highly informed decision-making processes with previously unseen efficiency and
accuracy [4].

For example, predictive maintenance algorithms in manufacturing show the efficiency of AI in enhancing
operational efficiency and improving manufacturing processes. Besides, predictive analytics performed
with AI is capable of helping companies to predict and fix failures in their equipment, thus saving on
downtime, resource allocations, and maintenance costs [5], [6]. In the health sector, AI-driven diagnostic
tools are changing the dimension of medical imaging, since they enable timely disease detection, quality
patient outcomes, and organised treatment procedures [7], [8].

AI algorithms also help to analyse critically the financial sector markets and forecast consumer behaviour
to let financial institutions make the most appropriate investment decisions and personalise their services
to clients [9], [10]. Another advantage of AI solutions is that they increase the delivery of public services
in many domains, such as traffic management and disaster response [11], [12]. Last, apart from industry,
AI has affected society in several ways, such as optimised traffic management, improved environmental
monitoring, and improved educational conditions [13], [14], [15].

Despite AI’s advantages, several ethical issues arise from its acceptance in society, including the biases it
causes and the transparency deficit it carries [16], [17]. For instance, a lack of clarity might arise when
making data-driven decisions, whereby there is a need to question who to hold responsible for any
untrustworthiness [8]. Other ethical problems and biases may also come through in decision-making, leading
to discriminatory decisions [17]. Last, broader risks threaten society, such as AI-infected viruses and a lack
of human autonomy. As a result, it is essential to implement strategies for using AI [18], [19].

Because of the risks that are posed by using AI for data-driven decision-making, an essential question is:
“Is the use of AI for competitive advantage worth the associated risks it poses?”

To answer this question, this study intends to analyse the implication of integrating AI into data-driven
decision-making with the help of the following research questions:
RQ1: What are the benefits of using AI in data-driven decision-making?
RQ2: What are the risks of using AI in data-driven decision-making?
RQ3: Which sectors mainly use AI for decision-making?
RQ4: What strategies do organisations use to address issues when using AI for data-driven decision-making?

Thus this paper should give clear insight into AI’s multifaceted economic, environmental, and social impacts
on decision-making and, ultimately, offer strategies for future research endeavours.

2. METHODOLOGY

2.1. Introduction

This section presents an amended version of the methodologies used by [20] and [21] to analyse the
literature and to extract its main findings. This methodology first outlines the strategy to narrow the
literature search, then screens the literature and extracts and analyses the relevant data. The methodology
is carried out this way to ensure clarity and comprehensiveness.

196
2.2. Search Strategy

As this research paper aimed to determine the implications of integrating AI into decision-making, an in-
depth examination of ‘artificial intelligence for data-driven decision-making to gain a competitive
advantage’ was initially conducted to lay the foundation for formulating the literature review questions.

Once the review questions had been formulated, they were broken down into distinct keywords to create
the search terms illustrated in Table 1. This involved using Boolean operators, since these would ensure
that the retrieved literature consisted only of articles containing one or both search terms in order to
explore the relevant sources comprehensively [22].
Table 1: Search terms used to obtain literature

Search terms
1 Artificial intelligence) AND (data-driven decision-making)
2 (Artificial intelligence) AND (decision-making) AND (risks)
3 (Artificial intelligence) AND (decision making) AND (advantage)
4 (Industry 4.0) AND (data-driven decision making)

The search terms were then run through three search engines – Scopus, ScienceDirect, and Web of Science
– since those databases are compatible with Boolean operators and have been considered to be of a high
standard [23]. Apart from the abovementioned platforms, the search terms were run through Google
Scholar to broaden the search and to ensure that no scholarly works were left unexamined.

2.3. Study selection

After the literature retrieval, a thorough screening and selection process followed the steps according to
the preferred reporting of items for systematic reviews and meta-analyses (PRISMA) statement, as shown
in Figure 1. According to the statements proposed by [24] and [25], the PRISMA statement outlines the steps
used to limit the initial set of studies into a final set and, in so doing, to make the review sound by increasing
its methodological quality.

