THE USA JOURNALS
THE AMERICAN JOURNAL OF AGRICULTURE AND BIOMEDICAL ENGINEERING (ISSN
–
2689-1018)
VOLUME 06 ISSUE07
11
https://www.theamericanjournals.com/index.php/tajabe
PUBLISHED DATE: - 27-07-2024
DOI: -
https://doi.org/10.37547/tajabe/Volume06Issue07-03
PAGE NO.: - 11-28
EXPLORING BENEFITS, OVERCOMING
CHALLENGES, AND SHAPING FUTURE
TRENDS OF ARTIFICIAL INTELLIGENCE
APPLICATION IN AGRICULTURAL INDUSTRY
Sanchita Saha
Department Of Business Administration, Westcliff University, 17877 Von Karman Ave 4th
Floor, Irvine, CA 92614, United States
Ashok Ghimire
Department Of Business Administration, Westcliff University, 17877 Von Karman Ave 4th
Floor, Irvine, CA 92614, United States
Mia Md Tofayel Gonee Manik
Department Of Business Administration, Westcliff University, 17877 Von Karman Ave 4th
Floor, Irvine, CA 92614, United States
Anamika Tiwari
Department Of Business Administration, Westcliff University, 17877 Von Karman Ave 4th
Floor, Irvine, CA 92614, United States
Md Ahsan Ullah Imran
Department Of Business Administration, Westcliff University, 17877 Von Karman Ave 4th
Floor, Irvine, CA 92614, United States
RESEARCH ARTICLE
Open Access
THE USA JOURNALS
THE AMERICAN JOURNAL OF AGRICULTURE AND BIOMEDICAL ENGINEERING (ISSN
–
2689-1018)
VOLUME 06 ISSUE07
12
https://www.theamericanjournals.com/index.php/tajabe
INTRODUCTION
Geopolitical events and climatic disruptions are
increasingly straining supply chains and
undermining the resilience of food systems,
presenting significant challenges to global efforts
to end hunger and food insecurity [1]. The COVID-
19 pandemic has exacerbated these vulnerabilities,
exposing critical weaknesses in agri-food systems
and deepening societal inequalities [2]. As the
global population is projected to reach around 10
billion by 2050, the Food and Agriculture
Organization (FAO) estimates that food demand
will surge by 70% [3]. This dramatic rise
necessitates innovative solutions to enhance food
production and distribution. Artificial intelligence
(AI) offers a promising avenue for addressing these
challenges by optimizing agricultural processes. AI
techniques can improve crop yields, reduce waste,
and streamline supply chains, thereby enhancing
the overall resilience of food systems. By
leveraging AI, we can develop more efficient and
sustainable agricultural practices that are better
equipped to meet the growing food demands of the
future while mitigating the impacts of geopolitical
and climatic disruptions.
AI, a key discipline within computer science,
focuses on developing algorithms that mimic the
cognitive,
physiological,
or
evolutionary
phenomena observed in nature and human beings
[4]. Unlike traditional models that require explicit
knowledge of problem-solving paths, AI relies on
data, examples, and relationships to facilitate
diverse problem resolutions. This approach allows
AI to exhibit intelligent behavior akin to human
experts in specific tasks. Presently, AI is
predominantly directed towards solving problems
involving large, dynamic data sets that often
contain
inaccuracies
and
contradictions.
Techniques such as iterative methods and
interconnected neural network architectures,
commonly referred to as "Machine Learning" and
"Deep Learning," form the backbone of modern AI
[5-7]. These methods are widely applied across
various domains, unified by their ability to analyze
vast and complex data structures influenced by
temporal and uncertain factors.
Agriculture, an intricate sector blending science,
engineering, and economic principles, has
undergone significant advancements through the
integration of AI. Comprehensive reviews by Ruiz-
Real, et al. [8] and Jha, et al. [9] highlight the
transformative impact of expert systems and
decision
support
systems
in
optimizing
agricultural
processes
and
supply
chain
management. These systems simulate agricultural
operations, enhancing resource allocation and
process efficiency. AI's applications in quality
control, as explored by Nair and Mohandas [10],
Abstract
THE USA JOURNALS
THE AMERICAN JOURNAL OF AGRICULTURE AND BIOMEDICAL ENGINEERING (ISSN
–
2689-1018)
VOLUME 06 ISSUE07
13
https://www.theamericanjournals.com/index.php/tajabe
enable precise monitoring through artificial vision
systems. Additionally, AI facilitates policy
formulation, exemplified by Bryceson and
Slaughter [11] analyzed AI as a collaborative tool
among agri-food chain stakeholders. Economic
studies applying neural networks and machine
learning techniques to agri-food product pricing
reveal AI's potential in predicting and stabilizing
market fluctuations. In climate science, researchers
like Siegert, et al. [12] and Kosovic, et al. [13] used
AI to model and predict solar radiation, aiding in
climate management and agricultural planning.
The interest in AI's agricultural applications has
surged due to its robust data analysis capabilities.
The rise of Agriculture 3.0 introduced robotics and
automation, revolutionizing traditional practices
with sophisticated machinery capable of
autonomous planting, spraying, and harvesting
[14-16]. Agriculture 4.0 further enhances efficiency
through intelligent farms and interconnected
systems, focusing on precision agriculture to
optimize water, fertilizer, and phytosanitary
product use [17-19]. This approach, combined with
genetic engineering and big data analytics,
addresses challenges like climate change
adaptation and resource optimization. The
technification of agriculture, driven by Industry 4.0
concepts, has heightened AI's role, enabling agri-
food companies to streamline operations and drive
economic growth [20, 21]. This technological
evolution not only meets the rising demand for
richer diets but also enhances employment and
economic activities in industrial regions,
solidifying AI's crucial role in modern agriculture
and global food security.