As described by the PRISMA statement, the retrieved literature was subjected to strict screening against
the pre-defined inclusion and exclusion criteria shown in Table 2, during which we examined the relevance
of each study to the research objective. The specific inclusion and exclusion criteria were chosen to ensure
that only recent articles of a high academic quality were included, as the research questions required
credible information on the latest implications of AI. According to [26], the PRISMA screening process is
used, as it includes only the most relevant studies for analysis so that the literature that is obtained is as
fine-tuned as possible.
Table 2: Inclusion and exclusion criteria used to narrow down literature

Inclusion criteria Exclusion criteria

Studies published between 2018 and the present. Studies published as conference proceedings.
Studies are available in full text in English. Studies printed in hard copy.
Studies published in journals.
Studies are peer reviewed.

197
 Identification of studies via databases and registers

Records removed before

screening:
Records identified from*:
Identification

Duplicate records removed

Databases (n = 9668 )
(n = 767)

Records removed for other

reasons (n = 172)

Records screened Records excluded**

(n = 8729) (n = 6447)

Reports sought for retrieval Reports not retrieved

(n = 2282)
Screening

(n = 857)

Reports assessed for eligibility Reports excluded: 1393

(n = 1425) Incorrect content (n = 358)

Replicated theme (n = 1035)

Studies included in review

Included

(n = 32)

Figure 1: PRISMA statement used to narrow down literature

2.4. Data collection

Once the literature had been refined using the PRISMA statement, the qualitative data from the selected
studies was extracted and presented using an author matrix. This was used to compare the primary findings
and limitations across the reviewed literature to ensure thoroughness and accuracy.

Following the data extraction, a quality assessment, as summarised in Table 3, was carried out to ascertain
the reliability and relevance of the literature. This quality assessment used a set of criteria from [27] and
[28] to assess the quality of every study, including having a well-defined research aim, a suitable
methodology, and reporting its limitations.

The quality assessment also examined each study’s strengths and limitations to evaluate the credibility of
the extracted data and thereby enhance the overall reliability of the findings.

198
 Table 3: Quality assessment criteria used to measure the reliability of the literature

Question
Does the study have a well-defined research aim?
Does the study have appropriate research questions?
Does the study have a suitable methodology?
Does the study have an appropriate data collection process?
Does the collection of data in the study happen in an unbiased way?
Does the study have a suitable data analysis process?
Does the study interpret its results efficiently?
Does the study report its limitations or validity threats?
Does the collected data in the study support its results and conclusions?

2.5. Data analysis

After the data had been analysed, the author matrix was converted into a concept matrix by categorising
the data into key concepts and themes. According to [29], a concept matrix allows an extensive
comparative analysis of reviewed literature to be carried out by examining its similarities, differences, and
emerging patterns. Furthermore, the matrix visualises the relationships between the main themes,
providing essential implications for the findings.

3. LITERATURE REVIEW RESULTS

This section presents the results of the literature review, from which the literature trends and key findings
concerning the integration of AI into data-driven decision-making are discussed.

3.1. Literature trends

In order to analyse the chosen articles’ publication patterns and to understand our findings more clearly,
we examined several trends. Initially we analysed the number of studies published per year, as seen in
Figure 2, which indicates a peak in publications in 2023, followed by 2021, 2022, and 2024, proving the
articles to be relevant. Second, we analysed the publication country of the studies, as seen in Figure 3,
with some publications occurring in more than one country. When examining Figure 3, we observed that
most of the publications were in the United States and the United Kingdom, but that none were in African
countries. This trend indicates the need for more research into integrating AI into decision-making in
African countries. Last, we examined the different research methodologies employed by the studies, as
depicted in Figure 4, which showed the literature review to be the most common methodology used. The
evidence of the literature review indicates that there is an ongoing contribution to the development of
knowledge about the integration of AI into data-driven decision-making.