Bibliometric studies that connect various
disciplines have become increasingly significant in
understanding the impact and future potential of
interdisciplinary synergies within the research
community. These studies serve as vital indicators
of interest and engagement in specific fields,
offering a comprehensive view of scientific
production and collaboration patterns. For
instance, Gu [22] demonstrated the global
structure of scientific output, highlighting the
intricate relationships between quality, references,
and author synergies. In the realm of artificial
intelligence (AI), Cobo, et al. [23] delved into its
evolution using various bibliometric indicators,
focusing on citations related to knowledge-based
systems. Despite the wealth of research in AI,
similar bibliometric studies in the agri-food
industry or agriculture are less common.
Nevertheless, the growing number of publications
in these fields, as evidenced by trends in Google
Scholar over the past five years, suggests a
burgeoning interest and a promising future for
interdisciplinary studies that explore the interplay
between AI and agriculture.
Literature reviews play a crucial role in
synthesizing the existing knowledge base, offering
insights into the application of AI in agriculture
[24]. Notable reviews include works on crop yield
prediction using machine learning [25], advanced
agricultural
disease
image
recognition
technologies [26], IoT solutions for smart farming,
and big data applications in agriculture [27-29].
These reviews highlight the diverse and innovative
applications of AI aimed at addressing various
agricultural challenges. In our study, we conducted
extensive research on AI applications in
agriculture, identifying seven key domains: crop
management,
water
management,
soil
management, fertilization, crop prediction, crop
classification, and disease and pest management.
From 176
studies selected for descriptive analysis, over 20
different AI techniques were identified. A
subsequent qualitative analysis distilled 17 articles
that provided detailed insights into the application,
challenges, and benefits of AI in agriculture,
underscoring the significant potential and ongoing
advancements in this field.
Research Purpose
SLR, as defined by [11], utilizes systematic methods
to collate and synthesize the findings of studies
addressing a formulated question, reported in
sufficient detail to enable replication of the
review's findings. This SLR aims to identify and
analyze recent studies on AI techniques applied in
agriculture, addressing specific questions and
recognizing emerging trends. The methodological
steps involved identification, screening, eligibility,
THE USA JOURNALS
THE AMERICAN JOURNAL OF AGRICULTURE AND BIOMEDICAL ENGINEERING (ISSN
–
2689-1018)
VOLUME 06 ISSUE07
14
https://www.theamericanjournals.com/index.php/tajabe
and inclusion of relevant studies. Initially, research
questions were defined, followed by the
establishment of criteria for study inclusion and
exclusion. A comprehensive search in scientific
databases was then conducted to extract relevant
studies, which were subsequently analyzed to
answer the research questions. The SLR ensured
clarity and transparency through a four-phase
verification flowchart adapted from [12]. This
work aims to expand current research knowledge
by focusing on AI technologies in agriculture. Table
1 presents the questions formulated for this
review.
Table 1. Questions for study
ID
Research question
Justification
R1
Which nations have had the most impact in the field
of research universities, publications, and
seminal works in AI approaches used in
agriculture?
To provide the necessary background
information regarding AI research in
the agricultural sector.
R2
What are the most common AI methods used for
agricultural work?
Application
of
AI
methods
for
identifying key areas of agriculture.
R3
When it comes to using AI in agriculture, what are
the biggest pros and cons?
Highlights trends and problems while
revealing potential avenues for future
research and improvement.
This research aimed to contextualize the topics of
AI and agriculture by addressing the research
question posed in Q1. To tackle Q2 and Q3, a
framework adapted from existing literature was
employed. Figure 1 illustrates the intersections
between AI technology's potential impacts on
agricultural applications (Q2) and highlights the
associated challenges and benefits (Q3). The
framework concentrates on AI technologies, their
application domains, and the ensuing challenges
and benefits, building on the methodologies
proposed by previous studies.
METHODOLOGY
This section outlines the review principles of the
Systematic Literature Review (SLR), the criteria for
selecting studies, and the quality assessment of the
chosen studies. To examine the evolution of AI in
the agricultural industry as reflected in scientific
publications, a comprehensive bibliometric
analysis was conducted. The study employs a
systematic bibliographical approach centered on a
specific topic, adhering to a sequence of methodical
steps. These steps include: (a) defining the search
criteria, keywords, and time frame; (b) selecting
relevant databases; (c) refining the research
criteria; (d) exporting the full set of results; and (e)
analyzing and discussing the findings (see Figure
1). This structured process ensures a rigorous and
thorough exploration of the literature, facilitating a
clear understanding of AI's impact on agriculture
over time.
Figure 1. Stages of bibliometric analysis
DATA COLLECTION
For the systematic exploration of scientific
information, a comprehensive array of literature
THE USA JOURNALS
THE AMERICAN JOURNAL OF AGRICULTURE AND BIOMEDICAL ENGINEERING (ISSN
–
2689-1018)
VOLUME 06 ISSUE07
15
https://www.theamericanjournals.com/index.php/tajabe
databases was meticulously utilized. These
databases, including ScienceDirect, Scopus,
Springer, IEEE Xplore, and MDPI, were selected for
their expansive coverage across various disciplines
and their robust repositories of scholarly works.