Figure 2: Document publications per year

199
 Figure 3: Document publications per country

Figure 4: Research methodologies used

3.2. Main findings

The primary findings obtained during the systematic examination of the literature are discussed below. We
begin with an exploration of the benefits and risks of integrating AI into decision-making, followed by an
identification of the sectors using AI in decision-making, and ending with the various methods that
companies use to mitigate the risks posed by AI in decision-making.

3.2.1. Advantages of using artificial intelligence for data-driven decision-making

AI for decision-making offers organisations many benefits, as depicted in Table 4, and gives companies a
competitive advantage by improving their efficiency and effectiveness.

200
 Table 4: Summary of the benefits that artificial intelligence has for decision-making

Benefits Description Sources

Improves business outcomes Predictive maintenance, data science, and data [1], [3], [4], [8],
analytics lead to improved processes and reduced [30], [31], [32]
costs, which result in better results in all sectors.
Reduces costs Predictive maintenance, data science, and AI lead [1], [3], [4], [32]
to reduced expenses owing to streamlined
operations in all sectors.
Improves patient care Data science and AI applications improve decision- [4], [7]
making, which in turn is responsible for better
health care outcomes and regulation processes.
Enhances operation Advanced technologies optimise processes, promote [1], [8], [13],
efficiency sustainability, and improve environmental [30], [31], [33]
performance in various sectors.
Improves customer AI-driven personalisation and quality product [5], [33]
satisfaction improvement enhance user satisfaction across all
platforms and manufacturing sectors.
Enhances adaptability and AI and Industry 4.0 technologies lead to adaptability [5], [9], [12],
competitiveness and competitiveness through data-driven decision- [34], [35]
making and innovations.
Improves time of assistance Advanced technologies such as AI and big data [12]
to victims analytics lead to enhanced response strategies in
humanitarian operations.
Improved and faster AI, machine learning, and data analytics enable [1], [4], [5], [8],
decision-making quick and informed decision-making in many [10], [13], [35],
industries.
Increased growth Advanced technologies in the manufacturing, [3], [8], [13],
finance, and research sector increase growth owing [31], [33], [35]
to improved decision-making.
Improves sustainability AI and big data contribute to sustainability in [3], [8], [14],
manufacturing, urban development, and [30], [36], [37]
environmental governance.
Improves policies and public Technologies and data analytics enhance evidence- [11], [17], [36],
values based decision-making, refine policies, and [38],
strengthen societal values.

Strengthens governance Advanced technologies help make better and more [7], [9], [11],
transparent governance decisions. [15], [18], [33],
[36], [38]
Advanced performance Data science, artificial intelligence, and Industry 4.0 [4], [6], [8],
methodologies enhance decision-making and [11], [35], [42]
efficiency in the sectors.
Increased monitoring and Data-driven approaches and artificial intelligence [8], [14], [30],
managing environmental enhance the monitoring and management of [37]
challenges environmental challenges.

3.2.2. Risks of using artificial intelligence for data-driven decision-making

Although the areas of application of artificial intelligence are numerous, they have risks, as illustrated in
Table 5. As these consequences are severe, companies must ensure that they are actively managing such
risks to realise the full benefits of the technologies.

201
 Table 5: Summary of the risks that artificial intelligence brings to decision-making

Risks Description Sources

Bias and The biases built into AI systems lead to unfair outcomes in [2], [16], [17],
discrimination the decisions made in different sectors. [38], [39], [40]
Privacy concerns The application of artificial intelligence and data-driven [16], [39], [40]
decision-making in almost all sectors leads to privacy
issues, affecting personal data security and democratic
principles.
Environmental A big challenge exists in applying advanced technologies [8], [37], [41]
governance such as artificial intelligence and big data when addressing
environmental governance.
AI-infected viruses The increased existence of viruses infected with artificial [19]
intelligence leads to destructive risks for computer systems
that may negatively impact members of society.
Ethical concerns The presence of biases in AI causes ethical issues such as [16], [39], [40]
privacy concerns and threats to democratic values.
Technological The use of AI in decision-making raises concerns about who [5]
control is in control and what impact the technology has on
individual autonomy.

3.2.3. Sectors using artificial intelligence for decision-making

As detailed in Table 4, various companies in many sectors use AI and its associated benefits in making data-
driven decisions. Given the wide adoption of AI in multiple domains, Figure 5 depicts the frequency of the
various sectors mentioned in the literature, with some literature studies affecting more than one sector.
As seen in Figure 5, the sectors most dependent on AI are the manufacturing, finance, and healthcare
industries, proving the importance of technology globally.
Table 6: Sectors using artificial intelligence for data-driven decision-making

Sector How artificial intelligence assists

Manufacturing It makes manufacturing processes more effective, predicts maintenance needs,
ensures product integrity, and suggests environmentally sustainable practices.
Climate change It helps to mitigate challenges to achieving carbon neutrality by identifying
adaptation strategies, seeking improvements, and using holistic approaches.
Streaming It personalises the content and services to increase user satisfaction and
platforms engagement.
Finance It analyses risks, detects fraud, and enhances processes by automating them.
Healthcare It helps efficiently to dispense medicine, diagnose diseases, and personalise
therapy.
Supply chain It helps to increase supply chain productivity and sustainability.
management
Governance It suggests the moral correctness of decisions.
Disaster relief It helps quickly to assess, coordinate, and distribute resources for providing
relief.
Education It supports customised learning, automates administrative tasks, and provides
access to data analysis in education.
Innovation It improves decision-making processes, encourages knowledge creation, and
management fosters innovation management by experimenting in unknown environments.
Transportation It enhances autonomous vehicles’ safety, environmental impact, and societal
acceptance.
Higher education It helps in resource allocation, performance evaluation, and research
institutions (HEIs) enhancement.
Materials science It accelerates research, helps in error detection, and improves structural
analysis.
Cybersecurity It helps to detect and prevent threats.

202
 Figure 5: Frequency of sector references in analysed literature

3.2.4. Methods employed to address risks when incorporating AI-driven decision-making

Owing to the many risks involved in using AI for decision-making, firms must devise strategies to counteract
these possible effects. Some approaches that companies have already taken to minimise AI’s impacts are
summarised in Table 7.
Table 7: Strategies used to account for risks when implementing AI-based decision-making

Strategy Source
Implementing ethical and legal principles [7]
Addressing biases in AI systems [7]
Implementing training and awareness [33]
Performing regular auditing and monitoring [15]
prioritising fairness, accountability, and transparency in machine learning [31]
Investing in data quality and security [36]
Ensuring transparency and accountability [15]
Implementing human oversight [33]

4. DISCUSSION

4.1. Principal findings

This section discusses the implication of integrating AI into data-driven decision-making by analysing the
literature trends described in section 3.1 and examining the research questions in section 3.2.

4.1.1. Literature trend analysis

As seen in section 3.1, literature trends can provide essential insights into the findings from the obtained
literature and give helpful information for further research endeavours. Initially, as seen in Figure 2, the
publication years of the literature are highly relevant, proving the findings to be reliable and credible, and
reflecting a rising research interest in AI-driven decision-making in various sectors. In addition, as seen in
Figure 3, the lack of articles published in African countries indicates an extreme need for more research
on integrating AI into data-driven decision-making in this region.

203
4.1.2. Discussion of research questions

RQ1: What are the benefits of using AI in data-driven decision-making?