Each database was meticulously navigated to
ensure a thorough search for pertinent research
material. Table 2 serves as a crucial guide,
presenting the inclusion indicator for the data
collection phase, thus ensuring a systematic and
transparent approach in the retrieval and selection
of relevant literature. This meticulous selection
process underscores the commitment to rigor and
comprehensiveness in acquiring the necessary
foundation for the subsequent stages of analysis
and synthesis.
Table 2. Data collection source
Indicator
Description
Search interval
2017 to 2022
Databases Screening
ScienceDirect; Scopus; Springer; IEEE
Xplore; MDPI
Document types
Title, abstract, DOI, and year
Language
Review and original article
The proposed solution
English
Screening
Applied on agriculture
Search term
The search string devised for querying databases
on artificial intelligence and its applications in
agriculture incorporates a nuanced blend of
keywords, synonyms, and logical operators to
optimize the retrieval of relevant literature. By
amalgamating terms like "artificial intelligence"
and "agriculture" with synonyms and subarea
descriptors, the search aims to cast a wide net,
capturing varied perspectives and facets within the
intersection of these domains (see Table 3). The
strategic use of logical operators such as OR and
AND enhances the precision of the search, allowing
for the inclusion of diverse terms while
maintaining coherence in the retrieved results.
Employing this comprehensive query string in
advanced search fields of databases ensures a
systematic exploration of pertinent literature,
facilitating a robust understanding of the
advancements,
challenges,
and
potential
applications of artificial intelligence in agricultural
contexts.
Table 3. Search term of this study
Search term
TITLE-ABS-KEY (agriculture*) AND ( TITLE-ABS-KEY ( artificial* AND intelligence* ) OR
TITLE-ABS-KEY ( AI*) AND TITLE-ABS-KEY (machine* AND learning* ) AND TITLE-ABS-KEY
(deep* AND learning* )) AND PUBYEAR > 2012 AND PUBYEAR < 2022 AND ( LIMIT-TO (
DOCTYPE , "review" ) AND (DOCTYPE , "original article")) AND ( LIMIT-TO ( LANGUAGE ,
"English" ) )
Selection criteria
The systematic literature review (SLR) followed a
meticulous process of identification and screening
to ensure the inclusion of pertinent studies, as
depicted in Figure 2 of the PRISMA flow diagram.
THE USA JOURNALS
THE AMERICAN JOURNAL OF AGRICULTURE AND BIOMEDICAL ENGINEERING (ISSN
–
2689-1018)
VOLUME 06 ISSUE07
16
https://www.theamericanjournals.com/index.php/tajabe
This paper's identification phase meticulously
considered
contemporary
knowledge
disseminated through peer-reviewed scientific
journals in English. Notably, exclusion criteria were
applied to filter out sources such as book chapters,
annals, and abstracts, thereby maintaining a focus
on robust, scholarly publications. By adhering to
these rigorous standards, the SLR aimed to
synthesize a comprehensive understanding of the
topic under investigation, rooted in the latest
empirical research findings available in the
academic landscape.
Figure 2. PRISMA flow diagram of this study
Selection criteria
The quality assessment of papers within the SLR
encompassed an in-depth evaluation of various
pertinent aspects, crucial for determining their
inclusion or exclusion during the screening phase.
This qualitative analysis was designed to
meticulously categorize and prioritize the articles
scrutinized within the SLR framework. Five distinct
quality evaluation criteria were employed,
encompassing both quantitative and content-
related dimensions. Among these criteria, three
were quantitative, directly linked to journal
metrics such as Impact Factor, Citescore, and
Citations, serving as indicators of scholarly impact
and relevance. Additionally, qualitative aspects,
notably the integration of AI technologies and
applications within the agriculture domain, were
THE USA JOURNALS
THE AMERICAN JOURNAL OF AGRICULTURE AND BIOMEDICAL ENGINEERING (ISSN
–
2689-1018)
VOLUME 06 ISSUE07
17
https://www.theamericanjournals.com/index.php/tajabe
assessed for their alignment with the study's
objectives. Each criterion was assigned one of three
response options
—
high, medium, or low
—
based
on
predefined
thresholds,
ensuring
a
comprehensive and standardized assessment
process (see Table 4).
Table 4. Quality appraisal criteria
Indicator
Description
Impact Factor
Assess the significance of scientific
publications
Citescore
Stands for the typical amount of
references
Citations
The number of times a publication
has been referenced
Artificial intelligence technologies
Taking into account the technique,
citations, usage, and outcomes,
artificial intelligence technology is of
extreme significance.
Agriculture domain applications
Taking into account the technique,
citations, usage, and outcomes,
agriculture application is of extreme
significance.
During the eligibility analysis phase, thorough
scrutiny of the 176 selected studies revealed
distinct ranges of values pertinent to various
criteria. For instance, in assessing the impact
factor, a spectrum from 11.8 to 6.1 delineated high-
quality studies, each garnering 1.0 points in this
criterion, as outlined in Table 5. This quantitative
approach facilitated a nuanced evaluation, wherein
each record could amass points ranging from 0 to 3
across multiple criteria. By meticulously
delineating
such
ranges
and
assigning
corresponding point values, the eligibility analysis
not only ensured a comprehensive assessment but
also facilitated a structured comparison among the
diverse pool of studies under consideration.
Table 5. Quantitative assessment criteria
Criterion
High
Medium
Low
ImpactFactor
[11.8–6.1]
[6.0–3.1]
[3.0–1.58]
Citescore
[18.7–6.5]
[6.4–4.0]
[3.9–2.7]
Citations
[1195–100]
[99–10]
[9–0]
For a quality appraisal, all 176 papers that were chosen during the eligibility stage were reviewed in their
entirety. The papers with the highest scores are shown in Table 7.
Table 7. Quality assessment with a high score
References
AI technology
Application
domain
Citescore Impact
Factor
Citation
Score
[30]
Robotics and
automation; Computer
Crop
management
5.002
8.7
135
5.5
THE USA JOURNALS
THE AMERICAN JOURNAL OF AGRICULTURE AND BIOMEDICAL ENGINEERING (ISSN
–
2689-1018)
VOLUME 06 ISSUE07
18
https://www.theamericanjournals.com/index.php/tajabe
vision; Convolutional
neural network
[31]
Genetic algorithm;
Internet of Things
(IoT)
Fertigation
management
11.072
15.8
25
5.5
[32]
Machine learning
Water
management
6.757
11.8
40
5.5
[33]
Digital twins
Crop
management
6.757
11.8
63
5.5
[34]
Machine learning
Crop prediction
8.171
12
12
5.5
[35]
Machine Learning
(ML); Internet of
Things (IoT)
Crop
management
3.476
7
12
5.0
[27]
Bigdata; Robotics
Crop
management
6.765
9.7
1195
5.0
[36]
Machine earning;
Computer vision
Crop
management
6.757
11.8
70
5.0
[37]
Deep learning;
Computer vision
Crop
Classification
3.889
5.0
19
4.5
RESULTS AND DISCUSSION
Descriptive Analysis(R1)
This section provides a quantitative analysis of the
176 selected studies, encompassing their
publication
and
citation
volumes,
the
methodologies employed in their identification,
and the predominant countries, journals, and
institutions driving their research. Figure 3 visually
represents the distribution of papers and citations
per year, offering insight into the temporal trends
of scholarly activity within the field. Through
meticulous examination, this section illuminates
the landscape of research, shedding light on the
prolific contributors, influential publications, and
the evolving discourse surrounding the subject
matter. By delineating these key metrics, it affords
a comprehensive understanding of the breadth and
impact of the studies under scrutiny, thereby
facilitating informed interpretation and guiding
future inquiry.
THE USA JOURNALS
THE AMERICAN JOURNAL OF AGRICULTURE AND BIOMEDICAL ENGINEERING (ISSN
–
2689-1018)
VOLUME 06 ISSUE07
19
https://www.theamericanjournals.com/index.php/tajabe
Figure 3. Publications and citations per year
Over the past three years, there has been a
remarkable surge in both the number of
publications and citations within the field,
indicating a substantial growth in research activity
and scholarly engagement. Data collected up to
December 2022 revealed a peak of 2440
publications referencing works from 2020. This
surge underscores the dynamic feedback loop
prevalent in this area of research, with subsequent
years building upon and citing earlier findings.
These studies have been systematically classified
into theoretical and empirical categories (see Table
8). Theoretical inquiries encompass reviews and
SLRs, while empirical investigations include
modeling and simulation studies, surveys, and case
studies. This structured classification offers
valuable insights into the diverse methodologies
employed within the field, reflecting its
multidisciplinary nature and the breadth of
approaches undertaken to advance knowledge and
understanding.
Table 8. Identified studies
Studies
Category
Total
%
Theoretical
Reviews
59
34
Systematic reviews
16
9
Total
75
43
Empirical
Modeling and simulations
68
39
Case studies
13
7
Surveys
20
11
Total
101
57
Overall total
176
100
The distribution of studies was balanced, with 43%
classified as theoretical and 57% as empirical. The
theoretical studies predominantly focused on
literature reviews, accounting for 34% of the total
with 59 papers. Among the empirical studies,
modeling and simulation were particularly
prominent, comprising 39% of 68 papers. This
distribution is mirrored in the quality appraisal
THE USA JOURNALS
THE AMERICAN JOURNAL OF AGRICULTURE AND BIOMEDICAL ENGINEERING (ISSN
–
2689-1018)
VOLUME 06 ISSUE07
20
https://www.theamericanjournals.com/index.php/tajabe
steps, where notable articles such as [38] and [39]
exemplify the applied research in the field. These
studies respectively developed innovative systems
for robotic strawberry picking and smart
irrigation, highlighting the practical implications of
both theoretical and empirical research in
advancing technological applications.
When evaluating the significance of a dissertation
Bibliometric analysis sheds light on the quantity
and influence of publications by use of citation
metrics; it assesses the relevance of research
articles by means of a number of variables. The
Netherlands stands out as the most influential
country in this domain, achieving 1629 citations
from only 5 publications. India is a close contender
with 37 publications accumulating 1499 citations.
Greece, although contributing just 3 publications,
has a substantial impact with 1002 citations. China,
with 23 publications, has garnered 899 citations.
Table 9 provides a comprehensive overview of
countries with more than 100 citations,
highlighting the global reach and influence within
this research area.
Table 9. Quality assessment with a high score
Score
Country
Publications Citations
1
Netherlands
5
1629
2
India
37
1499
3
Greece
3
1002
4
China
23
899
5
Spain
5
868
6
USA
10
679
7
Australia
5
649
8
Brazil
9
556
9
France
2
232
10
Egypt
3
185
11
New Zealand
2
168
12
Italy
7
157
13
Pakistan
5
144
14
Malaysia
7
115
15
Portugal
2
107
16
Canada
3
105
17
Chile
4
100
Over the last three years, the publications and
citations of artificial intelligence (AI) techniques
applied to agriculture have increased almost
sixfold, underscoring the growing significance and
relevance of this research area. The most
influential countries in this domain are among the
world's largest food producers, highlighting the
global recognition of AI's potential to revolutionize
agricultural practices. Additionally, numerous
high-impact journals have emerged as key
platforms for disseminating groundbreaking
research in this field, further demonstrating the
scholarly and practical importance of integrating
AI in agriculture.