AI has many advantages for various fields and sectors, as shown in Table 4. In maintenance planning and
risk mitigation, AI uses predictive maintenance to optimise processes and to reduce costs in order to
enhance business outcomes [1], [4], [8], [30]. In addition, in healthcare, AI improves patient care by
improving diagnostic accuracy and operational efficiency [4], [7]; and in humanitarian efforts, it refines
response strategies to ensure the rapid delivery of aid to disaster victims [12]. AI, too, enhances customer
satisfaction by providing personalised services [5], [33], and it fosters competitiveness by using data-driven
insights [5], [9], [34]. Moreover, AI encourages growth and sustainability and is pivotal in forming
government policies [11], [36], [38] and in environmental monitoring. Hence, as a result, the technology
can provide a more sustainable and efficient future across various sectors [8], [30].

RQ2: What are the risks of using AI in data-driven decision-making?

AI-driven decision-making processes have several risks for various sectors, as shown in Table 5. The
outstanding issue with the technology is societal discrimination, because biased algorithms might lead to
unfair outcomes owing to race or gender issues [16], [38], [39]. In using AI in health and finance, there is a
risk of privacy being violated, causing ethical and autonomy problems [39], [40]. In environmental
governance, AI leads to complexities in data and results in security problems that affect sustainability
efforts [8], [41]. In addition, systems infected by ‘AI viruses’ are the most significant cybersecurity threat
and threaten human and infrastructural safety [19]. Finally, human autonomy is substantially threatened
by AI through its influence on consumer freedom [5]. Therefore, the risks that AI poses must be vigilance
maintained to ensure that the technology will be safe to use.

RQ3: Which sectors mainly use AI for decision-making?

As AI offers many advantages, many sectors use the technology, as depicted in Table 6. The most prominent
sector using technology for data-driven decision-making is the manufacturing sector, as seen in Figure 5; it
uses AI to enhance the efficiency of manufacturing processes, predict the need for maintenance, ensure
the quality of the products they produce, and encourage environmentally friendly practices [1], [8], [33],
[35]. The second most prominent sectors using AI are the finance and healthcare sectors, which use AI in
assessing risks, detecting fraud, improving medical diagnoses, and personalising treatment [3], [4], [7],
[35]. The remaining sectors – supply chain management, governance, disaster relief, education, innovation
management, transportation, higher education institutions, materials science, and cybersecurity – rely on
AI for decision-making to address specific problems and to improve operational efficiency, as noted in Table
6.

RQ4: What strategies do organisations use to address issues when using AI for data-driven decision-
making?

Since incorporating AI into data-driven decision-making risks having serious consequences, organisations
must strategically manage these ethical concerns, biases, and privacy issues to enhance their competitive
advantage, as detailed in Table 7. Some of the strategies are enforcing moral and legal standards [7],
conducting training sessions to raise awareness among staff [33], performing regular audits and monitoring
to ensure compliance, emphasising accountability in machine-learning processes [31], investing in data
security to prevent privacy infringements [36], and ensuring human oversight in maintaining control [33].
These strategies empower an organisation to handle effectively the risks that AI poses so that it gains the
benefits of the technology.

4.2. Implications

The implications from the key findings in section 4.1 highlight that AI is of great value to different sectors,
since it provides them with an opportunity to achieve a competitive edge. The application of AI in decision-
making processes in the manufacturing, finance, healthcare, and education sectors allows them to increase
their efficiency and effectiveness [1], [4], [7], [15], [39], [40]. However, at the same time, the risks linked
with AI call for mitigative strategies to ensure the responsible deployment of the technology. Thus there is
a call to business leaders to harness the full potential of AI while addressing the risks related to privacy,
fairness, and transparency [31]. Policymakers are urged to integrate AI into government frameworks to
204
provide social benefits, ensuring that ethical considerations are taken care of [7]. Therefore, although AI
has many transformative benefits, its risks must be researched through proactive strategies to ensure that
it is used sustainably and responsibly in different sectors.