THE USA JOURNALS
THE AMERICAN JOURNAL OF AGRICULTURE AND BIOMEDICAL ENGINEERING (ISSN
–
2689-1018)
VOLUME 06 ISSUE07
21
https://www.theamericanjournals.com/index.php/tajabe
Artificial Intelligence in Agriculture (R2)
Agriculture domain
Agriculture, defined as the science of cultivating
land and raising livestock, is essential for
producing the food and resources necessary for
human survival. Central to this practice is the
physical environment, which serves as the
foundational resource base, and the cultivated crop
plant, which is the primary unit of production [40].
The success of agricultural endeavors hinges on the
effective management of the physical environment
to meet the biological demands of these crops. Key
factors influencing crop yield include soil
productivity, water availability, climate conditions,
and the control of pests and diseases [41].
Mastering these variables is crucial for optimizing
agricultural output and ensuring the sustainability
and resilience of food systems.
Artificial intelligence is revolutionizing the
agricultural sector by optimizing processes and
resources. This review identified seven key
applications:
crop
management,
water
management,
soil
management,
chemical
application, fertigation, crop prediction, and crop
classification, as summarized in Table 10. The
primary goal of crop management is to rationalize
resource use [42]. Water management focuses on
optimizing irrigation and water use on farms, while
soil management is crucial for the success of site-
specific cropping systems. Proper chemical
application is vital for environmental and economic
sustainability [43]. Fertigation, the use of irrigation
systems for fertilizer application, has been shown
to enhance fertilizer effectiveness. Crop prediction
and classification, utilizing image processing and
deep learning [44], are essential for sustainable
resource utilization. Effective management of
diseases and pests is critical for improving crop
yield, quality, and overall food security.
Table 10. Uses in the agricultural sector
References Domain
Details
[45]
Crop management
Included are planting, tending, harvesting, storing,
and distributing seeds
[46]
Water
management
Streamlining irrigation methods and processes to
maximize utilization of water
[47]
Soil management
Make sure plants get enough nutrients.
[48]
Fertirrigation
Technology with the goal of providing data for a
different application sector and a worldwide
perspective on crop distribution
[34]
Crop prediction
The logistical management of farmers relies heavily
on crop productivity projection
[49]
Crop classification The purpose of crop categorization is to provide a
comprehensive picture of crop dispersion and
related data for a different field of usage
[50]
Disease and pest
management
Impair crop yields and quality while decreasing the
efficiency of resource utilization. Protecting
agricultural crops from the many different kinds of
weeds,
animals,
and
microbes
necessitates
technological solutions.
Artificial Intelligence Technologies
The integration of AI into agriculture, aimed at
enhancing food production while mitigating the
effects of climate change, poses significant
challenges rooted in the analysis of AI technologies.
Originating from the conceptualization of cognitive
processes and neurobiology in the 1950s, AI's
evolution has delineated four distinct categories of
intelligent systems: those that emulate human
THE USA JOURNALS
THE AMERICAN JOURNAL OF AGRICULTURE AND BIOMEDICAL ENGINEERING (ISSN
–
2689-1018)
VOLUME 06 ISSUE07
22
https://www.theamericanjournals.com/index.php/tajabe
thought, behavior, rational thinking, and rational
action. Success in these categories is gauged by
their fidelity to human performance or rationality
[8]. AI systems are proficient in data storage,
manipulation,
knowledge
acquisition,
representation, and deduction of new knowledge
from existing information. Within the realm of
agriculture,
AI
technologies are diverse,
encompassing cognitive science, robotics, and
natural interface applications. These technologies,
as identified across numerous studies, are
bolstered by auxiliary technologies like IoT, big
data, and cloud computing, enabling the
implementation of specific AI techniques such as
computer vision, robotics, machine learning,
augmented reality, and virtual reality.
The analysis of 17 selected articles in the quality
assessment stage revealed a spectrum of AI
technologies and their applications in agriculture,
as summarized in Table 11. Notably, the focus
encompassed innovations such as irrigation,
disease, and pest management systems, which
emerged
prominently
in
the
literature.
Additionally, a comprehensive review of these
advancements, as presented in [24], provided
valuable insights into the field's progression.
Furthermore, the development of a smart
irrigation system, detailed in [39], showcased
practical implementations of AI in enhancing
agricultural practices. Moreover, the exploration of
emerging technologies like agricultural digital
twins, highlighted in [33], underscored the ongoing
efforts to capture the intricate interactions
between living systems and their environment.
Table 10. AI technologies that were mentioned in the chosen papers
References Domain
Details
[30]
Robotics and
automation
The utilization of machinery, software, and other
technology to carry out activities that replace or
mimic human movements
[51]
Drones and
unmanned aerial
vehicles (UAVs)
Unmanned aerial vehicles that are capable of being
commanded remotely
[32]
Machine learning
(ML)
This system has the ability to adapt its behavior on
its own using various algorithms that can be used
to evaluate performance and make accurate
predictions.
[52]
Artificial neural
networks(ANNs)
Computer systems that mimic human intelligence,
which is able to acquire new knowledge and adjust
to novel and ever-changing circumstances.
[37]
Deep learning:
convolutional
neural network
CNN)
A collection of algorithms connected to machine
learning forms its basis. CNNs use picture patterns
to identify objects, classifications, and categories.
[30]
Genetic algorithm
(GA)
Machine learning algorithms that simulate evolving
processes and resolve issues by studying biological
evolution.
[31]
Computer vision
Automatic picture acquisition, analysis, and
comprehension are all part of computer vision,
which includes issues like object detection, motion
tracking, and action recognition.
THE USA JOURNALS
THE AMERICAN JOURNAL OF AGRICULTURE AND BIOMEDICAL ENGINEERING (ISSN
–
2689-1018)
VOLUME 06 ISSUE07
23
https://www.theamericanjournals.com/index.php/tajabe
[53]
Digital twins
The goal of this virtual representation is to
maximize efficiency and production
[36]
Internet of Things
(IoT)
Establishes connections between various intelligent
devices, making crop management easier
[54]
Cloud computing
Assets like storage for information and computer
power are made available on request
[55]
Big data
Information is gathered, processed, and analyzed
In our comprehensive review of the literature, we
have discerned seven primary applications that
underscore the diverse utility of artificial
intelligence (AI) in agricultural contexts. These
applications encompass crop management, water
management, soil management, fertigation, crop
prediction,
crop
classification,
and
the
identification and mitigation of diseases and pests.
Furthermore, our analysis has unveiled a rich
tapestry of AI techniques, numbering twenty-four
distinct methodologies. Notably, prevalent among
these techniques are big data analysis, Internet of
Things (IoT) integration, and cloud computation.
Of these applications, crop management, water
management, and disease and pest control emerge
as the most prevalent areas of focus, indicating
their significance in addressing contemporary
agricultural challenges. Within the realm of AI
techniques, machine learning, robotics, deep
learning, and IoT technologies stand out as the
most frequently employed tools, underscoring
their pivotal role in advancing agricultural
innovation and sustainability.
Benefits, Challenges and Trends (RQ3)
The integration of machine learning, deep learning,
and computer vision techniques within agriculture
has heralded a transformative era known as
Agriculture 4.0 or Digital Agriculture. This
paradigm shift leverages advanced technologies
like precision agriculture, IoT, and cloud
computing to revolutionize farming practices.
Precision
agriculture
optimizes
resource
utilization and minimizes environmental impact
through real-time monitoring aided by remote
sensing technologies. Agriculture 4.0, akin to
Industry 4.0, harnesses big data and novel
technologies across the supply chain, aiming to
enhance productivity while reducing waste.
Moreover, the emergence of next-generation
agriculture, denoted as 5.0 and 6.0, emphasizes
deep learning and robotic advancements to
balance
production
and
environmental
sustainability. Underpinning these advancements
is the pervasive use of artificial intelligence,
facilitating efficient crop management, irrigation,
and disease detection by processing vast datasets.
This synergy of technologies underscores a pivotal
shift towards intelligent farming systems, where
data-driven insights empower farmers to make
informed
decisions,
fostering
agricultural
sustainability and productivity.
UAVs have revolutionized data collection in
agriculture, offering vast and intricate datasets.
Leveraging big data analytics tools and cloud
computing can markedly enhance data processing
efficiency, bolster data security, ensure scalability,
and reduce operational costs. Furthermore, the
integration of ML, ANN-based, and deep learning
techniques shows promise, particularly in crop
prediction, owing to the abundance of data from
diverse sources. Amidst these advancements,
emerging technologies like DT are reshaping the
agricultural landscape. While PA remains
prominent, newer terms such as Agriculture 4.0
and smart farming are gaining traction in scholarly
discourse. Notably, there's a call for tailored
research endeavors, considering regional climate
and crop specifics, particularly in countries like
Brazil, which are yet to fully harness their scientific
potential
in
this
domain.
Additionally,
understanding the intricacies of crop production
chains is crucial for effectively applying AI
technologies and optimizing production (see
Figure 4). These technologies, with their common
reliance on digitized data, lie at the heart of the
digital revolution in agriculture. As we chart future
research trajectories, it's imperative to focus on
THE USA JOURNALS
THE AMERICAN JOURNAL OF AGRICULTURE AND BIOMEDICAL ENGINEERING (ISSN
–
2689-1018)
VOLUME 06 ISSUE07
24
https://www.theamericanjournals.com/index.php/tajabe
enabling technologies that facilitate the practical
application and dissemination of research
outcomes.
Figure 4. The top 10 technologies and phrases that emerged from the examination of 175
publications
CONCLUSIONS
This systematic literature review, conducted using
the PRISMA methodology, comprehensively
examines the application of artificial intelligence
(AI) technologies in agriculture. Analyzing 176
papers through bibliometric analysis and assessing
17 papers for quality, the review identifies seven
primary agricultural applications: crop, water, soil
management, fertigation, prediction, classification,
and disease/pest management. Notably, twenty-
four AI techniques, prominently including machine
learning, deep learning with convolutional neural
networks, robotics, and the Internet of Things, are
identified as pivotal in advancing agricultural
practices. These technologies offer substantial
benefits such as optimizing management systems,
irrigation, and disease identification. However,
challenges persist, particularly in digitizing
production processes and addressing the
associated costs, which remain prohibitive for
many farmers. Labor qualification and the need for
supportive public policies in food-producing
nations are also highlighted. Moreover, the
integration of computer vision with robotics and
UAVs for crop classification and disease detection,
alongside emerging technologies like digital twins,
signifies promising avenues for agricultural
optimization. As precision agriculture transitions
towards smart farming frameworks, incorporating
telecommunications and data infrastructure, the
advent of agriculture 4.0 and 5.0 further
underscores the evolving landscape driven by AI
and UAV technologies. Despite these insights, the
review acknowledges limitations such as language
bias and database selection, suggesting avenues for
future research to enhance the comprehensiveness
and validity of findings.