4.3. Strengths and limitations

The structured PRISMA protocol followed in this study offers reliability and methodological clarity. In
addition, the study considers recently published and peer-reviewed articles (Figure 3) to capture the latest
developments in the field, strengthening the relevance of the findings. Nevertheless, despite the thorough
review process followed to ensure the quality of the identified articles, the study’s results rely on published
articles, and as a result, publication bias remains a limitation.

5. CONCLUSION

The integration of AI into data-driven decision-making in the manufacturing, finance, healthcare,

education, and transport sectors has a number of advantages and risks. AI-driven decision-making can
improve the performance, progress, and sustainability of various industries. However, its associated risks
must be addressed with strategic measures to harness responsibly the technology’s full potential and to
gain a competitive advantage in the data-driven landscape.

AI contributes to business growth and sustainability by smoothing out operations, increasing efficiency, and
fostering innovation. However, the risks related to bias, discrimination, privacy concerns, and AI-driven
security threats require attention. Organisations could implement measures to respond to these challenges
by employing ethical and legalistic norms, bias mitigation methods, and periodic monitoring and audit
practices. These measures would provide accountability and transparency by reducing the risks and
maximising the benefits of the AI-enabled decision-making process.

6. RECOMMENDATIONS

In future research efforts, researchers should study the effect of AI on additive manufacturing, as there is
an increased research interest in AI-enabled decision-making in manufacturing industries. As this study has
also recognised the need to identify articles from the African region, further studies may also focus on AI’s
effect on decision-making in the area. In addition, one way to demonstrate the impact of AI on decision-
making in various sectors would be by comparing the outcomes of decisions using AI with those made
without AI.

REFERENCES

[1] A. Bousdekis, K. Lepenioti, and G. Mentzas, “A review of data-driven decision-making methods for
Industry 4.0 maintenance applications,” Electronics, vol. 10, no. 7, 828, 2021.
[2] Y. Duan, J. S. Edwards, and Y. K. Dwivedi, “Artificial intelligence for decision making in the era of
big data: Evolution, challenges and research agenda,” International Journal of Information
Management, vol. 48, pp. 63-71, 2019.
[3] G. Soni, S. Kumar, R. V. Mahto, S. K. Mangla, M. L. Mittal, and W. M. Lim, “A decision-making
framework for Industry 4.0 technology implementation: The case of FinTech and sustainable supply
chain finance for SMEs,” Technological Forecasting and Social Change, vol. 180, 121686, 2022.
[4] Sarker, I. H. “Data science and analytics: An overview from data-driven smart computing, decision-
making and applications perspective,” SN Computer Science, vol. 2, no. 5, 377, 2021.
[5] A. R. Gonçalves, D. C. Pinto, S. Shuqair, M. Dalmoro, and A. S. Mattila, “Artificial intelligence vs.
autonomous decision-making in streaming platforms: A mixed-method approach,” International
Journal of Information Management, vol. 76, 102748, 2024.
[6] E. Mahmoodi, M. Fathi, M. Tavana, M. Ghobakhloo, and A. H. C. Ng, “Data-driven simulation-based
decision support system for resource allocation in Industry 4.0 and smart manufacturing,” Journal
of Manufacturing Systems, vol. 72, pp. 287-307, 2024.
[7] R. Jago et al., “Use of artificial intelligence in regulatory decision-making,” Journal of Nursing
Regulation, vol. 12, no. 3, pp. 11-19, 2021.
[8] C. Li, Y. Chen, and Y. Shang, “A review of industrial big data for decision making in intelligent
manufacturing,” Engineering Science and Technology, An International Journal, vol. 29, 101021,
2022.