REFERENCES
1.
D. Aminetzah et al., "A reflection on global food
security challenges amid the war in Ukraine and
the early impact of climate change," McKinsey’s
Agriculture Practice, 2022.
2.
S. Shiratori, Y. Tobita, and E. M. Sawadogo-
Compaoré, "Food Security, Nutritional Supply,
and Nutrient Sources in Rural Burkina Faso,"
THE USA JOURNALS
THE AMERICAN JOURNAL OF AGRICULTURE AND BIOMEDICAL ENGINEERING (ISSN
–
2689-1018)
VOLUME 06 ISSUE07
25
https://www.theamericanjournals.com/index.php/tajabe
Nutrients, vol. 15, no. 10, p. 2285, 2023, doi:
10.3390/nu15102285
3.
N. Alexandratos and J. Bruinsma, "World
agriculture towards 2030/2050: the 2012
revision," 2012, doi: 10.22004/ag.econ.288998.
4.
B. A. King, T. Hammond, and J. Harrington,
"Disruptive technology: Economic consequences
of artificial intelligence and the robotics
revolution," Journal of Strategic Innovation and
Sustainability, vol. 12, no. 2, pp. 53-67, 2017.
5.
F. S. Aditto et al., "Fresh, mechanical and
microstructural behaviour of high-strength self-
compacting concrete using supplementary
cementitious materials," Case Studies in
Construction Materials, vol. 19, p. e02395, 2023.
6.
J. A. Jabin, M. T. H. Khondoker, M. H. R. Sobuz, and
F. S. Aditto, "High-temperature effect on the
mechanical behavior of recycled fiber-
reinforced concrete containing volcanic pumice
powder: An experimental assessment combined
with machine learning (ML)-based prediction,"
Construction and Building Materials, vol. 418, p.
135362,
2024/03/08/
2024,
doi:
https://doi.org/10.1016/j.conbuildmat.2024.1
35362.
7.
M. H. R. Sobuz et al., "Optimization of recycled
rubber self-compacting concrete: Experimental
findings
and
machine
learning-based
evaluation," Heliyon, vol. 10, no. 6, 2024, doi:
https://doi.org/10.1016/j.heliyon.2024.e2779
3.
8.
J. L. Ruiz-Real, J. Uribe-Toril, J. A. Torres Arriaza,
and J. de Pablo Valenciano, "A look at the past,
present and future research trends of artificial
intelligence in agriculture," Agronomy, vol. 10,
no.
11,
p.
1839,
2020,
doi:
10.3390/agronomy10111839
9.
K. Jha, A. Doshi, P. Patel, and M. Shah, "A
comprehensive review on automation in
agriculture
using
artificial
intelligence,"
Artificial Intelligence in Agriculture, vol. 2, pp. 1-
12,
2019/06/01/
2019,
doi:
https://doi.org/10.1016/j.aiia.2019.05.004.
10.
B. B. Nair and V. Mohandas, "Artificial
intelligence
applications
in
financial
forecasting
–
a survey and some empirical
results," Intelligent Decision Technologies, vol.
9, no. 2, pp. 99-140, 2015, doi: 10.3233/IDT-
140211.
11.
K. Bryceson and G. Slaughter, "Integrated
Autonomy A Modeling-Based Investigation of
Agrifood Supply Chain Performance," in 2009
11th International Conference on Computer
Modelling and Simulation, 2009: IEEE, pp. 334-
339, doi: 10.1109/UKSIM.2009.42.
12.
C. Siegert, D. Leathers, and D. Levia, "Synoptic
typing: interdisciplinary application methods
with three practical hydroclimatological
examples," Theoretical and Applied Climatology,
vol.
128,
pp.
603-621,
2017,
doi:
https://doi.org/10.1007/s00704-015-1700-y.
13.
I. N. Kosovic, T. Mastelic, and D. Ivankovic, "Using
Artificial Intelligence on environmental data
from Internet of Things for estimating solar
radiation: Comprehensive analysis," Journal of
Cleaner Production, vol. 266, p. 121489,
2020/09/01/
2020,
doi:
https://doi.org/10.1016/j.jclepro.2020.121489
.
14.
G. Ren, T. Lin, Y. Ying, G. Chowdhary, and K. C.
Ting, "Agricultural robotics research applicable
to poultry production: A review," Computers
and Electronics in Agriculture, vol. 169, p.
105216,
2020/02/01/
2020,
doi:
https://doi.org/10.1016/j.compag.2020.10521
6.
15.
S. Fountas, N. Mylonas, I. Malounas, E. Rodias, C.
Hellmann Santos, and E. Pekkeriet, "Agricultural
robotics for field operations," Sensors, vol. 20,
no. 9, p. 2672, 2020, doi: 10.3390/s20092672
16.
J. Lowenberg-DeBoer, I. Y. Huang, V. Grigoriadis,
and S. Blackmore, "Economics of robots and
automation in field crop production," Precision
Agriculture, vol. 21, no. 2, pp. 278-299, 2020, doi:
https://doi.org/10.1007/s11119-019-09667-5.