205
[9] D. Korayim, V. Chotia, G. Jain, S. Hassan, and F. Paolone, “How big data analytics can create
competitive advantage in high-stake decision forecasting? The mediating role of organizational
innovation,” Technological Forecasting and Social Change, vol. 199, 123040, 2024.
[10] M. Stone et al. “Artificial intelligence (AI) in strategic marketing decision-making: A research
agenda,” The Bottom Line, vol. 33, no. 2, pp. 183-200, 2020.
[11] V. Charles, N. P. Rana, and L. Carter, “Artificial intelligence for data-driven decision-making and
governance in public affairs,” Government Information Quarterly, vol. 39, no. 4, 101742, 2022.
[12] R. Dubey, D. J. Bryde, Y. K. Dwivedi, G. Graham, and C. Foropon, “Impact of artificial intelligence-
driven big data analytics culture on agility and resilience in humanitarian supply chain: A practice-
based view,” International Journal of Production Economics, vol. 250, 108618, 2022.
[13] S. Gaftandzhieva, S. Hussain, S. Hilcenko, R. Doneva, and K. Boykova, “Data-driven decision making
in higher education institutions: State-of-play,” International Journal of Advanced Computer
Science and Applications, vol. 14, no. 6, 2023.
[14] L. Fu, J. Li, and Y. Chen, “An innovative decision making method for air quality monitoring based
on big data-assisted artificial intelligence technique,” Journal of Innovation & Knowledge, vol. 8,
no. 2, 100294, 2023.
[15] Y. Wang, “When artificial intelligence meets educational leaders’ data-informed decision-making:
A cautionary tale,” Studies in Educational Evaluation, vol. 69, 100872, 2021.
[16] F. Z. Borgesius, “Discrimination, artificial intelligence, and algorithmic decision-making,”
Strasbourg: Directorate General of Democracy, Council of Europe, 2018.
[17] E. Ntoutsi et al., “Bias in data-driven artificial intelligence systems — An introductory survey,” WIREs
Data Mining and Knowledge Discovery, vol. 10, no. 3, e1356, 2020.
[18] M. C. Pietronudo, G. Croidieu, and F. Schiavone, “A solution looking for problems? A systematic
literature review of the rationalizing influence of artificial intelligence on decision-making in
innovation management,” Technological Forecasting and Social Change, vol. 182, 121828, 2022.
[19] A. Turchin and D. Denkenberger, “Classification of global catastrophic risks connected with artificial
intelligence,” AI and Society, vol. 35, no. 1, pp. 147-163, 2020.
[20] K. Khan, R. Kunz, J. Kleijnen, and G. Antes, “Five steps to conducting a systematic review,” Journal
of the Royal Society of Medicine, vol. 96, no. 3, pp. 118-121, 2003.
[21] R. W. Wright, R. A. Brand, W. Dunn, and K. P. Spindler, “How to write a systematic review,” Clinical
Orthopaedics and Related Research, vol. 455, pp. 23-29, 2007.
[22] S. Gupta et al., “Systematic review of the literature: Best practices,” Academic Radiology, vol. 25,
no. 11, pp. 1481-1490, 2018.
[23] M. Gusenbauer and N. R. Haddaway, “Which academic search systems are suitable for systematic
reviews or meta-analyses? Evaluating retrieval qualities of Google Scholar, PubMed, and 26 other
resources,” Research Synthesis Methods, vol. 11, no. 2, pp. 181-217, 2020.
[24] N. Panic, E. Leoncini, G. de Belvis, W. Ricciardi, and S. Boccia, “Evaluation of the endorsement of
the preferred reporting items for systematic reviews and meta-analysis (PRISMA) statement on the
quality of published systematic review and meta-analyses,” PLoS ONE, vol. 8, no. 12, e83138, 2013.
[25] C. Stern, Z. Jordan, and A. McArthur, “Developing the review question and inclusion criteria,”
American Journal of Nursing, vol. 114, no. 4, pp. 53-56, 2014.
[26] T. Meline, “Selecting studies for systemic review: Inclusion and exclusion criteria,” Contemporary
Issues in Communication Science and Disorders, vol. 33, pp. 21-27, 2006.
[27] H. Edison, X. Wang, R. Jabangwe, and P. Abrahamsson, “Innovation initiatives in large software
companies: A systematic mapping study,” Information and Software Technology, vol. 95, pp. 1-14,
2018.
[28] M. Ivarsson and T. Gorschek, “A method for evaluating rigor and industrial relevance of technology
evaluations,” Empirical Software Engineering, vol. 16, no. 3, pp. 365-395, 2011.
[29] R. Klopper, S. Lubbe and H. Rugbeer, “The matrix method of literature review,” Alternation, vol.
14, pp. 1-17, 2007.
[30] S. Fosso Wamba, M. M. Queiroz, and L. Trinchera, “The role of artificial intelligence-enabled
dynamic capability on environmental performance: The mediation effect of a data-driven culture in
France and the USA,” International Journal of Production Economics, vol. 268, 109131, 2024.
[31] Q. Li, R. Xing, L. Li, H. Yao, L. Wu, and L. Zhao, “Synchrotron radiation data-driven artificial
intelligence approaches in materials discovery,” Artificial Intelligence Chemistry, vol. 2, no. 1,
100045, 2024.
[32] F. Sattari, L. Lefsrud, D. Kurian, and R. Macciotta, “A theoretical framework for data-driven
artificial intelligence decision making for enhancing the asset integrity management system in the
oil & gas sector,” Journal of Loss Prevention in the Process Industries, vol. 74, 104648, 2022.

206
[33] L. Kumar, N. U. Ibne Hossain, S. A. Fazio, A. Awasthi, R. Jaradat, and K. Babski-Reeves, “A data
driven decision model for assessing the enablers of quality dimensions: Context of Industry 4.0,”
CIRP Journal of Manufacturing Science and Technology, vol. 35, pp. 896-910, 2021.
[34] H. Nouinou et al., “Decision-making in the context of Industry 4.0: Evidence from the textile and
clothing industry,” Journal of Cleaner Production, vol. 391, 136184, 2023.
[35] M. Schmitt, “Automated machine learning: AI-driven decision making in business analytics,”
Intelligent Systems with Applications, vol. 18, 200188, 2023.
[36] L. Huang, M. Ladikas, J. Schippl, G. He, and J. Hahn, “Knowledge mapping of an artificial
intelligence application scenario: A bibliometric analysis of the basic research of data-driven
autonomous vehicles,” Technology in Society, vol. 75, 102360, 2023.
[37] W. Zhu, R. Gui, and R. Guo, “Unveiling the nexus and promoting integration of diverse factors:
Prospects of big data-driven artificial intelligence technology in achieving carbon neutrality in
Chongming District,” Water-Energy Nexus, vol. 6, pp. 111-121, 2023.
[38] V. Marda, “Artificial intelligence policy in India: A framework for engaging the limits of data-driven
decision-making,” Philosophical Transactions: Mathematical, Physical and Engineering Sciences,
vol. 376, no. 2133, pp. 1-19, 2018.
[39] K. Manheim and L. Kaplan, “Artificial intelligence: Risks to privacy and democracy,” The Yale
Journal of Law & Technology, vol. 21, pp. 106-188, 2019.
[40] F. A. Raso, H. Hilligoss, V. Krishnamurthy, C. Bavitz, and L. Kim, “Artificial intelligence & human
rights: Opportunities & risks,” Berkman Klein Center Research Publication Publication No. 2018-6,
2018.
[41] M. A. El-Brawany, D. A. Ibrahim, H. K. Elminir, H. M. Elattar, and E.A. Ramadan, “Artificial
intelligence-based data-driven prognostics in industry: A survey,” Computers & Industrial
Engineering, vol. 184, 109605, 2023.
[42] R. Colombari et al., “The interplay between data-driven decision-making and digitalization: A firm-
level survey of the Italian and U.S. automotive industries,” International Journal of Production
Economics, vol. 255, 108718, 2023.

207