17.
D. C. Rose, R. Wheeler, M. Winter, M. Lobley, and
C.-A. Chivers, "Agriculture 4.0: Making it work
for people, production, and the planet," Land Use
Policy, vol. 100, p. 104933, 2021/01/01/ 2021,
doi:
https://doi.org/10.1016/j.landusepol.2020.104
THE USA JOURNALS
THE AMERICAN JOURNAL OF AGRICULTURE AND BIOMEDICAL ENGINEERING (ISSN
–
2689-1018)
VOLUME 06 ISSUE07
26
https://www.theamericanjournals.com/index.php/tajabe
933.
18.
Z. Zhai, J. F. Martínez, V. Beltran, and N. L.
Martínez, "Decision support systems for
agriculture 4.0: Survey and challenges,"
Computers and Electronics in Agriculture, vol.
170, p. 105256, 2020/03/01/ 2020, doi:
https://doi.org/10.1016/j.compag.2020.10525
6.
19.
S. K. Roy and D. De, "Genetic Algorithm based
Internet of Precision Agricultural Things (IopaT)
for Agriculture 4.0," Internet of Things, vol. 18, p.
100201,
2022/05/01/
2022,
doi:
https://doi.org/10.1016/j.iot.2020.100201.
20.
M. Ryan, "Agricultural big data analytics and the
ethics of power," Journal of Agricultural and
Environmental Ethics, vol. 33, pp. 49-69, 2020,
doi:
https://doi.org/10.1007/s10806-019-
09812-0.
21.
M. Mokarram and M. R. Khosravi, "A cloud
computing
framework
for
analysis
of
agricultural big data based on Dempster
–
Shafer
theory," The Journal of Supercomputing, vol. 77,
pp.
2545-2565,
2021,
doi:
https://doi.org/10.1007/s11227-020-03366-z.
22.
Y. Gu, "Global knowledge management research:
A bibliometric analysis," Scientometrics, vol. 61,
pp.
171-190,
2004,
doi:
https://doi.org/10.1023/B:SCIE.0000041647.0
1086.f4.
23.
M. J. Cobo, M. A. Martínez, M. Gutiérrez-Salcedo,
H. Fujita, and E. Herrera-Viedma, "25years at
Knowledge-Based Systems: A bibliometric
analysis," Knowledge-Based Systems, vol. 80, pp.
3-13,
2015/05/01/
2015,
doi:
https://doi.org/10.1016/j.knosys.2014.12.035.
24.
T. van Klompenburg, A. Kassahun, and C. Catal,
"Crop yield prediction using machine learning: A
systematic literature review," Computers and
Electronics in Agriculture, vol. 177, p. 105709,
2020/10/01/
2020,
doi:
https://doi.org/10.1016/j.compag.2020.10570
9.
25.
Y. Yuan, L. Chen, H. Wu, and L. Li, "Advanced
agricultural
disease
image
recognition
technologies: A review," Information Processing
in Agriculture, vol. 9, no. 1, pp. 48-59,
2022/03/01/
2022,
doi:
https://doi.org/10.1016/j.inpa.2021.01.003.
26.
M. S. Farooq, S. Riaz, A. Abid, T. Umer, and Y. B.
Zikria, "Role of IoT technology in agriculture: A
systematic literature review," Electronics, vol. 9,
no.
2,
p.
319,
2020,
doi:
10.3390/electronics9020319
27.
S. Wolfert, L. Ge, C. Verdouw, and M.-J. Bogaardt,
"Big Data in Smart Farming
–
A review,"
Agricultural Systems, vol. 153, pp. 69-80,
2017/05/01/
2017,
doi:
https://doi.org/10.1016/j.agsy.2017.01.023.
28.
F. Maffezzoli, M. Ardolino, A. Bacchetti, M.
Perona, and F. Renga, "Agriculture 4.0: A
systematic literature review on the paradigm,
technologies and benefits," Futures, vol. 142, p.
102998,
2022/09/01/
2022,
doi:
https://doi.org/10.1016/j.futures.2022.102998
.
29.
S. Araújo, R. Peres, J. Barata, F. Lidon, and J.
Ramalho, "Characterising the Agriculture 4.0
Landscape
—
Emerging Trends, Challenges and
Opportunities. Agronomy 2021, 11, 667," ed,
2022.
30.
H. A. M. Williams et al., "Robotic kiwifruit
harvesting using machine vision, convolutional
neural networks, and robotic arms," Biosystems
Engineering,
vol.
181,
pp.
140-156,
2019/05/01/
2019,
doi:
https://doi.org/10.1016/j.biosystemseng.2019.
03.007.
31.
N. Lin, X. Wang, Y. Zhang, X. Hu, and J. Ruan,
"Fertigation management for sustainable
precision agriculture based on Internet of
Things," Journal of Cleaner Production, vol. 277,
p.
124119,
2020/12/20/
2020,
doi:
https://doi.org/10.1016/j.jclepro.2020.124119
.
32.
M. M. Reis, A. J. da Silva, J. Zullo Junior, L. D. Tuffi
Santos, A. M. Azevedo, and É. M. G. Lopes,
"Empirical and learning machine approaches to
estimating reference evapotranspiration based
on temperature data," Computers and
Electronics in Agriculture, vol. 165, p. 104937,
2019/10/01/
2019,
doi:
THE USA JOURNALS
THE AMERICAN JOURNAL OF AGRICULTURE AND BIOMEDICAL ENGINEERING (ISSN
–
2689-1018)